Using Social Network Analysis to Support Collective Decision-Making Process

IJDSST Editorial Board Editor-in-Chief: Pascale Zaraté, Toulouse U., France International Advisory Board: Frederic Adam, U. College Cork, Ireland Hojjat Adeli, The Ohio State U., USA David Paradice, Florida State U., USA Nava Pliskin, Ben-Gurion U., Israel Daniel Power, U. of Northern Iowa, USA Andrew B. Whinston, U. of Texas - Austin, USA Associate Editors: James Courtney, U. of Central Florida, USA Jeet Gupta, U. of Alabama - Huntsville, USA James R. Marsden, U. of Connecticut, USA Manuel Mora, Autonomous U. of Aguascalientes, Mexico Vicki L. Sauter, U. of Missouri, USA Victoria Yoon, U. of Maryland Baltimore County, USA IGI Editorial: Heather A. Probst, Director of Journal Publications Jamie M. Wilson, Assistant Director of Journal Publications Chris Hrobak, Journal Production Editor Gregory Snader, Production and Graphics Assistant Brittany Metzel, Production Assistant International Editorial Review Board: Ezendu Ariwa, London Metropolitan U., UK Sandro Bimonte, U. of Lyon, France Patrick Brezillon, Pierre and Marie Curie U., France Ricardo Colomo-Palacios, Universidad Carlos III de Madrid, Spain Fátima C.C. Dargam, Simtech Simulations Technology, Austria, and ILTC Research Institute, Brazil Carlos Escobar, Universidad Nacional Autómoma de México, Mexico Leonardo Garrido, Monterrey Tech, Mexico Carina Gonzales, U. of the Laguna, Spain Miro Gradisar, U. of Ljubljana,Slovenia Zhiling Guo, U. of Maryland - Baltimore County, USA Ashish Gupta, Minnesota State U. Moorhead, USA Payam Hanaizadeh, Allameh Tabatabae’i U., Iran Luca Iandoli, Università degli Studi di Napoli Federico II, Italy IGIP Miroljub Kljajic, U. of Maribor, Slovenia Carlos Legna, Universidad de la Laguna, Spain Katty Murty, Michigan State U., USA Daniel O’Leary, U. of Southern California, USA Zita Zoltayne Paprika, Budapest U., Hungary Doncho Petkov, Eastern Connecticut State U., USA Roger Pick, U. of Missouri - Kansas City, USA R. Venkato Rao, Sardar Vallabhbhai National Institute of Technology (SV NIT), India Rita Ribeiro, UNINOVA, Portugal Clive Roberts, U. of Birmingham, UK J P Shim, Mississippi State U., USA Ralph Sprague, U. of Hawaii at Manoa, USA David Ullman, Robust Decisions Inc., USA P. Vasant, U. Technology Petronas, Malaysia Xuan F. Zha, NITS, USA IGI PublIshInG www.igi-global.com CALL foR ARtICLeS International Journal of Decision Support System Technology An official publication of the Information Resources Management Association The Editor-in-Chief of the International Journal of Decision Support System Technology (IJDSST) would like to invite you to consider submitting a manuscript for inclusion in this scholarly journal. MISSION: The primary objective of the International Journal of Decision Support System Technology (IJDSST) is to provide comprehensive coverage for DMSS technology issues. The issues can involve, among other things, new hardware and software for DMSS, new models to deliver decision making support, dialog management between the user and system, data and model base management within the system, output display and presentation, DMSS operations, and DMSS technology management. Since the technology’s purpose is to improve decision making, the articles are expected to link DMSS technology to improvements in the process and outcomes of the decision making process. This link can be established theoretically, mathematically, or empirically in a systematic and scientiic manner. COVERAGE/MAJOR TOPICS: • Context awareness, modeling, and management for DMSS • DMSS computer hardware • DMSS computer systems and application software • DMSS data capture, storage, and retrieval • DMSS feedback control mechanisms • DMSS function integration strategies and mechanisms • DMSS model capture, storage, and retrieval • DMSS network strategies and mechanisms • DMSS output presentation and capture • DMSS system and user dialog methods • DMSS system design, development, testing, and implementation • DMSS technology evaluation • DMSS technology organization and management • Public and private DMSS applications • Web-based and mobile DMSS technologies • Other related technology issues that impact the overall utilization and management of DMSS in modern life and organizations ISSN 1941-6296 eISSN 1941-630X Published quarterly All submissions should be emailed to Pascale Zaraté, Editor-in-Chief zarate@irit.fr Ideas for Special theme Issues may be submitted to the editor-in-Chief. Please recommend this publication to your librarian. For a convenient easy-to-use library recommendation form, please visit: http://www.igiglobal.com/ijdsst and click on the "Library Recommendation Form" link along the right margin. InternatIonal Journal of DecIsIon support system technology April-June 2011, Vol. 3, No. 2 Table of Contents Special Issue on Models for Collaborative Decision Making Processes and Cases Studies on Decision Support Systems Editorial Preface i Pascale Zaraté, Toulouse University, France Research Articles 1 Discrepancies and Analogies in Artiicial Intelligence and Engineering Design Approaches in Addressing Collaborative Decision-Making Marija Jankovic, Ecole Centrale Paris, France Pascale Zaraté, Toulouse University, France 15 Using Social Network Analysis to Support Collective Decision-Making Process Simon Buckingham Shum, The Open University, UK Lorella Cannavacciuolo, University of Naples Federico II, Italy Anna De Liddo, The Open University, UK Luca Iandoli, University of Naples Federico II, Italy Ivana Quinto, University of Naples Federico II, Italy 32 Strategic Development of a Decision Making Support System in a Public R&D Center Carlos E. Escobar-Toledo, Universidad Nacional Autónoma de México, Mexico Héctor A. Martínez-Berumen, Universidad Nacional Autónoma de México and CIATEQ, Mexico 44 Decision Support for Crisis Incidents Daniel J. Power, University of Northern Iowa, USA Roberta M. Roth, University of Northern Iowa, USA Rex Karsten, University of Northern Iowa, USA 57 Understanding Organisational Decision Support Maturity: Case Studies of Irish Organisations Mary Daly, University College Cork, Ireland Frederic Adam, University College Cork, Ireland i Editorial PrEfacE Pascale Zaraté, Toulouse University, France In July 2010 the previous Editor-in-Chief, Prof. G. Forgionne, asked me to take in charge the function of Editor-in-Chief of the International Journal of Decision Support System Technology. I assume this function since this date and for this occasion I asked the Editorial Review Board Committee members to freely express their views on DSS Technologies. This issue is comprised of papers on the topics of Models for Collaborative Decision Making Processes and Cases Studies on Decision Support Systems contributed by the editorial board of IJDSST. I received ive papers suitable for this issue. These papers broadly relect the current research work of their authors. All these authors greatly contribute to the DSS community, but the content of this issue could be seen as very heterogeneous. I think that the heterogeneity of contributions to the DSS ield is not harmful since the provided support is really eficient for decision makers. A new trend proposed by Jankovic, Zaraté et al. (2008) supports decision makers in their Collaborative Decision Making Processes. Cases Studies are also very interesting for researchers in order to learn about real cases and how DSS can be used in real life. This issue is organized in two sections: (1) Models for Collaborative Decision Making Processes, and (2) Cases Studies. Models for Collaborative Decision Making Processes The irst paper authored by Marija Jankovic and Pascale Zaraté proposes an analytical analysis through discrepancies and analogies in artiicial intelligence and engineering design approaches in addressing collaborative decision-making. The second paper written by Simon Buckingham Shum, Lorella Cannavacciuolo, Anna De Liddo, Luca Iandoli, and Ivana Quinto shows how using social network analysis can support a collective decision-making process. Cases Studies The third paper is authored by Carlos E. EscobarToledo and Héctor A. Martínez-Berumen, and proposes a strategic development of a decision making support system in a public R&D center. The fourth paper is written by Daniel J. Power, Roberta M. Roth, and Rex Karsten, and describes decision support for crisis incidents. The ifth paper, authored by Mary Daly and Frederic Adam, proposes organisational decision support maturity, using cases studies of Irish organisations. All these contributions provide new trend for Decision Support Systems Technologies. Pascale Zaraté Editor-in-Chief IJDSST ii RefeRenCeS Jankovic, M., Zaraté, P., Bocquet, J. C., & Le Cardinal, J. (2008). Collaborative Decision Making: Complementary Developments of a Model and an Architecture as a Tool Support. International Journal of Decision Support System Technology, 1(1), 35-45. Pascale Zaraté is a Professor at Toulouse 1 Capitole University. She conducts her researches at the IRIT laboratory (http://www.irit.fr). She holds a Ph.D. in Computer Sciences / Decision Support from the LAMSADE laboratory at the Paris Dauphine University, Paris (1991). She also holds a Master degree in Computer Science from the Paul Sabatier University, Toulouse, France (1986); as well as a Bachelors degree Toulouse, France (1982). Pascale Zaraté’s current research interests include: Decision Support Systems; distributed and asynchronous decision making processes; knowledge modelisation; cooperative knowledge based systems; cooperative decision making. She is the Editor-in-Chief of the International Journal of Decision Support System Technology (IGI Global). Since 2000, she is head of the Euro Working Group on DSS (www.euro-online.org). She published several studies and works: one book, edited two books, edited 11 special issues in several international journals, two proceedings of international conferences, 22 papers in several international journals, two papers in national journals, ive chapters in collective books, 26 papers in international conferences. She belongs the Editorial Scientiic Committee of three International Journals: Journal of Decision System (Lavoisier), ComSIS, Intelligent Decision Technologies (IOSPress). She was chairing the IFIP TC8/WG8.3 conference devoted to Collaborative Decision Making (http://www.irit.fr/CDM08). 2011 International Journal of Applied Industrial Engineering International Journal of Art, Culture and Design Technologies International Journal of Aviation Technology, Engineering and Management International Journal of Biomaterials Research and Engineering International Journal of Chemoinformatics and Chemical Engineering International Journal of Cloud Applications and Computing International Journal of Computer Vision and Image Processing International Journal of Computer-Assisted Language Learning and Teaching International Journal of Cyber Behavior, Psychology and Learning International Journal of Cyber Ethics in Education International Journal of Cyber Warfare and Terrorism International Journal of Fuzzy System Applications International Journal of Game-Based Learning International Journal of Information Retrieval Research International Journal of Intelligent Mechatronics and Robotics International Journal of Interactive Communication Systems and Technologies International Journal of Knowledge-Based Organizations International Journal of Manufacturing, Materials, and Mechanical Engineering International Journal of Measurement Technologies and Instrumentation Engineering International Journal of Online Marketing International Journal of Online Pedagogy and Course Design International Journal of People-Oriented Programming International Journal of Privacy and Health Information Management International Journal of Public and Private Healthcare Management and Economics International Journal of Quality Assurance in Engineering and Technology Education International Journal of Signs and Semiotic Systems International Journal of Social and Organizational Dynamics in IT International Journal of Space Technology Management and Innovation International Journal of Technology and Educational Marketing International Journal of User-Driven Healthcare International Journal of Wireless Networks and Broadband Technologies For subscription information, please visit: www.igi-global.com/journals International Journal of Decision Support System Technology, 3(2), 1-14, April-June 2011 1 Discrepancies and Analogies in Artiicial Intelligence and engineering Design Approaches in Addressing Collaborative Decision-Making Marija Jankovic, Ecole Centrale Paris, France Pascale Zaraté, Toulouse University, France AbStRACt One of the trends in the decision-making ield in the past 20 years has been the migration from individual decision-making to collective one. Several changes of working conditions inluenced this trend: geographical dispersion due to the business internationalisation, concurrent work in order to satisfy time delays, facilitation of the information sharing induced by the development of local area networks (LAN), and internet. This study examines the discrepancies and analogies in addressing the collaborative decision making in two scientiic ields: artiicial intelligence and engineering design. These two ields have different considerations and approaches in view to the decision-making support. This paper exposes a comparative study concerning two research studies, both decision support oriented: the irst one concerns the collaborative decision-making in early design stages in vehicle development projects (Jankovic, Bocquet, Stal Le Cardinal, & Bavoux, 2006) and the second one concerns the development of an architecture of a Cooperative decision Support Systems (CDSS) (Zaraté, 2005). Keywords: Collaborative Decision-Making, Cooperative DSS, DSS, Engineering Design, Product and Process Development 1. IntRoDuCtIon One of the trends in the decision-making field in the past 20 years have been the migration from individual decision-making to collective one (Shim et al., 2002). We can state several DOI: 10.4018/jdsst.2011040101 changes of working conditions that influenced this trend: geographical dispersion due to the business internationalisation, concurrent work in order to satisfy time delays, facilitation of the information sharing induced by the development of local area networks (LAN) and the internet. Several research fields address the issues of group, cooperative and collaborative decision- Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 2 International Journal of Decision Support System Technology, 3(2), 1-14, April-June 2011 making. In this study we particularly focus on two fields: artificial intelligence and engineering design. These two fields have different definitions concerning these types of decision-making and different approaches when addressing the problem of decision support. Therefore, we explore two research studies in order to make a comparative study of results and discuss further implication leading towards a more integrated approach. The first study concerns the collaborative decision-making in early design stages in vehicle development (Jankovic, Bocquet, Stal Le Cardinal, & Bavoux, 2006). The main objectives of this study are: 1) identifying key parameters for collaborative decision making in order to support the design team and 2) proposing an adequate support tool integrated into project management tools, already existing. The second study concerns a Cooperative Decision Support Framework (CDSF) (Zaraté, 2005). This framework is under development at the IRIT laboratory and was partially used in industrial context (i.e. Airbus). Main objective of this research work is to identify the discrepancies and analogies in addressing the decision-support concerning the two proposed research studies. Therefore, the authors try to address: 1) key parameters or data that are indentified and supported by both approaches and 2) differentiating elements with the aim of discussing potential further developments concerning decision-support tools. Identifying these discrepancies and analogies might be relevant in order to propose more integrated approaches and identify the difference in decision-making processes in different domains. Therefore, to address these issues we propose to discuss the definitions and approaches concerning cooperative and collaborative decision making, mostly in the field of artificial intelligence and engineering design. In the second part of this paper, we expose the characteristics and specificities of collaborative decision-making in early design stages, i.e. conceptual design. The third part of the paper gives an overview of the proposed cooperative decision support framework. At the very end of this paper, we propose to discuss these two approaches and conclusions concerning the differentiating elements in these two studies. 2. LIteRAtuRe RevIew The necessity of using the information technologies for supporting business processes and decision making has been growing in the past two or three decades (Kim, Godbole, Huang, Panchadhar, & W., 2004). Moreover, the development of the world wide web has been accelerating these process, introducing new application. For example, decision-support tools integrating different consensus management techniques to develop a solution (Kim, Godbole, Huang, Panchadhar, & W., 2004). This development of information technologies and the change i, working organisations have also been raising interest for supporting group decision-making and developing cooperative decision support systems (CSDS) (Zaraté, 2005). Nevertheless, the research literature does not give a uniform definition and characterization of different types of decision making, especially when it comes to cooperative and collaborative decision making. We can observe that difference between the artificial intelligence (AI) field addressing the decision-support systems and engineering design. In AI, the most integrated decision-making, in view to decisionmaking objectives sharing, is considered to be cooperative decision making where different actors have the same objectives and might be geographically distributed (Zaraté, 2005). In engineering design, due to the system engineering objectives, the most integrated decision-making process is considered to be the collaborative decision-making where decision makers do have common global objectives but also have objectives in the decision-making process that concern their own design domain (Jankovic, 2006). This is due to the complexity of design processes and the necessity to perform the cascading of design objectives and therefore the decision-making objectives. Some of the work in AI fields when speaking of collaborative decision-making Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 1-14, April-June 2011 3 considers distributed asynchronous decisionmaking (Aldunate, Pena-Mora, & Robinson, 2005; Chim, Anumba, & Carrillo, 2004; Cil, Alpturk, & Yazgan, 2005), which is often similar to the definition given for cooperative decision-making. Some research studies refer to it as multi-actor decision-making where actors have different goals (Karacapilidis & Papadias, 2001; Panzarasa, Jennings, & Norman, 2002). Panzarasa and Jennings (Panzarasa, Jennings, & Norman, 2002) consider collaborative decision-making as a multi-agent socio-cognitive process. Thus they incorporate beliefs, goals, desires, intentions, and preferences in what they call mental modelling. The authors also adopt a prescriptive approach in order to give a set of possible actions at every step of collaborative decision-making. The model is developed using social mental shaping, the process by which the mere social nature of agents may impact upon their mental states and motivate their behaviour. Their collaborative decision-making model consists of four phases: the practical starting-point, group generation, social practical reasoning, and negotiation. Another very important research stream is considering the decision-making as an argumentation process in order to integrate and support conflict prevention. Karacapidilis and Papadis (2001) consider collaborative decisionmaking to be a process of “collaboratively considering alternative understandings of the problem, competing interests, priorities and constraints”. Their research work towards the definition of Collaborative Decision Support Systems (CDSS) is based upon the definition of Kreamer (1988). Karakapidilis and Papadias (2001) develop the “Hermes” system to support collaborative decision-making. They define this system as a “generic active system that efficiently captures users’ rationale, stimulates knowledge elicitation and argumentation on the issues under consideration, while it constantly (and automatically) checks for inconsistencies among users preferences and considers the whole set of the argumentation items asserted to update the discourse status”. Munkvold, Eim, and Husby (2005) focuses on collaborative IS decision-making processes: “complex decision-making processes, involving multiple stakeholder groups”. This research work concerns the specification, selection and acquisition of a new IT solution for collaboration and information management. The authors consider this process to be a collaborative decision-making process. One of the points underlined by the research study is also identification of potential challenges when using different collaborative technologies in the decision-making process:: 1) ensuring the continuity in the project; 2) ensuring effective communication among different stakeholder groups and 3) gaining involvement and commitment from the business areas. In order to build a decision support system, Hamel, Pinson, and Picard (2005) underlines the need to explicitly model behaviour including the interactions as well as the actions of the actors. The authors propose the use of a Multi-Agent Based Simulation (MABS) in order to take into account the interactions between individuals. This is because MABS has the ability to cope with simple entities as well as the organisations and interactions between entities and groups. The ACKA model that is proposed is based upon two principles: the definition of modelling and conceptual roles and a definition of the decision-making process as a series of interactions between stakeholders. As for the engineering design fields, the definition as well as the specificities of the collaborative decision-making are a consequence of the definition of the collaboration process and integration of systems engineering tools in the design (Browning & Eppinger, 2002; Lindemann, Maurer, & Braun, 2008; Wyatt, Eckert, & Clarkson, 2009) The collaborative decision-support tools are mostly oriented towards obtaining a reasonable trade-offs in the design process (Holley, Yannou, & Jankovic, 2010) integrating tools like DSM (design structure matrices) based tools. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 4 International Journal of Decision Support System Technology, 3(2), 1-14, April-June 2011 3. CoLLAboRAtIve DeCISIon-MAkIng In vehICLe DeveLoPMent PRojeCtS This first study concerns the collaborative decision-making in early stages of the vehicle development process (Jankovic, 2006; Jankovic, Stal Le Cardinal, & Bocquet, 2009). The proposed model to support the design process concerns two major aims: 1) identifying key parameters for collaborative decision making in order to support the design team and 2) proposing an adequate support tool integrated into project management tools. The developed model is a base for project design support tool already discussed in previous work of authors (Jankovic, Stal Le Cardinal, & Bocquet, 2009). A. Industrial Context: Project Definition Phase The New Product and Process Development (NPPD) is one of the key processes contributing to the enterprise success and future development (Marxt & Hacklin, 2004). The beginning of this process is the identification of client needs resulting from the market research. This first phase, the conceptual design phase, has a crucial importance for the overall design process. It is already widely accepted that the conceptual design process consists of generation of concepts, exploitation of these concepts and evaluation (Pahl, Beitz, & Wallace, 1996; Ulrich & Eppinger, 1995). The extreme importance of these stages is underlined by: 1) the fact that it is a value definition stage (80% of all life cycle costs are engaged in this stage), 2) the changes in this stage impacting the whole product life cycle, like manufacturing or distribution processes, and 3) the innovation which represents the added value generated and integrated within this phase. The Project Definition phase is at the very beginning of the conceptual design and is considered to be very complex. During this phase all aspects of one project are to be defined. Project organisation and management are set up throughout the fulfillment of functions assigned to every project team member. The beginning of this phase is a clear definition of client’s needs defined by the marketing department addressing the characteristics of the market share that is targeted. Based upon these definitions, different departments defined strategic orientations for the given development project. The involved departments are marketing, production, innovation, and strategy (Figure 1). The project team’s mission is to transcribe these strategic orientations onto the project objectives, to decompose them on sub-objectives and discern their global incoherence in order to propose coherent ones. In order to fulfill its mission, the project team has to collaborate with several departments that are responsible for the definition of global enterprise orientations and the ones that detain the enter- Figure 1. Project objectives: context definition (JDS) Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 1-14, April-June 2011 5 prise know-how (knowledge poles). The definition of project objectives is a difficult task due to the compromise that must be found between enterprise knowledge and enterprise ambitions, i.e. strategic orientations. Because a vehicle is a complex system, one of the difficulties of this phase is that there are more than hundred objectives to take into account on the vehicle system level. In order to consider more formal decision-support tools, in this study we have also identified different decision making objectives. More than 150 objectives were identified. The correlations between these objectives are not often determined, therefore there is no certainty about the influence that one objective could have on another. Furthermore, the Project Definition phase is crucial for the introduction of innovation. In this phase, the project team has to select the innovations that are interesting according to the type of vehicle but also feasible for the given time line. This innovation introduction increases the difficulty of identification of possible correlations between project objectives. b. Decision-Making in the Project Definition Phase The Project Definition phase is also a collaborative decision-making phase. The decision makers in the Project Definition phase are experts for one aspect of the product development (motorisation, economic aspects, architecture, etc.), each of them having different information and knowledge concerning the problem (Figure 2). The difficulties in collaborative decisionmaking in this phase come from: 1. The difficulties to asses and evaluate the uncertainties that are inherent to the early design phases, Figure 2. Collaborative decision-making (these) Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 6 International Journal of Decision Support System Technology, 3(2), 1-14, April-June 2011 2. 3. The difficulties of evaluating the potential impacts of different decision-making alternatives and The necessity to cascade design objectives which induces that the decision-makers have common objectives but also their own, domain specific, objectives that are often contradictory. The collaborative decision-making represents a rich way for decision alternatives generation. In the design project, it is already accepted that the collaborative decision-making is creating favorable conditions for synergy development. Previous studies on collaboration processes in engineering design have already explored the benefits and difficulties in these processes (Rose, 2000). Collaborative decision-making is also a necessary approach in design projects. Development of new products demands a very large body of knowledge in different scientific domains and a comprehension of interactions between these domains. Therefore, it is a decision-making process that involves a large number of decision-making actors having a specific domain oriented point of view and different knowledge concerning the same problem. Moreover, due to this fact, the collaborative decision-making is considered to be a better informed process. Upper statements sustain the possibility of a better-quality decision-making process. As actors all together have knowledge in several project fields, more information and are influenced by synergy effects, the collaborative decision-making process is resulting in a bigger number of alternatives and thus a possibility of a higher quality of decision. Nevertheless, this type of decision-making, even though showing great advantages is not without some inconveniences and problems that must be taken into account: 1. 2. Every decision maker has preferences concerning the decision. These preferences are due to the fact that each actor has his own domain specific objectives. We can see some problems of different value judgments that every decision maker 3. 4. has for the same decision. The decisionmakers have different backgrounds and different information, and therefore have different preferences. Negotiations among themselves are then necessary. The specificity of collaborative decision making process is the existence of different objectives. Every actor has specific objectives that are important to satisfy, otherwise the project has chance to be stopped. As collaborative decision-making is a multi-actor process, the problem of postcontrol is an important issue. Development projects are situated in the dynamic environment and it is necessary to follow-up the coherence between the chosen solution(s) and developing solutions. 4. ConCePtuAL MoDeL of CoLLAboRAtIve DeCISIon-MAkIng The collaborative decision-making model is developed in order to identify key parameters in collaborative decision making and therefore identify intrinsic elements and data necessary in the process. This model is based upon the systems theory developed by Le Moigne (1990). Le Moigne defines the concept of General System as a representation of an active phenomenon comprehended as identifiable by its project in an active environment. Therefore we have developed four different points of views in our collaborative decision-making model: Objectives View, Process View, Transformation View and Environment View. These views are interdependent and are not to be taken separately. The model and its implementation have already been object of publications (Jankovic, Stal Le Cardinal, & Bocquet, 2009). Therefore, we will expose only the essential elements necessary for the understanding of the model. Objectives View concerns objectives in collaborative decision-making processes. This view takes into account different objectives influencing the process, as well as their relationships. The collaborative decision-making objectives are also influenced by actors’ preferences. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 1-14, April-June 2011 7 Environment View relates to the relevant information concerning the development process, in our case the conceptual design phase, and captures the knowledge on one company. It is considered to represent the company “knowhow”. Three different environments influence collaborative decisions in the New Product and Process Development: 1. 2. 3. Decision environment, Project environment, Enterprise environment. Each of these environments is identified by its context, determining the influencing factors of collaborative decision-making and different actors relevant for collaborative decision-making. Therefore, Decision Environment is identified by decision-making context and actors participating in the collaborative decision-making process. This environment is influenced by the Project Environment, equally defined by Project Context and Project Influence Groups. Project and Decision Environment are influenced by the Enterprise Environment, identified by its context and actors. The Process View represents the general decision-making process proposed for the conceptual design phase. Collaborative decisionmaking process is a complex human-interaction and human-cognition process. Therefore, we have identified 3 general phases of collaborative decision-making process: 1. 2. 3. Identification of the need for decision-making, Decision-making phase, Implementation and Evaluation. In the model we underline that every process implies the utilisation of the resources, human or material. The main objective of the clearly defined collaborative decision-making process is to ensure its quality. The correlation between the formalization of the decisionmaking processes and its outcome has been explored and confirmed by several studies (Courtright, 1978; Kameda & Sugimori, 1993; Neck & Moorhead, 1995). The Transformation View represents a dynamic view of the decision-making process. The main objective is to support information flow and changes that can be: spatial (transfer of information) or formal (transformation of the information into new information). These transformations can be grouped in two groups: 1. 2. Preparatory transformations and Implementing transformations. Preparatory transformations represent information required in order to dispose to necessary elements for decision making. Implementing transformations represent information concerning transformation the implementation of the made decision. 5. A CooPeRAtIve DeCISIon SuPPoRt fRAMewoRk Cooperative Decision Making is seen as a process and in order to support it efficiently, we propose a general framework composed by several tools. For this kind of framework, cooperation is defined at two levels: Man/Machine cooperation and Cooperation among several decision makers. In her work, Zaraté (2005) proposes a Cooperative Decision Support framework. It is composed by several tools: − − − − An interpersonal communication management system, A task management system, A knowledge management tool, A dynamical man/machine interactions management tool. This framework is described in the Figure 3. The interpersonal communication management tool is able as in every kind of Computer Supported Collaborative Work (CSCW) tool, Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 8 International Journal of Decision Support System Technology, 3(2), 1-14, April-June 2011 Figure 3. Architecture for the cooperative decision support framework to help users and decision-makers in the interactions among themselves. The dynamical man/machine interactions management tool guides the users in their processes of solving problems in order to solve the misunderstanding problems. This package is able to avoid misunderstanding between the system and the user by proposing new solutions and is under development. The knowledge management tool storages the previous decision made by the group or by other groups. The system proposes solutions or part of solutions to the group in very similar situations. In the case of a different situation the system must be able to propose the solution the most appropriated and the users could accept it or not. This tool is based on a knowledge management tool. Based on the DSSs’ architecture defined by Sprague and Carlson (1982), the system includes also a Data Base Management System, a Model Base Management System. Nevertheless, this system is based on the development of Knowledge Based System and more particularly Cooperative Knowledge Based System. Thus, the proposed system includes a Knowledge Base. The task management tool is based on a Cooperative Knowledge Based System developed at the IRIT laboratory. This Cooperative Knowledge Based Architecture is developed through libraries of models: users’ models, domain models (or problems models) and contextual models. The calculation of the proposed solutions is based on several techniques: planning tools (Camilleri, 2000), linear programming (Dargam, Gachet, Zaraté, & Barnhart, 2004). The main usage principle of this kind of tool is based on the interaction between the system and the users. The system proposes a solution to the group, the group takes in charge some tasks and then the system recalculates a new solution and proposes the new one to the group and etc. The problem or the decision to made is solved steps by steps each actors (system and users) solving parts of the problem (Camilleri, Soubie, & Zaraté 2008). This tool has for objective to propose solutions or part of solutions to users. It calculates the scheduling of tasks and sub-tasks and each role that are assigned to each tasks. It also proposes an assignment of tasks to users or to the system itself. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 1-14, April-June 2011 9 The main idea of this work is to find complementary aspects of the conceptual model and the Collaborative Decision Support Framework. 6. CoMPLeMentARy StuDy of the MoDeL AnD the CoLLAboRAtIve DeCISIon SuPPoRt fRAMewoRk The collaborative decision-making model represented using the UML 2.0 modeling language is represented in the Figure 4. The main question is this comparison study is what parameter or data are supported by Cooperative Decision Support framework (CDSF) and what is to be done in order to extend this support in the conceptual design phase. In order to address these issues, a quick overview of different data is given as well as the possibility of their support. This is represented in the Figure 4. The objectives view gives a global overview of the collaborative decision-making objectives and their relationships. As we already stated in the part 2 of this paper, the collaborative decisions are decisions for which actors have different operational objectives. Therefore, the class Actor is associated with the class Operational_Objective. One actor can have one or more operational objectives and one objective can concern several actors. Each actor has a role in the collaborative decision-making Figure 4. Data supported by CDSF Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 10 International Journal of Decision Support System Technology, 3(2), 1-14, April-June 2011 process, thus the Actor class has an association relationship with the class Role. The Collaborative_Decision_Objective class represents an aggregation class of several Operational_Objective. They are defined by their name (Decision_Obj_Name), their value (Defined_Value) and the deadline for their realisation (Objective_Milestone) so that the project does not have any delay. Collaborative decision-making objectives are defined in view to satisfy different client needs identified during the NPPD. Therefore the Collaborative_Decision_Objective class has an association relationship with the class Client. The Enterprise_Goal is a link class of the relationship between the Collaborative_Decision_Objective and the Client class. As the NPPD process is a collaborative process and some objectives are given by different collaborators (like production or distribution), the class Actor represents an inheritance class of the class Client. The attributes of the Collaborative_Decision_Making_Process are Name, the Decision_ Importance and the Decision_Making_Phase. This class is composed by Ressources class, itself composed of Material_Resources and Human Resources classes. The Human_ressources is a generalised class of the Actor class. The Actor class has a directed association relationship with the Collaborative_Decision_Making class named “participates in”. The Collaborative_Decision_Making_ Process class has an association relationship with the Data and the Plan class. The Plan class is composed of different activities represented with the Tasks class. The Data class is defined by several attributes: Data_Name, Data_Responsible – the person qualified to give such an information, Data_Type – if the data is used in the preparation for the decision or in the implementation phase, Data_Storage – the place where this data can be found and Data_Criticity – the probability of obtaining the data on time for the decision. The Plan class has Plan_Name, Plan_Type – preparatory or implementing plan, Responsible and Plan_Storage – indication of the place where the given plan can be found. The Plan class is a parent class for the Task class (Task_Type, Task_Name, Criticality, Task_Responsible). Both Objectives and Transformations Views of the model are easily supported by Data Base management system. It is also interesting that the part concerning the design objectives in engineering studies is mostly oriented towards an expert-based systems or rule-based systems supporting the decision in design. Therefore, it might be interesting to combine theses two approaches in order to have a more efficient system addressing both points of views: decision and design. Environment view is only partially presented because this view concerns the knowledge of the collaborative decision-making process as well as the new product development process. This view represents the “know-how” of the organization addressing the issues of project organization, communication channels, and enterprise decision-making process. As it is the “know-how” it is very difficult to present it using the UML modeling language. Therefore, it may be useful to consider other languages for it representation. Nevertheless we consider possible to develop the Knowledge Base system supporting this kind of problems in the collaborative decision-making process. In short the information presented in the class diagram of the collaborative decisionmaking process we can identify different parts of the CDSF supporting the given information: information contained in the Transformations view can be supported by the Model Base Management System, Objectives and Collaborative Decision Making Process by the Data Base Management System and the Environment view by the Knowledge Base. The corresponding schema is given on the Figure 5. The information concerning the Objectives view and the Process view can be supported by the Data Base Management System (DBMS). This system contains all information concerning collaborative decision-making objectives, different actors participating and their roles in the decision-making process, different activities in the decision-making, as well as necessary resources for the decision. The Model Base Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 1-14, April-June 2011 11 Figure 5. CDSF support for collaborative decision making in engineering design Management System (MBMS) contains different model at the decision-makers’ disposal. Therefore it can support the Transformations View, addressing the issue of different plans and the tasks before and after the collaborative decision-making. The models can be proposed regarding the operational needs of one project team. 7. ConCLuSIon Artificial Intelligence and engineering design have different definitions and considerations concerning the cooperative and collaborative decision making. Whilst for AI the most integrated approach in term of sharing the decision-making objectives is cooperative decision-making, for engineering design field it is the collaborative decision-making due to the particularity of the product development process. Moreover, the characteristics of cooperative and collaborative decision-making are not the same in these two fields. Two research studies have been taken into account: one concerning the collaborative decision-making in early design stages in vehicle development and the second one concerning the proposition of the general Cooperative Decision Support framework. Main objective of this research work is to identify the discrepancies and analogies in addressing the decision-support concerning the two proposed research studies. Therefore, the authors try to address: 1) key parameters or data that are indentified and supported by both approaches and 2) differentiating elements with the aim of discussing potential further developments concerning decision-support tools. The comparison of this model and CSDF has pointed out that the information contained in the Objectives and Process view can be supported Data Base Management System (DBMS). The Transformations view can be supported by different model at the decisionmakers’ disposal in the Model Base Management System (MBMS). As for the Environment view, even though this part is only partially captured in this model and corresponds to the know-how, we consider to be possible to develop a Knowledge Management (KM) database in order to organise and capitalise the knowledge. The class diagram formalism has been used in order to structure the Data Base, the Model Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 12 International Journal of Decision Support System Technology, 3(2), 1-14, April-June 2011 Base and the Knowledge of the Collaborative Decision Support Framework. The authors can retrieve several conclusions from this comparative study. Firstly, the information identified for the collaborative decision-making support can be entirely supported by the CSDF. We also show that the type of information classified in different views of the model correspond to well identified parts of the framework. Secondly, we can conclude the possibility to apply the CSDF during the project management in the Project Definition phase. Nevertheless, some issues have to be dealt with: the cost of such a platform and ergonomic aspect that user oriented. Moreover, it seems interesting to integrate some of the expert-based systems or rule-base systems that address the collaborative decision-making support in design. Thirdly, even though the collaborative decision-making model was developed for the purpose of project management, the construction of the model with the types of information identified in the case of collaborative decision-making support can be generalised to other decision contexts and situation. RefeRenCeS Aldunate, R., Pena-Mora, F., & Robinson, G. (2005). Collaborative distributed decision making for large scale disaster relief operations: Drawing analogies from robust natural systems. Complexity, 11(2), 28–38. doi:10.1002/cplx.20106 Browning, T. R., & Eppinger, S. D. (2002). Modeling impacts of process architecture on cost and schedule risk in product development. IEEE Transactions on Engineering Management, 49(4), 428–442. doi:10.1109/TEM.2002.806709 Camilleri, G. (2000). Une approche, basée sur les plans, de la communication dans les systèmes à base de connaissances coopératif. Toulouse, France: Université Paul Sabatier. Chim, M. Y., Anumba, C. J., & Carrillo, P. M. (2004). Internet-based collaborative decision-making system for construction. Advances in Engineering Software, 35(6), 357–371. doi:10.1016/j.advengsoft.2004.03.007 Cil, I., Alpturk, O., & Yazgan, H. R. (2005). A new collaborative system framework based on a multiple perspective approach: InteliTeam. Decision Support Systems Collaborative Work and Knowledge Management, 39(4), 619–641. Courtright, J. A. (1978). A laboratory investigation of groupthink. Communication Monographs, 45(3), 229–246. doi:10.1080/03637757809375968 Dargam, F., Gachet, A., Zaraté, P., & Barnhart, T. (2004). DSSs for planning distance education: A case study. In Proceedings of the International Conference on Decision Support in an Uncertain and Complex World, Prato, Italy (pp. 169-179). Hamel, A., Pinson, S., & Picard, M. (2005, September 19-22). A new approach to agency in a collaborative decision- making process. In Proceedings of the International Conference on Intelligent Agent Technology (pp. 273-276). Holley, V., Yannou, B., & Jankovic, M. (2010). Using quality function deployment (QFD) to bring design department voice in the choice of concepts. Paper presented at the 12th International DSM Conference, Cambridge, UK. Jankovic, M. (2006). Collaborative decision making in new product development. Application to the car industry. Unpublished doctoral dissertation, Ecole Centrale Paris, France. Jankovic, M., Bocquet, J.-C., Stal Le Cardinal, J., & Bavoux, J.-M. (2006, May 15 - 18). Integral collaborative decision model in order to support project definition phase management. In Proceedings of the 9th International Design Conference, Dubrovnik, Croatia (pp. 1351-1358). Jankovic, M., Stal Le Cardinal, J., & Bocquet, J.-C. (2009). Collaborative decision in design project management. A particular focus on automotive industry. Journal of Decision Systems. Kameda, T., & Sugimori, S. (1993). Psychological entrapment in group decision making: An assigned decision rule and a groupthink phenomenon. Journal of Personality and Social Psychology, 65(2), 282–292. doi:10.1037/0022-3514.65.2.282 Karacapilidis, N., & Papadias, D. (2001). Computer supported argumentation and collaborative decision making: The HERMES system. Information Systems, 26(4), 259–277. doi:10.1016/S03064379(01)00020-5 Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 1-14, April-June 2011 13 Kim, S.-Y., Godbole, A., Huang, R., Panchadhar, R., & W., S. W. (2004). Toward an integrated humancentered knowledge-based collaborative decision making system. In Proceedings of the IEEE International Conference on Information Reuse and Integration. Kreamer, K. L., & King, J. L. (1988). Computer-based systems for cooperative work and group decision making. ACM Computing Surveys, 20(2), 115–146. doi:10.1145/46157.46158 Le Moigne, J.-L. (1990). La modélisation des systèmes complexes. Paris, France: Dunod. Lindemann, U., Maurer, M., & Braun, T. (2008). Structural complexity management: An approach for the field of product design. Berlin, Germany: Springer-Verlag. Marxt, C., & Hacklin, F. (2004, May 18-21). Design, product development, innovation: All the same in the end? A short discussion on terminology. In Proceedings of the 8th International Design Conference, Dubrovnik, Croatia (pp. 377-382). Munkvold, B., Eim, K., & Husby, O. (2005). Collaborative IS decision- making: Analyzing decision process characteristics and technology support. In H. Fuks, S. Lukosch, & A. C. Salgado (Eds.), Proceedings of the 11th International Workshop on Groupware: Design, Implementation and Use (LNCS 3706, pp. 292-307). Neck, C. P., & Moorhead, G. (1995). Groupthink remodeled: The importance of leadership, time pressure, and methodical decision-making procedures. Human Relations, 48(5), 537–557. doi:10.1177/001872679504800505 Pahl, G., Beitz, W., & Wallace, K. (1996). Engineering design: A systematic approach. London, UK: Springer. Panzarasa, P., Jennings, N. R., & Norman, T. J. (2002). Formalizing collaborative decision-making and practical reasoning in multi-agent systems. Journal of Logic and Computation, 12(1), 55–117. doi:10.1093/logcom/12.1.55 Rose, B. (2000). Proposition d’un référentiel support à la conception collaborative: CO²MED (Collaborative conflict management in engineering design), Prototype logiciel dans le cadre du projet IPPOP. Unpublished doctoral dissertation, U.F.R. Sciences et Techniques Mathématiques, Informatique et Automatique, Brest, France. Shim, J. P., Warkentin, M., Courtney, J. F., Power, D. J., Sharda, R., & Carlsson, C. (2002). Past, present, and future of decision support technology. Decision Support Systems, 33(2), 111–126. doi:10.1016/ S0167-9236(01)00139-7 Ulrich, K. T., & Eppinger, S. D. (1995). Product design and development. New York, NY: McGraw-Hill. Wyatt, D. F., Eckert, C. M., & Clarkson, P. J. (2009, August 24-27). Design of product architectures in incrementally developed complex products. In Proceedings of the 17th International Conference on Engineering Design, Stanford, CA. Zaraté, P. (2005). Des systèmes interactifs d’aide à la décision aux systèmes coopératifs d’aide à la décision. Unpublished doctoral dissertation, Institut National Polytechnique de Toulouse, Toulouse, France. Marija Jankovic is an Assistant Professor at Ecole Centrale Paris, Industrial Engineering Department. She holds PhD degree in Industrial Engineering titled “Collaborative decision making in new product development. Application in the automotive industry.” This research project was done with the collaboration of PSA Peugeot Citroen. The main research goals concern enhancing the performance and quality of early stages of new product and/or service development. The systems’ thinking is used as base for complex system modeling in addressing new product and/or service development process. The principal research fields concern quality engineering (using approaches like applied statistics) and system modeling on order to simulate, predict and improve global system performances in collaborative engineering environment. Marija Jankovic is also an Assistant Director of a Master Program in Design Engineering and Innovation at Ecole Centrale Paris. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 14 International Journal of Decision Support System Technology, 3(2), 1-14, April-June 2011 Pascale Zaraté is a Professor at Toulouse 1 Capitole University. She conducts her researches at the IRIT laboratory (http://www.irit.fr). She holds a Ph.D. in Computer Sciences / Decision Support from the LAMSADE laboratory at the Paris Dauphine University, Paris (1991). She also holds a Master degree in Computer Science from the Paul Sabatier University, Toulouse, France (1986); as well as a Bachelors degree Toulouse, France (1982). Pascale Zaraté’s current research interests include: Decision Support Systems; distributed and asynchronous decision making processes; knowledge modelisation; cooperative knowledge based systems; cooperative decision making. She is the Editor-in-Chief of the International Journal of Decision Support System Technology (IGI Global). Since 2000, she is head of the Euro Working Group on DSS (www.euro-online.org). She published several studies and works: one book, edited two books, edited 11 special issues in several international journals, two proceedings of international conferences, 22 papers in several international journals, two papers in national journals, 5 chapters in collective books, 26 papers in international conferences. She belongs the Editorial Scientific Committee of three International Journals: Journal of Decision System (Lavoisier), ComSIS, Intelligent Decision Technologies (IOSPress). She was chairing the IFIP TC8/WG8.3 conference devoted to Collaborative Decision Making (http://www.irit.fr/CDM08). Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 15-31, April-June 2011 15 using Social network Analysis to Support Collective Decision-Making Process Simon Buckingham Shum, The Open University, UK Lorella Cannavacciuolo, University of Naples Federico II, Italy Anna De Liddo, The Open University, UK Luca Iandoli, University of Naples Federico II, Italy Ivana Quinto, University of Naples Federico II, Italy AbStRACt Current traditional technologies, while enabling effective knowledge sharing and accumulation, seem to be less supportive of knowledge organization, use and consensus formation, as well as of collaborative decision making process. To address these limitations and thus to better foster collective decision-making around complex and controversial problems, a new family of tools is emerging able to support more structured knowledge representations known as collaborative argument mapping tools. This paper argues that online collaborative argumentation has the rather unique feature of combining knowledge organization with social mapping and that such a combination can provide interesting insights on the social processes activated within a collaborative decision making initiative. In particular, the authors investigate how Social Network Analysis can be used for the analysis of the collective argumentation process to study the structural properties of the concepts and social networks emerging from users’ interaction. Using Cohere, an online platform designed to support collaborative argumentation, some empirical indings obtained from two use cases are presented. Keywords: Argument Mapping Tool, Concept Network Analysis, Decision-Making Process, Group Decision Support System, Knowledge Management, Online Collaborative Tools, Social and Concept Network Visualization, Social Network Analysis IntRoDuCtIon In current dynamic and turbulent environment, organizations increasingly have to deal with complex problems. One way to deal with such increasing complexity is to create multidisciDOI: 10.4018/jdsst.2011040102 plinary and multi-stakeholders working groups in order to have diverse perspectives on the problem, different individual knowledge and skills, and alternatives approaches for problem solving (Beers et al., 2006). As highlighted in literature, in the last decades many organizational decisions migrated from individual decisions to distributed decisions based on the contributions Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 16 International Journal of Decision Support System Technology, 3(2), 15-31, April-June 2011 offered by large, diverse groups of individuals within a firm or even from multiple firms (Shim et al., 2002). Nowadays, cooperative and collaborative decision making is achieving increasing importance in organizations (Jankovic et al., 2009). Collective decision making processes are mainly the consequence of environmental complexity which compel firms to make decisions on increasing complex problems, and the evolution of information and communication technologies which have made feasible interaction and communication among several and, in many cases, dispersed workers at creasing, virtually zero costs. These expectations are confirmed by anecdotics as well as empirical evidence that under the right circumstances collective decisions made by large groups of competent and motivated individuals can be more effective than single experts’ or small group decision making (Page, 2008). According to Zaraté and Soubie (2004), it is possible to identify four different types of collective decision-making processes on the basis of two main criteria, namely time and space: • • • • Face to face decision making: different decision makers, involved in the decisional process, are physically in the same place and at the same time. Distributed synchronous decision making: different decision makers, involved in the decisional process, are not located in the same place but work together at the same time. Asynchronous decision making: different decision makers, involved in the decisional process, come in a specific place (also virtual place) to make decisions but not at the same time. Distributed asynchronous decision making: different decision makers, involved in the decisional process, do not work together at the same time and in the same place. Naturally, the collaborative decision making process in synchronous way is the richest, in terms of information exchanged among team members, and the most common way in organizations to make a decision especially in complex projects. Face to face group decision making is however prone to group thinking and well known group decision making pitfalls as information cascades, hidden profiles and polarization (Sunstein, 2006). The advent of Internet has given rise to many new and enriched applications of existing technologies able to better make deliberation and decision making process more efficient and effective. In particular, Internet has made feasible for organizations the overcoming of time and space constraints (Cramton, 2001) and to draw together knowledgeable individuals and wider amount of information sources on a scale that was unimaginable few years ago. The main Web’s impact on deliberation and decision making processes has been to increase information access and foster more rapid and deeper dissemination of relevant information to all decision makers implied in the process even if geographically dispersed as well as to make them less costly (Shim et al., 2002). Moreover, web-based tools promote more consistent and well-supported decision making by enabling a larger number of users to participate and, thus, promoting a broader exploration of the solution space. On the Internet there is a very large variety of tools to collect information and knowledge provided by many dispersed users in a cheap and efficient way. The most commonly used web-based technologies are wikis, blogs, and discussion forums. Although they are noticeably basic compared to group decision support systems (henceforth GDSS), they allow large groups of users to achieve outstanding results in knowledge sharing and accumulation. The successful emergence of web-based peer production platform such as Wikipedia and Linux, has encouraged an increasing number of organizations to exploit these technologies in their knowledge management processes (McAfee, 2006). However, current traditional technologies, while enabling effective knowledge sharing and accumulation, seem to be less supportive of knowledge organization, use and consensus formation (Iandoli et al., 2009). Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 15-31, April-June 2011 17 To address these limitations and thus to better foster collective decision-making around complex and controversial problems, we are now seeing the emergence of tools able to support more structured knowledge representations; among those, this paper focuses on argumentation technologies and their application to support distributed asynchronous deliberation and decision making process. Online argumentation platforms support users in the collaborative construction of arguments maps related to an issue object of discussion. The maps organize issues, possible solutions and their associated rationales expressed in terms of chains of pros and cons. As we discuss in the next section, argumentation technologies can help to overcome some of the limitations of current online collaborative technologies such as conversational tools (e.g. forums, wikis, blogs), basically by providing users with the possibility of building collective and reusable knowledge representations. In addition, as other web 2.0 like tools, an argumentation platform is able to keep track of the relationships that develop during the discussion through which users connect to other users and to different subtopics arising during the debate. We argue in this paper that collaborative online argumentation has the rather unique feature of combining knowledge organization with social mapping and that such a combination can provide interesting insights on the social processes activated within a collaborative decision making initiative. In particular, we explore how the analysis of the collective argumentation process can be used to investigate the structural properties of the concept and social networks emerging from the interaction. We present some empirical findings obtained from two use cases in which participants were involved in a collective deliberation task supported by Cohere, an online argumentation platform developed at the Knowledge Media Institute at The Open University of Milton Keynes (Buckingham Shum, 2008; De Liddo & Buckingham Shum, 2010.). Finally, we discuss the implications for the development and design and analysis of GDSS arising from an investigation of the concept and social networks emerging in the decision making process. StRuCtuRIng onLIne DeCISIon-MAkIng DebAte thRough ARguMentAtIon teChnoLogy Much effort has been spent in the development of tools to facilitate communication and indirectly support members of a group to discuss how to solve complex problems. Technological developments have continually enabled the development of more collaboration tools. The advent of Internet and World Wide Web has allowed organizations to use several diverse tools for many and different aims, such as e-mails, chat and teleconference to conduct real-time or asynchronous dialogues with external and internal employees, managers and shareholders, websites and newsgroups to gather information about stakeholder concerns and customer requirements, discussion forums, wikis and blogs to collect both technical and non-technical data about any organizational problem or need (Courtney, 2001). While such tools have been remarkably successful at enabling knowledge sharing at an unprecedented scale, they have also been criticized in many respects. One of the main criticisms regards their inability to organize and structure knowledge in a coherent and reliable way. Indeed, it is notorious that information collected through online collaboration, such as forum, wiki and blog, has often been considered overwhelming, redundant and of disputable quality, especially when produced in the course of controversial debates. Low quality and redundancy make hard to locate and isolate relevant and useful knowledge. Moreover, these tools do not inherently encourage or enforce any standard concerning what constitutes valid argumentation, so postings are often bias- rather than evidence- or logic-based. Finally, collaboration does not necessarily entail a convergence toward a shared decision or the definition of a ranking of the available choices. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 18 International Journal of Decision Support System Technology, 3(2), 15-31, April-June 2011 These shortcomings are major obstacles for the use of collaboration as tools to support group decision-making. Other kind of web-based technologies such as prediction markets and e-voting have been proved to be effective at aggregating individual opinions to determine the most widely/strongly held view (Wolfers & Zitzewitz, 2004), but provide little or no support for identifying what the alternatives selected among should be, or what their pros and cons are. In order to address these shortcomings, alternatives technologies, able to support a more structured knowledge and conflicting points of views representation, have been developed. In this paper we focus on argument mapping tools. These technologies try to fill the above mentioned gaps by helping groups to represent a debate as a visual map composed of a set of issues to be answered, positions (or ideas) as alternative solutions to issues and supportive or challenging arguments about proposed ideas. Debate is summarized into a visual map connecting Issues, Positions and Arguments through labelled links such as supports to, objects-to, suggested by, replaces (Figure 1). These tools are now finding application in several forms of knowledge work requiring clear thinking and debate, including learning (Toth, Suthers, & Lesgold, 2002), deliberation (van Gelder, 2003), knowledge management (Tergan, 2003), participatory planning (De Liddo & Buckingham Shum, 2010). Such tools allow users to represent complex reasoning in a concise, easy to follow, clear and unambiguous way, making the logic behind an analysis more evident. The main feature of argument mapping tools is to encourage careful critical thinking (Buckingham Shum et al., 2006; Van Gelder, 2007), by implicitly requiring that users express the evidence and logic in favour of the options they prefer. Moreover, the arguments are captured in a compact form that makes easy to understand what has been discussed and, if desired, add contributions to it without needless duplication. In this way they are expected to enable increased synergy across group members as well as non-redundant knowledge accumulation over time. Such tools are supposed to be particularly suited to foster deliberation and decision making processes around complex problems as they allow users to represent con- Figure 1. Argument Map (Source: http://labspace.open.ac.uk/mod/resource/view.php?id=350370) Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 15-31, April-June 2011 19 tentious and/or competing point of views in coherent structures made up of alternative positions on an issue at stake with their associated chains of pros and cons arguments. Numerous examples of web-based collaborative argument mapping tools exist (for a review of current tools see http://events.kmi. open.ac.uk/essence/tools/) which allow users to navigate, co-create and edit an argument map. Online collaborative argument mapping tools are designed to help diverse and geographically dispersed groups to systematically explore, evaluate, and come to decisions concerning complex problems. Moreover, by providing a logical-based debate representation, and by encouraging evidence-based reasoning and critical thinking, should significantly reduce the prevalence of some critical pitfalls that usually lead to deliberation failures in small scale groups and promote a more well-supported decision making. Argument mapping tools also face some important shortcomings. First, argument maps do not scale well: when the number of users increases the construction of proper collective maps appear to not be self-sustainable and selforganized and requires intensive moderation (Gürkan, Iandoli, Klein, & Zollo, 2010) Concerns have been raised about the effectiveness of argument mapping to mediate interaction: a central problem is the presence of communication formats too constraining and intrusive that disrupt the natural flow of free conversations. A further issue is the presence of a steep learning curve: a proficient use of argument mapping tools requires a certain amount of regular practice and training (Twardy, 2004). Scaling problems, ineffective mediation, need for practice and training imply more intense cognitive effort for users willing to participate to an argument-based conversation than it is required by current conversational technologies. Research on online argumentation has somehow underestimated these critical aspects and has focused mainly on knowledge representation issues with the objective to find suitable knowledge formats to represent users’ contributions while neglecting other important social and communication aspects involved in interaction among users. An unexplored advantage of online argument mapping, lying somewhat at the intersection between the use of knowledge representations and the analysis of social interaction happening through the representation itself, is the possibility to chart the social networks emerging among collaborators and relate it with the network of topics that is organized through the collective map in order to observe how the social and the content networks develop and possibly co-evolve. As with any other online collaborative tool, argument-based conversations can be analyzed to reconstruct the web of interactions among speakers, for instance by tracing a link each time a user replies to a post created by another member (Yan & Assimakopoulos, 2007). In addition, however, online maps offer the unique characteristics of make the relation between two concepts explicit by visualizing links describing certain functional roles (e.g. “supports”, “attacks”, etc.). Thus the social relationship and the contents network can be analyzed jointly to get insights on how the collective deliberation process develops. In order to investigate this aspect, in this paper we present an analysis of collectively generated argument maps with the aim to measure the structural properties of the social and concept networks generated during the debate. Our aim is twofold: first, we want to get empirical findings describing how the social network of collaboration evolves during the collective deliberation. Second, we argue that it could be desirable to provide users with additional information about the concept and social structure of the community that is involved in the discussion including measures such as who are the most central and active individuals or if there are subgroups clustering around specific positions or subtopics, who is talking to whom, who is talking about a certain topic, etc. Providing users of argument mapping tools with this kind of information could help to increase the social translucence of online users activities (Erickson & Kellogg, 2000): following previous studies (Danis, 2000; Shneiderman, 2000) we Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 20 International Journal of Decision Support System Technology, 3(2), 15-31, April-June 2011 argue that the visibility over the overall level of activity of the community members and a view of the system of relationships developing during the discussion contribute to increase the users’ knowledge and sense of the community, their degree of involvement and consequently the quality and the quantity of their participation level. In the next section we introduce Cohere, a web-based platform that has been used to support an argument-based discussion in two use cases. We analyzed the dataset collected in these two use cases through Social Network Metrics and Visualizations tools (UCINET and NodeXL). A web-bASeD ARguMent MAPPIng tooL: CoheRe Cohere is a web-based argument mapping tool whose purpose is to support an on-line collective argumentative debate. Viewed through the lens of contemporary web tools, Cohere sits at the intersection of web annotation (e.g. Diigo; Sidewiki), social bookmarking (e.g. Delicious), and mindmapping (e.g. MindMeister;), using data feeds and an API to expose content to other services (De Liddo & Buckingham Shum, 2010). This argument mapping tool uses the Issue Based Information System (IBIS) approach (Kunz & Rittel, 1970) in which a debate is represented as a tree structure composed of a set of issue to be answered, positions (or ideas) as alternative possible solutions to issues and supportive or challenging arguments about proposed ideas. With Cohere users can create posts to express their thoughts and pick up an icon to associate to them, which explain the rhetorical role of that post in the wider discussions. Moreover with Cohere users can explicitly connect their post to the post which is relevant to what they want to say. They can do so by making a connection between posts, which explain the rhetorical move they want to make in the conversation. Cohere augments the online conversation by making explicit information on the rhetorical function and relationship between posts (Figure 2). This web application allows representing debates in more compact way compared to traditional textual representation, creating semantic networks. By structuring and representing online discourse as semantic network of posts Cohere enables a whole new way to browse, make sense of, and analyze the online discourse. Indeed, through providing a logical rather than chronological organization of discussions, Cohere allows reflecting conceptual structure of debates and therefore to support the users in the answering to crucial questions such as: What are the key issues raised in the conversation? What are the emerging questions? How much support is there for this idea? Who disagrees, and what evidence do they use? What kind of argument is made to support this? Usually, this information is hidden in the free-text content, therefore participants have simply to read the entire online conversation along with all the possible “noise” to try picking up it. All this is supposed to improve the quality of collective decision making outcome, and more generally, knowledge representation and sharing as it should foster the emergence of more plausible, well-supported and shared conclusions about a given problem. In the next section we describe how Discourse and Social Network Analysis through Social Network Metrics and Visualizations could enable a deeper understanding of the online discourse, of participants to the debate and the social dynamics. DISCouRSe netwoRk AnALySeS AnD vISuALIzAtIon: how SuPPoRt onLIne DISCuSSIon gRouP Cohere is an Internet-based argument mapping tool, which allows to organize an argument as a map and to depict two superimposed networks that are assumed to be strongly connected: Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 15-31, April-June 2011 21 Figure 2. Cohere’s environment in which online dialogues is represented as semantic network of posts • • Concept network – which relates the nodes that users created. Social network – which relates users that participate to Cohere discussions posting Ideas, Questions and Arguments etc. The concept network gives a picture of the structure of discourse network. It allows to answer to some questions, such as “which is the most popular post?”, “Is the discourse network connected?”, “Is it possible to identify a pattern linking two posts?” For this network we consider the posts as nodes, and the semantic relations among posts as edges. The social network maps the pattern of interactions among actors in order to analyze users’ activity level and the most popular actor. In this network, we consider the users as nodes and we measure the edge between two users by counting the times that a user created a semantic connection that targeted a post authored by another user. The analysis of concept and social network gives insights on how the collaborative process is developing by tracking patterns among posts and among users. In the following tables, we show the Social Network Metrics that we calculated and their different meanings adapted to our context. In order to provide some concrete examples of how the above network metrics can be applied in online discussion groups we present two use cases. In the first use case (OLnet Team discussion) Cohere has been used by a group of researchers to annotate the documents of a project proposal, and to reflect on which areas of the proposal they were making a contribution. With Cohere’s Firefox sidebar users can annotate the documents and share their annotations in the group discussion environments. These annotations are initially presented as list of posts presented in reverse chronological order within the discussion group. After this initial phase of reading and annotating the document, participants were asked to have a group discussion on the main research questions addressed by the team, the main project achievements and how Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 22 International Journal of Decision Support System Technology, 3(2), 15-31, April-June 2011 they related to the project goal. In order to do this, they had to create new posts in which they described more general reflections on research questions, goals and activities and then they had to start creating semantic connections between the document annotations and the posts. This resulted into a discourse network in which document notes, open questions, ideas and other posts’ type are connected, and node’s icons and links express the rhetorical role and move played by each post into the online discourse. In the second use case (COP15 discussion) four researchers have used Cohere to collaborative annotate web news, documents, blog posts etc about the United Nation Climate Change Conference COP15. Results of the web annotations have then been used to inform an online dialogue on the main issues tackled during COP15, as reported by the press or as micro and macro blogged by participants to the conference. In order to have a specific focus for the discussion participants choose to discuss one of the public’s top questions that have been suggested on a Open University Platform (see page: http://www.open.ac.uk/platform/join-in/ your-votes/question-bypopular/Climate%20 Change); that is: • How do we know that climate change is real and we’re not just experiencing a weather cycle? Participants were asked to explore and annotate key Open Educational Resources (OER) and Social Media pages (such as Blogs, Wikis, Twitter streams, and web pages in general) with ideas to help answering the tackled question. Moreover, they were asked to make connections between their ideas and other participants’ ideas. In this process the main driving question and the identified relevant OERs have been used as evidences to base claims/ideas. This resulted in a web of ideas and annotated resources on the issues at stake, meaningfully connected into a discourse network. In the next paragraph we will describe how statistic on discourse network can provide insights on the collective decision making process of the group discussion and social interaction among group members. To analyze both the online group discussions and to compute some of above mentioned Network Analytics, we used UCINET tool (Borgatti et al., 2002); instead we used NodeXL tool (Smith et al., 2009) for both concept network and social network visualization. ConCePt netwoRk AnALySIS AnD vISuALIzAtIon In this section we introduce some empirical results deriving from the analysis and visualization of concept network by computing the metrics indicated in Table 1. In particular, with regard to concept network, we compute two main network measurements, namely link distribution and presence of components. While, concerning concept network visualization, we show as the visualization can facilitate the individuation of most attractive posts based on degree centrality measurement. Link Distribution: Are there hot Posts? The first analysis that we conducted on the datasets of the two use cases looks at links distribution to assess the presence of hub users and hub topics. The existence of hubs indicates the presence of hot topics/posts. Network’s hubs are nodes with the highest degree centrality. From the analysis of link distribution, emerges that both Olnet and Cop15 discussion groups are characterized by a power law distribution. The power law tail indicates that the probability of funding posts with a large number of links is rather significant; this means that the network connectivity is dominated by few highly connected posts. As illustrated in the two histograms below (Figure 3), in both the network it is possible to identify a hub with a highest degree followed by smaller ones. From the analysis emerges that the hub is a post labeled #COP15 and classified as “idea type”. The hub post has been connected to many other posts, which present annotations of various web resources. By analyzing the authors of posts, it Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 15-31, April-June 2011 23 Table 1. Network and visualization measurements for concept network Network measurement for Concept Network What do we measure? How is it possible to measure it? Are there hot posts? The link distribution It plots the distribution of node in function of the number of links. In this way, it is possible to visualize if all nodes have the same number of links or if there are few nodes, namely the hub, with higher number of links. The potential hub computed by link distribution are the hot posts Are there sub-topics? Component analysis The analysis of presence of components in a network allows assessing the degree to which a network is connected. A network which is fully connected has only one component The presence of components indicates the existence of different sub-topics discussed in a group What are the most attractive posts? Degree centrality It is defined as the number of links incident upon a node It plots the most interesting topics or the most debated. underlines that the user who created the post was in fact using the hub to cluster those resources under the # tag “COP15”. This highlights a use of Cohere in which the user, more than dialoguing is rather mapping out his notes on web resources and then sharing them with the group within the online discourse. A different case is the OLnet discussion group, which highlights a use of Cohere as tool for collective inquiry. In fact the discussion presents two hubs, both with degree equal to 8: • Which is its Concept Network meaning? “What motivates registered users to learn in socio-collaborative ways on OpenLearn?” • which was classifies by the author as “idea type” post; and “How can we build a robust evidence base to support and enhance the design, evaluation and use of OERs?” which has been classified as a “question type” post. The first thing that we can notice is that the two hubs are both posts which present an open question to the group. This seems to suggest that within all posts’ types, questions have a higher discourse power, in that they trigger users’ participation and interactions. Of course more systematic observations on wider and different online discussion groups are needed Figure 3. Link distribution histograms for COP15 and OLnet Team discussion groups Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 24 International Journal of Decision Support System Technology, 3(2), 15-31, April-June 2011 to appropriately test this hypothesis. Other considerations can be also done by looking at the hub posts’ type. The user who authored the first hub did not correctly classify the rhetorical role of his post within the wider conversation. The post clearly states a question but it has been classified as “idea”. This may be due to misunderstanding of the decision making task or to less confidence in the use of the technology; in any case this observation would alert a person in charge on the user performance. Summarizing, the analysis of results of link distribution puts in evidences different discourse elements which have been deepened to gain further insights on concept and social network, such as: 1. 2. The role of hub post: through a more deep analysis of hub posts, we can understand if it is a post that attracts much attention by other users or, as in COP15 case, it is a way to better explicit user’s opinions. These measures could be carried with the analysis of social network to better understand user’s activity and how users utilize COHERE. The nature of hub post: in this case, the analysis of hub post’s typology (e.g. idea, question, pro, con and so on) can give us an indirect measure of the quality of user’s activities. For instance, in OLNet case, the user stated in not correct way his post and this can reduce the quality of the collaborative decision making process. This impact could be more relevant for hub posts than other ones because they attract more attention in the discussion. Components Analysis: Are there Sub-topics? The second analysis which has been conducted on the two datasets consists in assessing the presence of components. A component is a connected subset that composes a disconnected network. Within networks’components there are no links/paths between the nodes belonged to different components. In network analysis, this measure assesses the degree to which a network is disconnected. For instance a network which is fully connected has only one component. Therefore, if we look at the concept network, this measure identifies the number of eventual isolated subsets of topics within the discourse network. From the analysis emerges that both networks present several components and this implies that the networks are weakly connected. In details, COP15 group presents 9 components but the bulk of nodes belong to two components. OLnet group presents 10 components but the bulk of nodes belong to one component. The presence of components in each group can be interpreted as the emergence of different sub-discussions independent among them. Analyzing the size of each component, (number of node in each component) emerges that not all the sub-discussions are developed by users in the same way. Bigger components can be interpreted as hot sub-topics which attracts a greater interest than others. We can also notice that the number of posts in the discussion group may have an influence on how hub topics distribute. For instance if we compare the two groups we can notice that in more developed discussion groups, such as COP15 group, two components absorb the bulk of nodes (161 out of 178). While in a group with less posts, as OLnet group, the bigger component absorb less than 45% of the total nodes. This could indicate that at the beginning of the discussion, users try to explore a wider deliberation space talking about different aspects of the same topic; then gradually, they start to focus on few sub-topics and to deepen them. This hypothesis would need to be proof/disproved by more in depth analysis, but consideration on the line of these give an example of how analyzing network metrics can inform the understanding of group dynamics. Summarizing, the analysis of presence of components in the concept network puts in evidences different discourse elements which have been deepened to gain further insights on concept and social network, such as: 1. the attractiveness of topics: the hot subtopic, to which the bulk of posts are re- Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 15-31, April-June 2011 25 2. ferred, emerges by analyzing the size of each component. This measure guides the users and newcomers in the exploration of online discussion and in the identification of the most debated sub-topics. The development level of discussion: by analyzing the size of the bigger component in each use case, it emerges that in the more developed discussion (COP 15), the bulk of nodes belongs to few components. It requires a deeper analysis to verify if this measure can indicate a development phase of the discussion. In fact, by comparing the two use cases we can hypothesizes that at the beginning of the discussion, users try to explore a wider deliberation space talking about different aspects of the same topic; then gradually, they start to focus on few sub-topics and to deepen them. Degree Centrality visualization: what Are the Most Attractive Posts? Specific network visualizations can be drawn to focus on the main network analysis metrics. The next figure (Figure 4) shows results of the concept network visualization for the OLnet Team discussion group, done with NodeXL (Smith et al., 2009). The concept network visualization is supposed to be very helpful for the users and in particular for newcomers, because they can gain a holistic view on all the debate and easily understand the dynamics on the basis of discussion without navigating the entire map. Therefore, this allows reducing cognitive effort to make sense of the whole discussion and foster a more effective users’ participation in it (Xiong & Donath, 1999). The next figure (Figure 4) shows results of the concept network visualization for the OLnet Team discussion group, done with NodeXL (Smith et al., 2009). Moreover, we have tried to convey further information on development of debate that, in future work, will be computed according to measurements of networks analysis. In particular in Figure 4: • • • Edge shape depends on link type (Positive: solid line; Neutral: dashed line, Negative: dotted lines). The final shape depends on the prevalence of one of two link type Edge width depends on the frequency of the relationship Vertex size depends on the degree centrality. Through this visualization, users can easily understand what are the “hot” posts, that is those Figure 4. Concept Network Visualization of OLnet Team discussion Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 26 International Journal of Decision Support System Technology, 3(2), 15-31, April-June 2011 posts are able to attract a greater attention or interest. Indeed, degree centrality depends on the number of links that each post has. Therefore, higher degree centrality, greater the link number of that post is. The network visualizations is a powerful tool since it facilitates the analysis of such measurements related to the node. Generally, by visualizing at glance all concept network, it is feasible to deduce how the discussion is developing (number of nodes and links), if the online debate is controversial (link shape), if there are “hot” posts (size of node). SoCIAL netwoRk AnALySIS AnD vISuALIzAtIon The phrase “social network” refers to the set of actors and the ties among them. The network analyst would seek to model these relationships to depict the structure of a group. One could then study the impact of this structure on behaviour of the group and/or the influence of this structure on individuals within the group (Wasserman & Faust, 1994). The ties among users are inferred by the created link among posts. If there is a link between a post A, created by user1, and a post B, created by user2, we infer that there is a tie between user1 and user2. In the following we present the Social Network Analysis (SNA) for the OLnet discussion group. The SNA measurements that we consider in our analysis are: out degree and in degree centrality. We adapted the meaning of these two measures to our case, indeed: • • Out degree measures the users’ activity level answering to the question “What are the most active users?” In degree is a sort of indirect measure of quality and relevance of a user’s posts, answering to the question “What are the most expert users?” Table 3 shows the results that emerge from the analysis of OLnet group social network. The user more active is User 1. Her outdegree is equal to 11. It means that she creates 11 links among different posts. While, User 6 is the user with the higher indegree value. Her indegree is equal to 11. It means that L6’s posts are considered more interesting and/or relevant by the group. By using the network metrics detailed in Table 2 the SN of the OLnet Team discussion group can be represented in Figure 5, NodeXL tool’s visualization (Smith et al., 2009). In particular in Figure 5: • • • • Link width indicates the frequency of relationship (reply). Edge shape indicates the link type (positive: solid line, neutral: dashed line, negative: dotted line). The final shape depends on the prevalence of one of two link type. Vertex size depends on the in-degree centrality of each users (bigger node have higher in-degree centrality). Vertex colour depends on the out-degree centrality (more black sphere have a higher in-degree value instead grey node have lower indegree value). Table 2. Network and visualization measurements for social networkpt Network Network and Visualization measurements for Social Network What do we measure? How is it possible to measure it? Which is its Social Network meaning? What are the most active users? Out-degree centrality It is the number of ties that the node directs to others. It measures/plots most active users computing the activity of a user in terms of how many posts s/he has linked What are the most expert users? In-degree centrality It is a count of the number of ties directed to the node It measures/plots Prestige and Expertise of a user computing how many links s/he receives from the others. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 15-31, April-June 2011 27 Table 3. Outdegree and indegree values for the Olnet team discussion (11 participants) Users Outdegree Users Indegree L1 11 L6 11 L2 7 L3 7 L3 6 L1 5 L4 4 L4 4 L5 4 L2 4 L6 3 L8 4 L7 3 L5 2 L8 2 L7 2 L9 2 L11 2 L10 1 L10 1 L11 0 L9 1 Figure 5. Social network representation of the OLnet Team discussion group By visualizing social network, users can easily know who is participating to the discussion and be awareness of the work of the others. Some studies (Danis, 2000; Erickson et al., 2002; Shneiderman, 2000) claim that, showing users’ presence in the community and their productivity, can stimulate users’ participation, support the development of a sense of membership, increasing users’ engagement level, make conversation smoother, more reflective and productive. Additionally, monitoring social interactions among users we can observe how the decision making process is developing, if and how much people exchange knowledge, who are the most active users, who are the “experts” on the topic and so on. As with concept network analysis and visualization, social network analysis and visualization is supposed to be very useful for newcomers. Indeed, they could immediately gain a view of the social dynamics and of users’ role in the discussion. This can make Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 28 International Journal of Decision Support System Technology, 3(2), 15-31, April-June 2011 participation more efficient, by allowing users to pitch in conversation easier and with a lower cognitive effort as they can gain an overall view both of past users’ interaction process and users’ activities. IMPLICAtIonS AnD next StePS In this article we try to tackle a complex, but important challenge: to create tools able to support distributed decision making and at the same time to facilitate sense-making and mediate interaction among participants in a satisfactory way. In this paper, through the presentation of some on-line debate, we try to investigate the capabilities of argument tools in structuring ideas and discourse and the abilities of Social Network Visualization in making visible the community and the social interactions among its members. These examples do not aim at presenting in depth analysis of the collected use case data, they are rather meant to give a proof of concept of the potential impact of Social Network Analysis and Visualization in the study of CSCW and in the design of tools to support distributed decision making. More specifically, using Cohere as an experimental platform, we have presented examples of Social Network Analysis and Visualization to better understand: • • Post and sub-topic distribution: by applying concept network analysis we can identify what are the hottest posts and sub-topics, what is the typology of posts and by who the posts have been proposed and discussed. Moreover we can see how topic and subtopic distribute within the online conversation. users’ participation level and participation modality: by calculating some social metrics, such as in-degree and out-degree centrality we can know the role of users in Cohere discussion and how much user participates in the debate. Moreover, in-degree centrality metric could be considered a reputational level that community recognizes to users. This could be considered as an urge to work more and better. Through these analyses, our intent is to develop in the near future tools for systematically monitoring the development of the collaborative process in order to have possible insights on how quality of the decision making process relates to the quality of the decision outcomes. In the next step, we should calculate more social networks metrics in order to deep the analysis of social and concept structures. Moreover, we should test if some hypothesis inferred by the analysis of concept measurement can be verified; such as if the analysis of component is an indicator of discussion development over the time and the analysis of post typology is an indicator of the user’s activity quality. RefeRenCeS Barabasi, A. L., Albert, R., & Jeong, H. (2000). Scale-free characteristics of random networks: The topology of the world-wide web. Physica A: Statistical Mechanism and its Applications, 281(1-4), 69-77. Beers, P. J., Boshuizen, H. P. A., Kirschner, P. A., & Gijselaers, W. H. (2006). Common ground, complex problems and decision making. Group Decision and Negotiation, 15, 529–556. doi:10.1007/s10726006-9030-1 Borgatti, S. P., Everett, M. G., & Freeman, L. C. (2002). Ucinet 6 for Windows: Software for social network analysis, analytic technologies. Boston, MA: Harvard University. Buckingham Shum, S. (2008). Cohere: Towards Web 2.0 argumentation. In Proceedings of the 2nd International Conference on Computational Models of Argument, Toulouse, France. Buckingham Shum, S., Selvin, A., Sierhuis, M., Conklin, J., Haley, C., & Nuseibeh, B. (2006). Hypermedia support for argumentation-based rationale: 15 Years on from gIBIS and QOC. In Dutoit, A., McCall, R., Mistrik, I., & Paech, B. (Eds.), Rationale management in software engineering: Concepts and techniques (pp. 111–132). Heidelberg, Germany: Springer-Verlag. doi:10.1007/978-3-540-30998-7_5 Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 15-31, April-June 2011 29 Conklin, J. (2006). Dialogue mapping: Building shared understanding of wicked problems. Chichester, UK: John Wiley & Sons. Conklin, J., & Begeman, M. L. (1988). gIBIS: A hypertext tool for exploratory policy discussion. ACM Transactions on Office Information Systems, 4(6), 303–331. doi:10.1145/58566.59297 Iandoli, L., Klein, M., & Zollo, G. (2009). Enabling on-line deliberation and collective decision-making through large-scale argumentation: A new approach to the design of an internet-based mass collaboration platform. International Journal of Decision Support System Technology, 1(1), 69–92. doi:10.4018/ jdsst.2009010105 Courtney, J. F. (2001). Decision making and knowledge management in inquiring organizations: Toward a new decision-making paradigm for DSS. Decision Support Systems, 31, 17–38. doi:10.1016/S01679236(00)00117-2 Jankovic, M., Zaraté, P., Bocquet, J.-C., & Le Cardinal, J. (2009). Collaborative decision making: Complementary developments of a model and an architecture as a tool support. International Journal of Decision Support System Technology, 1(1), 35–45. doi:10.4018/jdsst.2009010103 Cramton, C. D. (2001). The mutual knowledge problem and its consequences for dispersed collaboration. Organization Science, 12(3), 346–371. doi:10.1287/ orsc.12.3.346.10098 Kunz, W., & Rittel, H. W. (1970). Issues as elements of information systems. Berkeley, CA: Center for Planning and Development Research. Danis, C. M. (2000). Extending the concept of awareness to include static and dynamic person information. ACM SIGGROUP Bulletin, 21(3), 59–62. De Liddo, A., & Buckingham Shum, S. (2010). Capturing and representing deliberation in participatory planning practices. Paper presented at the International Conference on Online Deliberation, Leeds, UK. De Liddo, A., & Buckingham Shum, S. (2010). Cohere: A prototype for contested collective intelligence. In Proceedings of the ACM Computer Supported Cooperative Work Workshop on Collective Intelligence in Organizations - Toward a Research Agenda, Savannah, GA. Erickson, T., Halverson, C., Kellog, W. A., Laff, M., & Wolf, T. (2002). Social translucence: Designing social infrastructures that make collective activity visible. Communications of the ACM, 45(4), 40–44. doi:10.1145/505248.505270 Erickson, T., & Kellogg, W. A. (2000). Social translucence: An approach to designing systems that support social processes. ACM Transactions on Computer-Human Interaction, 7(1), 59–83. doi:10.1145/344949.345004 Gürkan, A., Iandoli, L., Klein, M., & Zollo, G. (2010). Mediating debate through on-line largescale argumentation: Evidence from the field. Information Sciences, 180, 3686–3702. doi:10.1016/j. ins.2010.06.011 McAfee, A. P. (2006). Enterprise 2.0: The dawn of emergent collaboration. MIT Sloan Management Review, 47(3), 21–28. Page, S. E. (2008). The difference: How the power of diversity creates better groups, firms, schools, and societies. Princeton, NJ: Princeton University Press. Shim, J. P., Warkentin, M., Courtney, J. F., Power, D. J., Sharda, R., & Carlsson, C. (2002). Past, present, and future of decision support technology. Decision Support Systems, 33(2), 111–126. doi:10.1016/ S0167-9236(01)00139-7 Shneiderman, B. (2000). Creating creativity: User interfaces for supporting innovation. ACM Transactions on Computer-Human Interaction, 7(1), 114–132. doi:10.1145/344949.345077 Smith, M., Shneiderman, B., Milic-Frayling, N., Rodrigues, E. M., Barash, V., Dunne, C., et al. (2009). Analyzing (social media) networks with NodeXL. In Proceedings of the Fourth International Conference on Communities and Technologies. Sunstein, C. R. (2006). Infotopia: How many minds produce knowledge. Oxford, NY: Oxford University Press. Tergan, S.-O. (2003). Managing knowledge with computer-based mapping tools. In Proceedings of the World Conference on Educational Multimedia, Hypermedia & Telecommunication (pp. 2514-2517). Toth, E. E., Suthers, D. D., & Lesgold, A. M. (2002). Mapping to know: The effects of representational guidance and reflective assessment on scientific inquiry. Science Education, 86(2), 264–286. doi:10.1002/sce.10004 Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 30 International Journal of Decision Support System Technology, 3(2), 15-31, April-June 2011 Twardy, C. (2004). Argument maps improve critical thinking. Teaching Philosophy, 2, 95–116. van Gelder, T. (2003). Enhancing deliberation through computer supported argument visualization. In P. A. Kirschner, S. J. Buckingham Shum, & C. S. Carr (Eds.) Visualizing argumentation: Software tools for collaborative and educational sense-making (pp. 97-115). London, UK: Springer. van Gelder, T. (2007). The rationale for Rationale. Law Probability and Risk, 6(1-4), 23–42. doi:10.1093/lpr/mgm032 Wasserman, S., & Faust, K. (1994). Social network analysis. Cambridge, UK: Cambridge University Press. Wolfers, J., & Zitzewitz, E. (2004). Prediction markets. The Journal of Economic Perspectives, 18(2), 107–126. doi:10.1257/0895330041371321 Xiong, R., & Donath, J. (1999). PeopleGarden: Creating data portraits for users. In Proceedings of the 12th Annual ACM Symposium on User Interface Software and Technology (pp. 37-44). Yan, J., & Assimakopoulos, D. (2007). The small world and scale-free structure of an internet technical community. In Proceedings of the Symposium on Computer Human Interaction for the Management of Information Technology (p. 10). Zaraté, P., & Soubie, J.-L. (2004). An overview of supports for collective decision making. Journal of Decision Systems, 13(2), 211–221. doi:10.3166/ jds.13.211-221 Simon Buckingham Shum is a Senior Lecturer at the Open University’s Knowledge Media Institute (KMI), a 70-strong research and development lab on Future Internet, Web Multimedia, Learning and Human-Centred Computing. His background is B.Sc. Psychology (1984, York), M.Sc. Ergonomics (1988, UC London) and PhD Human-Computer Interaction (1991, York). Dr Buckingham Shum leads KMI’s Hypermedia Discourse project, investigating the interplay between theory, technology and work practices for participatory, critical discourse. He is a co-founder of the Compendium Institute and GlobalSensemaking.net, and has edited two books by Springer: Visualizing Argumentation (2003) and Knowledge Cartography (2008). He cofounded and edited the Journal of Interactive Media in Education, and has contributed to the editorial boards of the International Journal of Human-Computer Studies and New Review of Hypermedia & Multimedia. Lorella Cannavacciuolo gained her Phd in Business and Managerial Engineering. She carries out her research activity within the Department of Business and Managerial Engineering at the University of Naples Federico II. Her research interests focus on the development of models to better manage organizational issues, such as design and implementation of Planning and Control Systems in health organizations and small firms. Currently, her research is also oriented to deepen the theory and models concerning the Social Network Analysis (SNA). She is applying theoretical concepts and models of SNA in different research fields, such as innovation of small firms, growth patterns in biotech sector, and online collaborative technologies. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 15-31, April-June 2011 31 Anna De Liddo is an Urban Planner and Designer; she took a MSc in Civil Engineering at Polytechnic of Bari, Italy and a MSc in Environmental Policy and Management at Institut National des Sience Appliquées in Lyon (France). She gained her PhD at Polytechnic of Bari, Italy, investigating ICT for Participatory Planning and Deliberation. In the course of the PhD she took individual responsibility in several international projects (Hypermediadiscourse, Palette and OpenParks) in the UK, Greece and Italy. She held a 12 month Post-Doc position at the Open University within the ESSENCE Project, investigating and evaluating human-centred computing tools to help tackling wicked problems such as Climate Change. She is currently Research Associate on Collective Intelligence at Knowledge Media Institute of the Open University, working in the Open Learning Network project (OLnet) on the design and development of a Collective Intelligence socio-technical infrastructure to enhance collaborative learning in Open Education. She is member of the GSm (Global Sense making) community and she developed links with research networks concerned with human-centred computing and the impact of computer technology on society, i.e. CPSR (Computers Professionals for Social Responsibility) and CIRN (Community Informatics Research Network). Luca Iandoli received his master’s degree in electronics engineering in 1998 from the University of Naples Federico II and a PhD in business and management from the University of Rome Tor Vergata in 2002. Since 2006, he is a professor in the Department of Business and Managerial Engineering, University of Naples Federico II. He got recently a Fulbright Scholarship in the category research scholar. His current research interests include application of soft computing techniques such as fuzzy logic and agent based systems to model organizational learning and cognition, such as in evaluation and decision making processes, and how to use collaborative internet technologies to support collective and organizational sense-making. He is a member of the editorial board of the Fuzzy Economic Review, Journal of Information Technology: Cases and Applications, and serves as associate editor for the Journal of Global Information Technology Management. Ivana Quinto received her master’s degree in SMEs Management in 2007 from the University of Naples Parthenope with full marks and honoris. She is a PhD student in Science and Technology Management (XXIV cycle) at the University of Naples Federico II, Department of Business and Managerial Engineering. Her research focuses on how web-based collaborative technologies are able to support online knowledge sharing and collective decision making processes. Recently, she was at Knowledge Media Institute - Open University (Milton Keynes – UK) as visiting student for six months. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 32 International Journal of Decision Support System Technology, 3(2), 32-43, April-June 2011 Strategic Development of a Decision Making Support System in a Public R&D Center Carlos E. Escobar-Toledo, Universidad Nacional Autónoma de México, Mexico Héctor A. Martínez-Berumen, Universidad Nacional Autónoma de México and CIATEQ, Mexico AbStRACt Decision making in new technologies is a crucial activity to raise competitiveness, especially for technology organizations. The decision-making process requires the use of information technology tools, since the information amount is large and requires reliable methods for collecting, accessing, storing, processing, distributing, and evaluating, in order to provide reliable information to decision makers. The strategy of an organization must take into account the integration of this aspect with other organizational functions. This paper presents a proposal to integrate new elements into the IT strategy, considering the interactions with other organizational functions, deining an implementation and transition plan that takes into account the organization dynamics, which has limited resources and, therefore, requires a gradual and long term transition plan. This paper refers as case study to a Mexican Public R&D Center, which has re-engineered its operating model with a systems approach. Keywords: Business Architecture, Decision Making, Decision Support Systems, Public R&D Center, Technology Planning IntRoDuCtIon The information system integration is an important topic in the strategic agenda of organizations. The term Enterprise Information System (EIS) refers to an information system that facilitates business processes and functionalities on an enterprise level (i.e., spanning across the enterprise) (Jukic, Jukic, & Velasco, 2009). In this regard being the key to ensure that computer systems are suitable and adapted to DOI: 10.4018/jdsst.2011040103 the organizational strategy, ensuring that when the latter changes, information systems enable and adapt, rather than becoming an obstacle for change (Cummins, 2002). One of the main challenges for the integration of computer systems is that, usually, organizations integrate isolated information systems (Sharum & Sage, 2002), since “organizations often make these decisions without formal analysis of existing systems and processes or without the clear understanding of the new system’s details” (Jukic, Jukic, & Velasco, 2009; OASIG, 1996). An approach which integrates Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 32-43, April-June 2011 33 organizational and technical change is perhaps missing in most organizations (Sedmak, 2010). One of the objectives of any organization is to adapt to its environment in an optimum sense, and to periodically reassess its strategy of adaptation as the environment changes (Kumar & Markeset, 2007). With this objective in mind, the organization needs to have methods and schemes in order to obtain, organize, process and analyze information about its operation and its environment (Maguire, Ojiako, & Robson, 2009) to gain a deeper understanding of their organizations and improve the decision-making process of users (Rockart, 1979). It is maybe surprising to note that “80% of the information needed to develop business intelligence already exists within organizations” (Rouach & Santi, 2001). However, Top managers of major corporations are frustrated by their information systems. They have difficulty getting information about how the business is running, as well as difficulty getting information to analyze the causes of major problems in order to develop solutions (Cummins, 2002). In this paper, we propose a theoretical alternative to strengthen the alignment between organizational strategy and Information Systems development, by defining a development strategy based on organizational architecture. Enterprise architectures, and frameworks which are offered as guidance for the construction of enterprise architectures, are not just theoretical constructs documented in the literature. They are in fact becoming commonly accepted in practice in both industrial and governmental institutions (Sharum & Sage, 2002). Enterprise architecting refers not just to the architecting performed at the Chief Information Officer’s level, but rather the larger enterprise context encompassing major program architectures, and down to the architecting of individual systems (Sharum & Sage, 2002). One of the main objectives to identify and define the organizational architecture as the basis for the development of information systems is to facilitate the interconnection of the various organizational elements. Communication and interconnection of organizational elements promotes synergy and collective intelligence. Each individual has some knowledge, both specific and global. This knowledge expands synergistically when shared in a collaborative environment (Baquero & Hernandez, 2008). It is also important to emphasize that collaboration and interdisciplinary are key to innovation, as they pose new methodological and conceptual situations, forcing a process of continual invention and experimentation (García, 2006). This paper also considers the case of a Mexican Public Research Center, which held a reengineering of its operating model with a systems approach (Martínez-Berumen, Baquero-Herrera, & Lizardi-Nieto, 2010). We propose such systems-oriented operation model as a basis for the strategic development of Information Systems. This paper deals specifically with the development of a Decision Making Support System (DMSS) to support decision making in new technologies. One aspect considered in the proposed methodology is that the reference organization has several information systems, many of which operate in a disjointed way (Baquero & Hernández, 2008). In addition, several inputs required for the methodology to support decision making on new technologies are obtained manually. Also, it is noted that the organization has limited resources for the acquisition or development of new information systems, so that the development and implementation of the DMSS should be considered as a gradual and long term process. We propose that an implementation plan should be designed considering these aspects. StRAtegIC DeveLoPMent of A DeCISIon SuPPoRt SySteM In A PubLIC R&D CenteR Value creation requires the design of a system of intangible resources, linked and articulated in a strategic manner and monitored continually (Baquero & Hernández, 2008). These resources result from the specialization of Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 34 International Journal of Decision Support System Technology, 3(2), 32-43, April-June 2011 different organizational functions. However, specialization may also result in decrement of interdisciplinarity and collaboration across the organization (Baquero & Hernández, 2008). The organization is then in a state in which knowledge workers are unable to access needed data and collaborate to develop enterprise solutions to key problems (Cummins, 2002). To support decision making, it is required to coordinate a strategy for knowledge assembly, which is given from rapid application of knowledge that already exists, but remains disconnected (Tiwana, 2002). Without ‘realworld’ intuition, the capabilities of computers are quite limited (Maguire, Ojiako, & Robson, 2009), making it clear that the role of information technology providing organized information cannot replace intuitive human analysis, which is based on prior knowledge assumptions, context, and experiences (Powell et al., 2004), as well as in tacit knowledge, understanding and learning (Baquero & Hernández, 2008), which are the most important resources for knowledge and organizational innovation (Teece, 1998), but they can support it, not only by organizing the data, but also to transform information into organizational knowledge. To this end, it is necessary to develop IT architecture consistent with organizational architecture, which allows integration of different information systems required by the organization to achieve its strategic objectives. In developing such an architecture, different areas of interdisciplinary collaboration may be identified (Saint-Onge, 2008), thus promoting information systems to developed across the entire organization (Baquero & Hernández, 2008) as an alternative way of designing “intelligent” systems, in which “autonomy, emergence, and distributed functioning replace control, preprogramming, and centralization” (Bonabeau, Dorigo, & Theraulaz, 1999) In the following sections we propose a methodology for the strategic development of a DMSS. This methodology is based on a systems analysis of the organization’s operational model, in order to ensure that information systems adapt to operation, and not the other way around. Methodology for the Strategic Development of the DMSS The field of information systems planning in the organization is a subject that is still in development (Galliers, 2004) and for which there are multiple perspectives and different approaches (Córdoba, 2009). It is important to consider that organizational information systems are embedded in an organizational strategy, and that development of these information systems, many of which are developed in order to support decision making, base on a decision made as part of the organizational strategy. Therefore, it is important to ensure that the decision to design and develop organizational information systems is taken in alignment with the overall organizational strategy. One of the most popular models of the decision-making process is that proposed by Simon (1960), which is consists of a three phases paradigm (intelligence, design and selection), complemented by an implementation phase. This paradigm contains most of the other proposed frameworks (Forgionne, 2000). An overview of each of the mentioned phases is described by Forgionne (2000). We propose the application of this paradigm at two levels, starting with the analysis of organizational functions with a systems approach, from which the organizational architecture and the main needs of organizational information systems are identified, in order to select and design the specific information system, defining an implementation phase, in line with organizational needs identified initially. Figure 1 summarizes the proposed methodology as an iterative process to strategically develop the DMSS in the Public Research Center. The following sections describe each of the stages mentioned in Figure 1, and briefly illustrate the experience of the mentioned Public Research Center for each of these stages: Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 32-43, April-June 2011 35 Figure 1. Methodology proposed to develop a strategic DMSS in the Public Research Center organizational Systems Analysis [Intelligence Phase] An organization’s success depends mainly on the alignment between its strategy and the various organizational elements. In order to ensure that such alignment is generated and maintained over time, a scheme to develop a systemsoriented Operation Model has been proposed, which is defined as a theoretical scheme that integrates all organizational elements for the achievement of common objectives and the strategic vision, by defining specialized systems within the institution, with the aim to create synergies and maximize organizational capabilities (Martínez-Berumen, 2007). This description is consistent with the definition of “System”, (INCOSE, 2010): “System: a combination of interacting elements organized to achieve one or more stated purposes an integrated set of elements, subsystems, or assemblies that accomplish a defined objective. These elements include products (hardware, software, and firmware), processes, people, information, techniques, facilities, services, and other support elements”. Figure 2 shows the general methodology for Organizational Systems Definition. The first step is to conduct a functional analysis with the aim of identifying the constituent systems, which in turn are integrated by processes. Systems interact with each other through the interactions of the processes within them (Martínez-Berumen et al., 2009). This activity is crucial, since the quality and further adaptation of the operation model depends on this step (Martínez-Berumen, Baquero-Herrera, & Lizardi-Nieto, 2010). Functional analysis depends on the Organizational Objectives and the diverse Management Systems that must be integrated, considering for example: Quality Management System, Health and Safety Management System, IT Security, Environmental Management System, Technology Management System, Project Management, Financial Accounting, Risk Management, among others. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 36 International Journal of Decision Support System Technology, 3(2), 32-43, April-June 2011 Failing in analyzing all this factors together, might lead to a weak integration and alignment of the operation model. Thus, it is possible to develop the business architecture, which is defined as “either the current or reengineered architecture of a business capturing its major components (and their responsibilities and relationships) as well as its major mechanisms” (Firesmith, 2005). business Architecture Construction [Design Phase] The organization of a Research and Development institution may be described as a complex system, given the large number of activities and interactions it maintains. The Public Research Centre referred to this work developed a new architecture for its operation model, in order to potentiate the organizational capabilities through a scheme that promotes systemic view of the organization (Martínez-Berumen, Baquero-Herrera, & Lizardi-Nieto, 2010). As a result of this development, it was concluded that the Systems oriented Operation Model consists of eight systems, which perform all the defined organizational functions. Figure 3 represents the main interactions between systems, after applying the described technique. One of the benefits when describing and mapping the organization as an integration of systems is that it is possible to define Goals and Indicators, aligned to the Main Strategic Objectives (Martínez-Berumen, Baquero-Herrera, & Lizardi-Nieto, 2010). Create the Integration Infrastructure Model [Design Phase] Once the business architecture is defined, an information systems architecture is required, which, in addition to support the strategy of the organization, provide a longitudinal and continuous monitoring of the operation (Baquero & Hernández, 2008). It is then necessary to design the infrastructure integration model, which is “the collection of shared facilities that allow the various business systems to work together” (Cummins, 2002). The importance of making this interactions map during the early phases of the project has been highlighted by Sage and Lynch (1998). In an organization, information flows are essential to ensure that all organizational elements operate together, so that “only a comprehensive consideration of all the elements of business and enterprise system architecture and their many complex interrelationships can reveal the true Figure 2. Systems definition Methodology (Martínez-Berumen et al., 2009) System A Organizational Objectives 1.2.3.4.n.- functions (necessary to achieve objectives ) 1.2.3.…n functions 2, 5, 6, … System B functions 1, 7, … System C functions 3, 4, … System D functions 8, 11, 9, … Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 32-43, April-June 2011 37 Figure 3. Main Organizational Systems Interactions. (Martínez-Berumen et al., 2010) Laws and Standards Government Performance Reports Directive System Associates Creation and development of Technology-based companies Associates Guidelines Strategy and Policies Knowledge Needs Technol ogy Environmen t Technology Foresight Technology System Tacit needs Market Market Market opportunities Explicit needs Customers Thematic Lines Technological Results Research Projects Innovation Projects Business System Proposal / Project Results Operations System Customers Specifications Proposals Project results Requirements Quality System Verification of requirements compliance Projects Budget Control Projects Budget Resources System Outreach Act ivities Research Act ivities HR Development Plan Knowledge Needs HR System magnitude of the strategic impact of information systems and resources on an organization and its strategic goals” (Jukic, Boris, Jukic, Nenad, Velasco, & Miguel, 2009). A Systems Architecture analysis is useful to identify and integrate all constituent elements from early design stages. It is common for many organizations that, after many years of developing computer applications in an evolving technological landscape, organizations have a significant IT infrastructure, which can lead to “fragmentation” of the operation, since they have been created gradually over time, and ad-hoc methods have been developed to interconnect them. The diversity of application architectures and technology, along with the local focus of application functionality, create major barriers to the capture, communication, and integration of management information necessary for the effective operation and improvement of the business (Cum- Operation and Maintenance of IT infrastructure Information System Knowledge/ Training Activities mins, 2002). The objective of starting with the definition of the systems architecture is that this architecting process is intended to ensure the interoperation of different workflow systems. Business processes often link the activities of different business units in an enterprise. If different units implement different workflow management systems, then they may be unable to link to each other’s business processes (Cummins, 2002), and to generate a balance between technical and business concerns, for building an enterprise architecture (Sharum,& Sage, 2002). Having a general enterprise level architecture serves as a map to develop each specific system, thus avoiding the development of isolated applications. Instead, “from conception through design, deployment and maintenance, enterprise systems are integrated, and interoperability across systems is achieved as required Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 38 International Journal of Decision Support System Technology, 3(2), 32-43, April-June 2011 by the mission of the enterprise” (Sharum & Sage, 2002). Identify and Prioritize the needs and Development Projects through MCDA [Choice Phase] From the union between business architecture and infrastructure integration model, it is possible to identify the information systems development needs in the organization. As mentioned, it is important to consider that the organization has limited resources for the development of the different organizational requirements, making it necessary to prioritize the development needs, in order to ensure that available resources are distributed among the priority initiatives. It is clear that each project has significant differences, making the evaluation and selection a complex process. In fact, these projects are multi-criteria in nature. We propose the use of Multicriteria Methods (Brans & Mareschal, 1984, 1992, 2002; Roy, 1985; Vincke, 1989) for the selection and prioritization of the information systems development initiatives. The selection of variables to consider will vary according to the priorities of each organization. Some variables that could be considered are for example: Required Investment, Strategic Impact of the information system to be developed, Internal Resources Availability, development complexity, development time, and intensity of use in the organization. Identify the Parameters and operation Scheme of the Selected Information System [Intelligence Phase] Once the specific project has been selected, the parameters and operation scheme of the information system are identified. In the case presented in this work, the project to develop is a Decision Support System for decision making in new technologies. As noted by Forgionne (2000): By coupling the intellectual resources of users with the capabilities of the computer, decision-making support systems (DMSS) are expected to improve the quality of decisions in this specific topic. The problem space within which this system is expected to function can be quite complex, involving a number of tangible and intangible variables from across a variety of organizational decision-making levels (Clark,& Jones, 2008; Singh et al., 2002; Watson et al., 2004). Complexity is determined by the size of the problem space (size and number of the related elements), its variability, its measurement difficulty, its velocity of change (Clark & Jones, 2008), as well as the number of interactions related to the operation of the information system and the effect of external agents to the variability of the system. Considering the results of organizational systems analysis, and the resulting Systems oriented Operation model (Figure 3), the methodology needed to support decision making in new technologies in a public research center was built. Figure 4 represents such methodology, which is used as reference to build the architecture of the information system to be developed. In this diagram, the mentioned Intelligence, Design, Choice and Implementation phases are indicated. The information system architecture is built with the mentioned elements. Figure 5 shows the DMSS for new technology decision architecture, built according to the model proposed by Forgionne (2000). This architecture contains all the elements described in Figure 4. The input data come from the organizational elements shown in Figure 3. Identify necessary, Available and Developing elements of the Architecture [Choice Phase] Once the DMSS architecture is defined, the next step is to identify what elements are available or under development. Each element’s functionality is reviewed and the best alternative for meeting each of them is chosen. For example, the Development and application of MCDA activity of the methodology shown in Figure 4 which in turn is shown in the DMSS architecture (Figure 5) as input, it requires the use of specialized software (eg Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 32-43, April-June 2011 39 Figure 4. Methodology to support decision making in new technologies for a Public R & D Center (Martínez-Berumen, Escobar-Toledo, 2011) Decision Lab) and a variety of Data Bases and Knowledge Bases, which concentrate information resulting from the earlier stages of the methodology to support decision making in new technologies (Figure 4). Similarly, the “Systems Dynamics Analysis” stage requires the use of software (eg iThink) and a variety of Data Bases and Knowledge Bases which will provide the information to perform system dynamics analysis. At this stage it is necessary to clearly identify the current “as-is” and desired (to-be) state. In the current state, there may be various information sources and activities that still require human analysis; the trend is to eliminate repetitive and tedious tasks, so that humans can concentrate on activities which require analysis, insight, context and tacit knowledge. As part of this activity, it is essential to identify the information necessary for the implementation of the methodology, some of which comes from outside the organization. The acquisition and integration of this information in Data Bases and Knowledge Bases is a sensitive activity, since these inputs are crucial to developing competitive strategies and prepare for new challenges. Develop Specific Implementation Plan [Implementation Phase]. The last stage of the proposed methodology is to determine the specific implementation plan. At this stage, it is possible to assign resources for the development of the elements required by the DMSS architecture, according to the needs identified in the previous stage. It is also time to identify the organizational change dynamics, define necessary training activities, implementation stages and to define the schemes Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 40 International Journal of Decision Support System Technology, 3(2), 32-43, April-June 2011 Figure 5. DMSS for new technology decision architecture (adapted from Forgionne, 2000) by which the effectiveness of implementation will be evaluated. At this stage it is vital to pay special attention to human factors issues such as education, motivation, attitude and commitment since these aspects determine the success or failure of an initiative like the one presented in this paper. The benefits of promoting networking and collaboration have been mentioned. In this regard, it is worth considering that in all systems under study “there is an interaction of highly complex components (i.e., intelligent human beings)” (Newman et al., 2006), either as elements of the system under study, or as “enforcers” of the study itself. A recent study (Sedmak, 2010) found that communication is a key factor for successful implementation, which reinforces the notion that success of such initiatives depends on the integration of human resources. To achieve this, the implementation of an initiative presented in this work requires the involvement and full commitment of top management. The implementation program should be developed in a realistic way, considering the magnitude of the changes. There must be a long-term work program, and ensure its continuous updating, identifying the issues which Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 32-43, April-June 2011 41 require reallocation of resources or modify the implementation strategy. The use of project management tools is critical to the success of change initiatives. Regarding the allocation of resources, and according to Clark and Jones (2008), “higher quality in these kind of initiatives results when small incremental investments over short periods are made rather than larger investment and longer periods between those investments”. Change is a necessary condition for the survival of organizations. An approach like the presented in this paper should position the enterprise to incorporate incremental changes in the business processes, the organization, the applications, and the supporting technology. The goal is not only to implement an improvement, but to ensure that the organization enters a cycle of continuous transformation. ConCLuSIon Change is the only constant in organizations. Change may stem from market dynamics, reorganization, globalization, or changes in the technological environment. The organizational systems architecture must be adaptable and flexible enough to permit and facilitate these changes. There must be a development plan that takes into account that these organizations are alive, so, a gradual and long term transition plan should be defined, taking into account the availability of resources and participation of the human factor. The effectiveness of organizational change depends largely on the level of knowledge that is available on the organization. Having a complete map of its functions and interactions is a useful tool to ensure that changes are aligned with organizational strategy. Systems approach is an efficient tool to ensure such an alignment. The success of an organization depends on the alignment between its strategy and the various organizational elements. In this paper we have presented a theoretical scheme to foster the alignment between organizational strategy and information systems. This proposal is based on the experience gained by a Mexican Public R&D Center, which is currently in an organizational transformation according to the topics discussed in this document. The effectiveness of the proposed approach should be assessed in the long term and will be reported in future works. RefeRenCeS Baquero, J., & Hernández, H. (2008). Articulación dinámica de la gestión de conocimiento con el proceso de innovación en la creación de valor: orientando la arquitectura informática. In Proceedings 3er. Simposio Iberoamericano en Comunicación del Conocimiento y Conferencias, en el contexto de la 7a. Conferencia Iberoamericana en Sistemas, Cibernética e Informática, Orlando, FL. Bonabeau, E., Dorigo, M., & Theraulaz, G. (1999). Swarm intelligence: From natural to artificial systems (1st ed.). New York, NY: Oxford University Press. Brans, J. P., & Mareschal, B. (1992). The PROMÉTHÉE – V MCDM problems with segmentation constraints. INFORMS, 30(2), 85–96. Brans, J. P., Mareschal, B., & Vincke, P. (1984). PROMETHEE: A new family of outranking methods in multicriterio analysis. Operations Research, 477–490. Brans. J. P., & Mareschal, B. (2002). PROMÉTHÉEGAIA Une méthodologie d’aide à la decision en presence de critères multiples. Brussels, Belgium: Editions de l’Université Libre de Bruxelles. Clark, T. D., & Jones, M. C. (2008). An experimental analysis of the dynamic structure and behavior of managerial support systems. System Dynamics Review, 24, 215–245. doi:10.1002/sdr.401 Córdoba, J. R. (2009). Critical reflection in planning information systems: A contribution from critical systems thinking. Information Systems Journal, 19, 123–147. doi:10.1111/j.1365-2575.2007.00284.x Cummins, F. A. (2002). Enterprise integration - an architecture for enterprise application and systems integration. New York, NY: John Wiley & Sons. Firesmith, D. (2005). The OPEN process framework repository organization glossary. Retrieved from http://www.opfro.org/index.html?Glossary/GlossaryA.html~Contents Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 42 International Journal of Decision Support System Technology, 3(2), 32-43, April-June 2011 Forgionne, G. A. (2000). Decision-making support systems effectiveness: The process to outcome link. Information Knowledge-Systems Management, 2, 169–188. Forgionne, G. A., & Rada, R. (2009). Effective DMSS guidance for financial investing. International Journal of Decision Support System Technology, 1(1). doi:10.4018/jdsst.2009010101 Galliers, R. (2004). Reflections on information systems strategizing. In Avgerou, C., Ciborra, C., & Land, F. (Eds.), The social study of information and communication technology: Innovation, actors and contexts (pp. 231–262). Oxford, UK: Oxford University Press. García, R. (2006). Sistemas complejos. Conceptos, método y fundamentación epistemológica de la investigación interdisciplinaria. Barcelona, Spain: Gedisa. International Council on Systems Engineering (INCOSE). (2010). Systems engineering handbook version 3.2. San Diego, CA: International Council on Systems Engineering. Jamshidi, M. (2009). System of systems engineering - innovations for the 21st century. New York, NY: John Wiley & Sons. Jukic, B., Jukic, N., & Velasco, M. (2009). Enterprise information systems change, adaptation and adoption. Information Resources Management Journal, 22(1), 63–87. doi:10.4018/irmj.2009010104 Kumar, R., & Markeset, T. (2007). Development of performance-based service strategies for the oil and gas industry: A case study. Journal of Business and Industrial Marketing, 22(4), 272–280. doi:10.1108/08858620710754531 Maguire, S., Ojiako, U., & Robson, I. (2009). The intelligence alchemy and the twenty-first century organization. Strategic Direction, 18, 125–139. Martínez-Berumen, H. A. (2007). Diseño e implantación de un Modelo Integral de Gestión orientado a Sistemas para Centros de Investigación y Desarrollo. Unpublished master’s thesis, Universidad Iberoamericana, Mexico City, México. Martínez-Berumen, H. A., Baquero-Herrera, F., & Lizardi-Nieto, V. (2010). Fostering innovation in an R&D center through systems engineering. In Proceedings of the 7th European Systems Engineering Conference, Stockholm, Sweden. Martínez-Berumen, H. A., & Escobar-Toledo, C. E. (2011). Decision making in new technologies for public research centers: A methodological proposal with systems approach. Mexico City, Mexico: UNAM. Martínez-Berumen, H. A., Escobar-Toledo, C. E., & Baquero-Herrera, F. (2009). Strengthening the technological aspects of an R&D centre using the systems approach. In Proceedings of the 7th Conference on Systems Engineering Research, Loughborough, UK. Newman, M., Barabasi, A. L., & Watts, D. J. (2006). The structure and dynamics of networks. Princeton, NJ: Princeton University Press. OASIG. (1996). The performance of information technology and the role of human and organisational factors. Swindon, UK: Economic and Social Research Council. Powell, G., Kokar, M., Matheus, C., & Lorenz, D. (2004). Understanding the role of context in the interpretation of complex battlespace intelligence. In Proceedings of the 9th International Conference on Information Fusion (pp. 1-8). Rockart, J. F. (1979). Chief executives define their own data needs. Harvard Business Review, 81–93. Rouach, D., & Santi, P. (2001). Competitive intelligence adds value: Five intelligence attitudes. European Management Journal, 19(5), 552–559. doi:10.1016/S0263-2373(01)00069-X Roy, B. (1985). Méthodologie multicritère d’aide à la décision. Paris, France: Economica. Sage, A. P., & Lynch, C. L. (1998). Systems integration and architecting: An overview of principles, practices, and perspectives. Systems Engineering, 1(3), 176–227. doi:10.1002/(SICI)15206858(1998)1:3<176::AID-SYS3>3.0.CO;2-L Saint-Onge, H. (2008). Collaborative networks and the new enterprise. Paper presented at the ADIAT Annual Congress, Acapulco, Mexico. Sedmak, M. (2010). Studying the processes during implementation of enterprise systems. Knowledge and Process Management, 17(1), 1–11. doi:10.1002/ kpm.336 Sharum, B. J., & Sage, A. P. (2002). Systems engineering and management of enterprise information systems: Ecological, architectural, and foundation program issues. Information Knowledge Systems Management, 3, 107–122. Simon, H. (1960). The new science of management decision. New York, NY: Harper and Row. Singh, S. K., Watson, H. J., & Watson, R. T. (2002). EIS support for the strategic management process. Decision Support Systems, 33, 71–85. doi:10.1016/ S0167-9236(01)00129-4 Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 32-43, April-June 2011 43 Teece, D. J. (1998). Capturing value from knowledge assets: The new economy, markets for know-how, and intangible assets. California Management Review, 40(3), 55–79. Tiwana, A. (2002). The knowledge management toolkit: Orchestrating IT, strategy, and knowledge platforms (2nd ed.). Upper Saddle River, NJ: Prentice Hall. Vincke, P. (1989). L’aide multicritère à la décision. Brussels, Belgium: Editions de l’Université Libre de Bruxelles. Watson, H. J., Fuller, C., & Ariyachandra, T. (2004). Data warehouse governance: Best practices at Blue Cross and Blue Shield of North Carolina. Decision Support Systems, 28, 435–450. doi:10.1016/j. dss.2003.06.001 Carlos E. Escobar-Toledo is a chemical engineer by the Faculty of Chemistry at UNAM; he received the degree of Master in Applied Sciences and PhD in Engineering (obtaining the specialty in Systems Theory) from the Universities of Louvain, Belgium and Aix-Marseille, France, respectively. His post-doctoral studies were conducted at the National Institute of Nuclear Sciences and Technology in Saclay, France. Dr. Escobar has extensive experience in the Energy and Petrochemical Industry. He worked at the Mexican Petroleum Company and in the Mexican Energy Ministry. He has been invited Professor at the Polytechnic Institute of Toulouse, France and at the Free University of Brussels, Belgium. He is currently a full-time professor at the Faculty of Chemistry at UNAM. His specialty areas are multicriteria decision making aid methods, mathematical modeling of complex systems and all intersections between the Systems Theory, Chemical Engineering and Operations Research. Dr. Escobar has published a lot of works in many international journals and assisted to international Conferences. He is President of the Specialty Commission on Systems Engineering at the Mexican Academy of Engineering and is a fellow of the Mexican Academy of Sciences. Héctor A. Martínez-Berumen is an Industrial Engineer by the Instituto Tecnológico de Aguascalientes (México), he received the degree of Master in Quality Engineering from the Universidad Iberoamericana (México). He has held various positions in CIATEQ, which is a Public Research Center in Mexico, where he currently works as Quality Manager. He is a member of INCOSE and ASQ. Currently, he is a doctoral student in Systems Engineering at UNAM. He has published some works in international journals and presented research papers in international conferences. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 44 International Journal of Decision Support System Technology, 3(2), 44-56, April-June 2011 Decision Support for Crisis Incidents Daniel J. Power, University of Northern Iowa, USA Roberta M. Roth, University of Northern Iowa, USA Rex Karsten, University of Northern Iowa, USA AbStRACt Crisis incidents occur in both business and public domains. This article focuses on non-routine incidents and explores uses of technologies for supporting crisis management tasks. A Crisis Incident Spiral of Decision Support helps identify useful decision support and information technologies. Additionally, a Crisis Incident Process/Decision Support Matrix categorizes processes of crisis planning, response and management with decision support technologies. Ideally, the matrix helps organize and stimulate thinking about novel DSS applications. Not all crises are of equal magnitude and different computerized decision support is needed in different types of crisis incidents. Keywords: Crisis Incidents, Crisis Preparedness, Crisis Prevention, Crisis Recovery, Decision Support, DSS, Incident Command System IntRoDuCtIon Both business and public domains continue to experience crisis incidents, and the magnitude of these incidents is sometimes much larger than any previously encountered. The recent Toyota vehicle recall, the Haiti earthquake, and the BP Gulf oil spill come immediately to mind. On a positive note, managers and politicians seem to desire more computerized decision support to help in both crisis incident planning and response. Checklists, vague contingency plans, and informal, ad hoc coordination are no longer considered adequate. Unfortunately, there are many unanswered questions about the “what?” and “how?” of DOI: 10.4018/jdsst.2011040104 disaster readiness and crisis and emergency decision support that need to be investigated and resolved. Recent events have brought new urgency to the topic of crisis and emergency decision support. Creating integrated decision support environments for command and control and emergency response is increasingly recognized as an important topic. The goal of this article is to discuss and explore some of the more creative and practical things we can use from the decision support systems (DSS) area to improve the capability for supporting all phases of crisis incident preparedness and management. The next section defines crisis incidents; then we explore a model for decision support and the possibilities for support in incident categories; finally, we conclude with relevant guidelines. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 44-56, April-June 2011 45 DefInIng CRISIS InCIDentS Not surprisingly, there are a variety of definitions and points of view on the term “crisis.” While these definitions vary in emphasis and detail, all agree that a crisis incident is an event or occurrence that creates difficulty or danger requiring a response and active management. For example, the World Health Organization (WHO) perspective emphasizes human health and safety concerns accompanying a crisis incident. According to the WHO website, crisis occurs when local and national systems are overwhelmed and are unable to meet basic needs. This may result from a sudden increase in demand (when food and water are in short supply), or because government and local services collapse because of staff shortage or lack of funds. Crisis triggers include sudden catastrophic events; complex, ongoing emergencies; or slow-onset events (World Health Organization, n.d.) The Munich Research Group identifies ten characteristics that are commonly associated with the term “crisis”: 1) an unusual volume and intensity of events, 2) ‘change of state’ in the flow of international political actions, 3) disruptive interactions between two or more adversaries, 4) abrupt or sudden change in one or more basic system variables, 5) change in the external or internal environment, 6) threat to basic values, 7) high probability of involvement in military hostilities, 8) awareness of finite time for response, 9) surprise, and 10) uncertainty (Power, 2005b). The Carnelian International Risks website focuses on business risk management. The site notes “The term Crisis is a complex proposition; since the word crisis will truly have different implications for different individuals and organisations. In reality the definition of crisis will vary depending on the constructs, limitations and perceptions of the crisis situation. What is important to recognise is that crisis events are not limited to, or defined by executive kidnappings, hijackings or product tampering, but by salient environmental variables that shape the situation into a crisis event for an organisation” (Power, 2005b). Carnelian consultants state the definition of crisis is dependent on perceptions of the value of possible losses, the probability of loss and the time pressure involved. Here, we see a business and private sector view of crises (Power, 2005b). Reh differentiates a crisis from a disaster. He argues a crisis and a disaster are very different. “A disaster is an event that results in great damage, difficulty, or death. A crisis is a situation that has reached an extremely difficult or dangerous point… Sometimes it is hard to know whether you are really in a crisis, but failure to handle a disaster properly can lead to a crisis” (Reh, 2010, p. 1). The Institute for Crisis Management notes “The most effective crisis management occurs when potential crises are detected and dealt with quickly--before they can impact the organization’s business. In those instances they never come to the attention of the organization’s key stakeholders or the general public via the news media” (Institute for Crisis Management, 2008). Tortella (2005) expanded on the business/ private sector crisis perspective by identifying eight characteristics of corporate crises. Characteristics included are surprise that is always driven by media exposure, and escalating flow of events that hampers the ability of management to understand quickly that, like it or not, they ‘own’ the problem and must quickly articulate a persuasive response. A number of authors have identified three stages of crisis management: 1) Prevention, 2) Preparedness, and 3) Recovery. They note that crisis prevention involves monitoring, anticipation, and taking pre-emptive actions to avert a crisis. Prevention is most problematic, difficult and expensive for low probability events. Crisis prevention activities can reduce threats. Crisis preparedness involves taking actions to reduce the impact and harm from a crisis when and if it should occur. It is important to identifying vulnerabilities and crisis scenarios. Planners need to identify what might go wrong and what the consequences would be if the worst case situation occurred. Crisis recovery encompasses damage assessment and the accounting, Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 46 International Journal of Decision Support System Technology, 3(2), 44-56, April-June 2011 reporting, and allocation of resources. Crisis recovery also offers the opportunity to learn how to respond more effectively to future crises (Grant & Powell, 2000). A MoDeL foR DeCISIon SuPPoRt In CRISIS InCIDentS Decision Support Systems can play an essential role in supporting the response and active management required when a crisis incident occurs. The model depicted in Figure 1 demonstrates our perspective on this role. Decision support systems can be used to help prevent some crisis incidents by providing a monitoring capability: gathering, collecting, organizing, and reporting on the status of incident indicators. Based on insights gathered in the monitoring phase, appropriate action steps can be taken to prepare for the crisis incident. Here, DSS can provide a variety of support mechanisms, including prediction models, communication and coordination, and planning models. Finally, the recovery from the crisis incident unfolds and again, DSS can play multiple roles in supporting recovery, including reporting, communication, and allocation models. Results of the recovery effort cycle back through the prevention phase, conveying feedback to the preparedness phase, which then may result in adjustments in the recovery phase. Hence, the “spiral” shape of the model. In the following two sections of the paper, we explore this model in terms of the Incident Command System Concept and through a discussion of six major categories of non-routine crisis incidents. InCIDent CoMMAnD SySteM ConCePt In the United States, the Incident Command System (ICS) is a broad approach for managing crisis/emergency situations (FEMA, n.d.; OSHA, n.d.). It is a system for managing emergencies. ICS is a “standardized on-scene incident management concept designed specifically to allow responders to adopt an integrated organizational structure equal to the complexity and demands of any single incident or multiple incidents without being hindered by jurisdictional boundaries” (Power, 2005b, p. 4) In the early 1970s, ICS was developed to manage rapidly moving wildfires. According to a number of sources, the system was intended to address the following eight problems: 1) too many people reporting to one supervisor; 2) different emergency response organizational structures; 3) lack of reliable incident information; 4) inadequate and incompatible communications; 5) lack of structure for coordinated planning among agencies; 6) unclear Figure 1. Crisis incident spiral of decision support Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 44-56, April-June 2011 47 lines of authority; 7) terminology differences among agencies; and 8) unclear or unspecified incident objectives. The key player in ICS is the Incident Commander on the scene, but a Command Center provides a reporting system to a unified command structure. ICS is a framework or template for creating and expanding a temporary organization for responding to an emergency or a crisis. ICS is oriented toward consolidating the efforts of public sector agencies, but conceivably it can mesh the efforts of public agencies, not-for-profits, private sector organizations and individual volunteers. Information technology must scale up and down as appropriate to an incident. DSS/IS/IT can serve particular responders on the scene of an incident, for example, supporting triage by a medical professional. DSS/IS/IT fulfills multiple roles in a permanent Command or Operations Center. Both commercial off-the-shelf software (COTS) and specialized, customized ICS software have a place in supporting the ICS. In general, software like Microsoft Access and Excel can be used to create templates that can be employed in a specific incident situation. As the scale of an incident increases, more specialized, web-based applications may be useful for distributed data gathering, data analysis and decision support in the temporary ICS organization. The web is an excellent means of gathering, maintaining and sharing data. Specialized software is needed for specific types of incidents like oil spills. Clearly, an Incident Commander and the crisis team need to be comfortable in a high technology “cocoon” of wireless interconnectivity, web access and stand-alone tools like MS Access and Excel. Moreover, an Incident Commander needs to be able to monitor and resolve hardware and software problems. Should any part of the technology infrastructure break down, an Incident Commander also must be able to improvise and continue functioning through the use of stand-alone computing or with no information technology support at all. Data must be gathered easily, inexpensively and reliably during an incident. Obviously, we need to gather data before we can use data-driven or model-driven DSS. Emergency response planners need to consider a variety of data-gathering methods. Computerized support can also assist in on-going emergency operations planning. An important job of an ICS is the initial development of an Incident Action Plan (IAP). If an incident continues for more than about 12-18 hours, a planning cycle is typically established by the Incident Commander and a Planning Section Chief is designated. A web-based Planning DSS can assist in development of an IAP for a particular operational period and help focus available resources on the highest priorities/ incident objectives. A web-based planning process can potentially speed up the planning process and better integrate staff inputs and identify critical shortfalls. In an extended crisis, technology-dependent planning and operation can become an issue. In a crisis, information technology can malfunction, break down, and create ancillary problems. These problems become more likely as the scale of the crisis grows in terms of number of people affected and the number of responders. Unfortunately, the current U.S. Incident Command System (ICS) does not adequately address how information technology will be supported, maintained and mobilized during an incident. More technology planning is needed for crisis/ emergency management as potential Incident Commanders will likely need an increasingly higher level of technology sophistication. The challenge is to develop a system that is complex enough to handle a wide variety of incidents while keeping it simple, robust, and less prone to the multifarious problems associated with emergencies. A RevIew of DSS uSe In CRISIS InCIDentS The expanded DSS framework (Power, 2002) identifies five categories of decision support systems: communication-driven, data-driven, document-driven, knowledge-driven, and Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 48 International Journal of Decision Support System Technology, 3(2), 44-56, April-June 2011 model-driven. Potentially, communicationsdriven DSS can reduce the negative effects of time pressure in a crisis situation. Data-driven DSS can help monitor the volume and intensity of events, abrupt or sudden change, insufficient information, and changes in the external or internal environment of an organization. Web-based, document-driven, group decision process structuring applications can improve contingency planning and action decisions during crisis preparation. Knowledge-driven DSS can potentially assist in understanding the constructs, limitations and perceptions of the crisis situation. For example, computerized DSS may help reduce cognitive biases during crisis management tasks. How values and preferences are elicited can impact their accuracy and how information is displayed in a DSS alters a decision maker’s perception in a situation. If we have built appropriate planning models, model-driven DSS can help reduce or manage uncertainty. Also, model-driven DSS can help identify vulnerabilities and evaluate crisis scenarios. In addition, Web portals and web-based DSS can help monitor news and events and help organizations share information with the media, stakeholders, and the general public. Improved communications technologies and handheld and portable computing technologies make it possible for first responders to bring decision support technologies into a crisis management setting whether that is in a nearby hotel room, a tent, or at the scene of an incident. A commander at the scene of an incident or crisis can conceivably have access to the entire range of DSS (FEMA, n.d.; OSHA, n.d.). In summary, a robust decision support infrastructure, coupled with appropriate training for response teams in the use of computerized decision support technologies, can help deal with the surprise, loss of control, panic, and psychological stress characteristic of crisis incidents. We will continue to encounter surprises that are both positive and negative. It is not possible and won’t be possible to anticipate every crisis incident that an organization, government, or public agency might encounter. That should not keep us from trying to anticipate crisis incidents. DSS can store scenarios, plans, and situation analyses and provide a starting point for more effective, rapid response. However, we must also ask the hard, troubling questions of crisis management. What if the computer systems fail? What if the data collected was inaccurate? What if the crisis is very serious and it was not anticipated? What if communications systems are not working? Are current plans for crisis management and response too dependent upon information technologies? PoSSIbILItIeS foR DSS In CRISIS InCIDentS Speculating about what might have been possible in specific exemplar situations can improve contingency planning and help us develop more sophisticated DSS. In much of the world, recurring emergencies of a small scale, like traffic accidents, are managed from centralized dispatch centers with computer-aided dispatch (CAD) tools. The first responders bring some decision support to the scene of an incident with them. There is a significant opportunity for expanding dispatch tools to include more decision support while also enhancing transaction processing role. More mobile decision support for triage and hazard management can also be developed. Improved data collection and sharing can also lead to more timely traffic safety and traffic management decision making at the management control level in local jurisdictions and enhanced monitoring and problem identification at more macro level government organizations. In this section of the paper, we direct our attention to six categories of non-routine crisis incidents: company and organizational crises, economic disruptions, natural catastrophes, political / terrorist acts, public infrastructure catastrophes, and public health crises. We provide several notable examples of each type of crisis, and then briefly summarize the support contributions that can be made by DSS in terms of preventing, preparing, and recovering from the crisis. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 44-56, April-June 2011 49 Company and organizational Crises Four examples of company and organizational crises are the Toyota Safety Recall, the Firestone Tire recall, Union Carbide gas leak in Bhopal, and the BP Oil Spill. In September 2009 Toyota issued a safety warning for 3.8 million Lexus and Toyota cars because of potentially deadly floor mats. Toyota recalled accessory all-weather floor mats in 2007 for similar problems. The US National Highway Traffic Safety Administration (NHTSA) noted reports of vehicles accelerating rapidly even after the release of the pedal (Yousef, 2009). The Firestone Tire recall associated with Ford Explorer crashes demonstrates a crisis that was mounting slowly for two large multinational companies. Data collected from traffic accidents was eventually used to demonstrate a cause and effect link that led the NHTSA to advise the companies involved to issue a recall of 6.5 million tires. Estimates of the impact of the faulty tires are approximately 250 deaths and more than 3000 catastrophic injuries. Most of the deaths occurred in accidents involving the Ford Explorer which tended to rollover when one of its tires had a blowout (CNN Money, 2000). In the early hours of December 3, 1984, methyl isocyanate (MIC) gas leaked from the Union Carbide India Limited (UCIL) plant in Bhopal, India. According to the state government, approximately 3,800 people died, approximately 40 people experienced permanent disability, and approximately 2,800 other individuals experienced partial disabilities (Singh, 2010). On April 20, 2010 an explosion and fire occurred on the BP-licensed Transocean drilling rig, Deepwater Horizon, in the Gulf of Mexico. . After the rig sank into the Gulf of Mexico on April 22, 2010, it dumped as much as 4.9 million barrels of oil into the sea before its broken well was capped in mid-July, 2010. Eleven people died and approximately 17 people were injured. Environmental and economic damage to the region has been extensive. Official estimates now say the Deepwater Horizon disaster is the largest accidental release of oil in world history (Yarett & Jones, 2010). Could DSSs have helped decision makers at companies facing disasters such as those listed here respond more effectively? For Prevention, DSS could potentially be used to identify the problem earlier and take action to avoid the problem. Business Intelligence systems would need to become much more sophisticated to help in this type of situation. Preparedness could have been improved through the use of planning DSS to develop contingency plans. Once the crisis occurred, Recovery could have been enhanced through communications-driven DSS, including simple bulletin boards, enabled improved coordination, gathered feedback and speeded decision-making, and helped the organizations control the communication flow and manage public perceptions. Major economic Disruptions Major economic disruptions are uncommon, but four notable occurrences are illustrative. On October 29, 1929 (known as Black Tuesday), the US stock market collapsed. In a single day, sixteen million shares were traded and thirty billion dollars of value vanished. For example, Westinghouse lost two thirds of its September 1929 value. DuPont dropped seventy points (Galbraith, 1954). The U.S. Savings and Loan Crisis created the greatest banking collapse since the Great Depression of the early 1930’s. By 1989, over half the Savings and Loan Associations had failed, along with the federal fund that was created to insure their deposits. Between 19861995, over 1,000 institutions with total assets of over $500 billion failed. By 1999, the Crisis cost $153 billion, with taxpayers footing the bill for $124 billion, and the S&L industry paying the rest (Amadeo, 2010). Japan suffered one of the biggest property market collapses in modern history. At the market’s peak in 1991, all the land in Japan, a country the size of California, was worth about $18 trillion, or almost four times the Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 50 International Journal of Decision Support System Technology, 3(2), 44-56, April-June 2011 value of all property in the United States at the time. Then came crashes in both stocks and property after the Japanese central bank moved too aggressively to raise interest rates. Both markets spiraled downward as investors sold stocks to cover losses in the land market, and vice versa, plunging prices into a 14-year trough (Fackler, 2005). A global financial crisis came to the forefront of the business world and world media in September 2008, with the failure and merging of a number of American financial companies. Consumer spending fell, and banks were much less likely to approve loans, and many countries wet into a recession. These events precipitated numerous problems in the economic and political world (Hinton, 2009). In the case of economic disruptions, there is a clear Prevention role for DSS, focused on risk assessment and prevention of disaster. Model-driven DSS play an important Preparedness role. Prediction models can provide the foundation for contingency plans. Rule-based DSS can provide rules to curb market trading. At present, Recovery is not well done and consists mainly of ad hoc responses. natural Catastrophes Three significant events provide good examples of natural catastrophes. On Tuesday, January 12, 2010, a 7.0 magnitude earthquake struck Haiti. The Haitian government estimates 200,000 people have died as a result of this incident. As of January 21, 2010, 2,000,000 people were homeless and 3,000,000 people in need of emergency aid (Haiti Earthquake Facts, n.d). The earthquake that generated the great Indian Ocean tsunami on December 26, 2004 is estimated to have released the energy of 23,000 Hiroshima-type atomic bombs. By the end of the day more than 150,000 people were dead or missing and millions more were homeless in 11 countries, making it perhaps the most destructive tsunami in history (National Geographic, 2005). Hurricane Katrina is the most costly natural catastrophe ever to strike the United States, and the deadliest since the Lake Okeechobee hurricane of September, 1928. Katrina caused widespread, massive devastation along the central Gulf Coast states of the U.S. The flooding of New Orleans, LA following the passage of Katrina was catastrophic, resulting in the displacement of more than 250,000 people. As of early August 2006, the death toll exceeded 1800 and total damages/costs were estimated to be around $125 billion (Graumann et al., 2005). Earthquakes, tsunamis, hurricanes, flooding, wildfires, mudslides, avalanches, and tornados cannot be avoided. Weather forecasting involves extensive computerized decision support. Prevention through better early warning and notification systems can be built to improve decision support for these situations (Kirlik, 2007). Since the impact of natural disasters can be very large, civil emergency and not-for-profit agencies need to improve Preparedness by investing in a wide range of DSS for a wide range of disasters. Recovery is enhanced through DSS that support Incident Management and First Responders and assist in the follow up of such disasters. Web portals can help gather relief items and notify the public about facts following a natural disaster. Communications-driven DSS can be created to inform, notify and consult with individuals, including potential victims. Political/terrorist Acts Terrorist attacks with political overtones are illustrated with four examples. On September 5, 1972 Black September terrorists launched a terrorist attack during the 1972 Olympic Games. Eight Palestinian terrorists killed two members of the Israeli Olympic team and then took nine others hostage. The situation was ended by a huge gunfight that left five of the terrorists and all of the nine hostages dead (Rosenberg, n.d.). On September 11, 2001, two planes struck the World Trade Center towers in New York City. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 44-56, April-June 2011 51 A third plan struck the Pentagon, and a fourth plane crashed in rural Pennsylvania. More than 2,600 people died at the World Trade Center; 125 died at the Pentagon; 256 died on the four planes. The death toll surpassed that at Pearl Harbor in December 1941. The planes also destroyed the New York Emergency Operations center. Command center location is obviously an important consideration. In February 2002, New York City opened a “new” temporary Emergency Operations Center (EOC). The new center replaced the EOC that former Mayor Giuliani had built at a cost of more than $13 million. It was located on the 23rd floor of 7 World Trade Center. That 40-story building collapsed about 7 hours after the Twin Towers. The EOC was criticized as lavish and ornate, but more importantly, it was poorly located (Global Security, n.d; Walton, 2003). On July 7, 2005, al-Qaeda associated militants detonated four bombs, three in London Underground trains in quick succession, a fourth bomb an hour later on a double-decker bus. Fifty-two people in addition to the four bombers were killed in the attacks and over 700 were injured (Muir & Cowan, 2005). On December 21, 1988, a terrorist bomb exploded on board Pan Am flight 103, destroying the aircraft over the Scottish town of Lockerbie and killing 270 people (Guardian News, 2009). Terrorism remains an on-going concern throughout the world. Information technologies and decision support systems have the potential to help officials Prevent, Prepare for and Recover effectively from terrorist attacks (AHRQ, n.d.) New and existing detection, diagnostic, management, and prevention DSS, along with surveillance, reporting and communication systems can help mitigate terrorist threats and improve response. Implementing structural solutions to reduce risks when possible are better than hoping that improved computerized decision support will identify and avoid terrorist threats. The case study at DSSResources. COM by Walton (2003) documents DSS used in response to the 9/11 crisis. Public Infrastructure Catastrophes We have created a complex public/private infrastructure that can fail and lead to “manmade” disasters. Crisis incidents produced by public infrastructure failures are illustrated with five examples. New York City experienced electrical blackouts in 1965, 1977 and 2003 (CNN U.S., 2003). The Aug. 14, 2003 blackout demonstrated that a failure in control and decision support systems can have wide-ranging consequences. Then U.S. President George W. Bush said the power outages across the Northeast and Midwest were a “wake-up call” to the antiquated state of the nation’s electrical grid. The St. Francis Dam Flood in California on March 12, 1928 killed 306 people. The failure of the Teton Dam in southeastern Idaho resulted in the loss of 11 lives and millions of dollars in property damage. In China in August 1975, the worst dam disaster occurred. The Chinese called it “Chu Jiaozi” (The river dragon has come!). Altogether, 62 dams broke in this incident. Downstream the dikes and flood diversion projects could not resist the flood of water from the initial dam collapse. The flood spread over more than a million hectares of farm land throughout 29 counties and municipalities. Eleven million people throughout the region were severely affected and more than 85 thousand died as a result of the dam failures (Democratic Underground, n.d.). According to Watkins (1975) “there was little or no time for warnings.” The Chernobyl accident killed more than 30 people immediately, and as a result of the high radiation levels in the surrounding 20-mile radius, 135,000 people had to be evacuated. (World Nuclear Association, 2011). Just after 6 p.m. on the evening of August 1, 2007, the 40-year old I-35 bridge in Minneapolis, Minnesota, collapsed into the river and its banks without warning, killing 13 and injuring 121 others (U. S. Fire Administration, 2007). Decision automation and DSS need to be built to help limit the consequences of infra- Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 52 International Journal of Decision Support System Technology, 3(2), 44-56, April-June 2011 structure failures. Prevention can be enhanced through the collection and analysis of inspection data. The first responders to such crises can be better Prepared through the continued use of computerized command centers and better incident management decision support to reduce the loss of life and property. Recovery efforts are strengthened through DSS that provide accounting, reporting, and resource allocation distribution models. Public health Crises Four recent public health crises are used to illustrate this type of crisis incident. South Africa is already home to 5.7 million HIV-positive people, more than any other nation, and can expect an additional five million to become infected during the next two decades even if the nation more than doubles its already considerable financing for treatment and prevention and gives prevention a higher priority (Bearak, 2010). Severe acute respiratory syndrome (SARS) is a serious form of pneumonia, caused by a virus isolated in 2003. Probable SARS cases with onset of illness from November, 2002 through July 2003 led to 8096 documented cases and 774 deaths (World Health Organization, n.d.). Ebola hemorrhagic fever (EHF) is one of the most deadly viral diseases, causing death in 50-90% of all clinically ill cases. Since it was first reported in 1976, along the Ebola River in Zaire (now the Democratic Republic of the Congo), there have been several outbreaks in various Central African countries. These outbreaks have been several years apart and occurred in different locations. After a sudden onslaught of deaths, the outbreaks end, leaving scientists baffled. Public health officials worry that a local outbreak could turn into an international epidemic if an infected person flies to another part of the world (Johnson, 2007). In 2009-2010, a new and very different flu virus (called 2009 H1N1) spread worldwide causing the first flu pandemic in more than 40 years. On April 26, 2009, The World Health Organization declared a deadly new strain of swine flu to be a “public health emergency of international concern,” as health officials identified possible new cases in two additional U.S. states and called the disease widespread. A WHO panel declared the developments thus far a public health emergency and urged governments around the world to intensify surveillance for unusual outbreaks of flu-like illness and severe pneumonia (McKay, Luhnow, & Goldstein, 2009). Public health crises have been a problem for humankind for thousands of years. Plagues and epidemics have ravaged nations and communities. Collecting data has helped the Prevention of disease and identification of the causes of such events. New approaches to data collection and analysis, such as Google’s Flu Trends web site, are encouraging. Better prediction models can assisted in Preparedness for these events. Computerized decision support has taken on an increasing role in this crisis management and response domain. DSS could not have helped avoid these crises, but the goal of new DSS must be to help decision-makers identify outbreaks more quickly. Recovery is improved through Web-based information portals, providing faster and more appropriate response can be; triage support; and the allocation and distribution of vaccines. ConCLuSIon The list of potential crisis exemplars goes on, such as computer failures, computer virus attacks, hazardous material spills, product tampering, and political crises like the overthrow of a government or the Cuban Missile Crisis. For now, however, what can we conclude? Only some emergencies and crises require or will benefit from elaborate computerized decision support. DSS are not especially relevant, helpful or useful in some crisis situations. We need a typology of crisis situations to analyze DSS needs and gaps for crisis planning, response and management. We need to critically examine who “owns” the crisis-related DSS capabilities and how such capabilities should be funded Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 44-56, April-June 2011 53 Table 1. Crisis incident process / decision support matrix Types of Decision Support Crisis Prevention Crisis Preparedness Crisis Recovery Monitoring Data-Driven DSS Anticipation Planning DSS Model-driven DSS Pre-Emptive Action Knowledge-driven DSS Reduce impact and harm Communication-driven DSS Document-driven DSS Identify vulnerabilities and crisis scenarios Model-driven DSS Knowledge-driven DSS Document-driven DSS Damage assessment and rebuilding Communication-driven DSS Knowledge-driven DSS Document-driven DSS Model-driven DSS Data-driven DSS and maintained. Also, we need to critically assess what DSS are needed by public sector first responders, by both private and not-forprofit sector organizations, and by national and international government agencies. It is important to identify some shared characteristics of “crisis” situations where it may be helpful to introduce additional decision support and information technologies. Also, conceivably some categorization of the various situations and of the process of crisis planning, response and management can help sort and organize our thinking about the various DSS possibilities. The matrix presented in Table 1 is a useful starting point for categorizing crisis management tasks and identifying appropriate decision support. Not all crises are of equal magnitude and different computerized decision support is needed in different types of crisis incidents. Grappling with the complexity of generalizing about decision support systems (DSS) for crisis, emergency, disaster and hazard situations is challenging. Our general guidance for building computerized decision support systems is anchored in Murphy’s Laws and Corollaries: “Anything that can go wrong will go wrong”; “Everything takes longer than you think”; and “Nothing is as easy as it looks”. The same wisdom holds true for crisis incidents. Nevertheless, we must persist in building and using decision support and information systems to help people in crisis planning, response and management. This review has led us to conclude the following: • • • • • • • An Incident Commander can and should have access to the entire range of DSS. DSS capabilities must scale up and down as appropriate to an incident. DSS should be targeted to serve specific responders on the scene of an incident. DSS can be used in a permanent Command or Operations Center for multiple tasks. An Incident Commander needs to be comfortable in a high technology “cocoon” of wireless interconnectivity, web access and stand-alone tools like MS Access and Excel. COTS like MS Access and Excel can be used to create effective small scale DSS for crisis decision support. As the scale of an incident increases, more specialized, web-based applications may be useful for distributed data gathering, data analysis and decision support in the temporary ICS organization. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 54 International Journal of Decision Support System Technology, 3(2), 44-56, April-June 2011 • • • • • • • • • • • • Communications-driven DSS can potentially reduce the negative effects of time pressure in a crisis situation. Data-driven DSS can help monitor the “volume and intensity of events”, “abrupt or sudden changes”, and changes in the “external or internal environment” during a crisis incident. Web-based, document-driven, group decision process structuring applications can improve contingency planning. Knowledge-driven DSS can potentially assist in understanding the “constructs, limitations and perceptions of the crisis situation”. Checklists can become more sophisticated. Model-driven DSS based upon quantitative planning models can help reduce or manage uncertainty. Also, model-driven DSS can help identify vulnerabilities and evaluate crisis scenarios. Both model-driven and data-driven DSS can support crisis prevention activities. Web portals and web-based DSS can help crisis decision makers monitor news and events and help organizations share information with the media, stakeholders and the general public. Improved communications technologies and handheld and portable computing technologies make it possible for first responders to bring decision support technologies into a crisis management setting. Communication and information technologies are likely to breakdown in crisis incidents, and backup and redundant systems should be in place. Assessing the appropriateness of specific DSS and decision support technologies in various crisis situations must be an ongoing activity of crisis management professionals and academic researchers. We can use the Internet and World-Wide Web to extend the reach and range of many general purpose DSS for crisis planning, response and management. Policy makers must examine who “owns” or should “own” crisis-related DSS capa- bilities and how such capabilities should be funded and maintained. RefeRenCeS Agency for Healthcare Research and Quality (AHRQ). (n. d.). Bioterrorism preparedness and response: Use of information technologies and decision support systems. Retrieved from http:// www.ahrq.gov/ Amadeo, K. (2010). Savings and loans crisis. Retrieved from http://useconomy.about.com/od/ grossdomesticproduct/p/89_Bank_Crisis.htm Bearak, B. (2010, November 19). South Africa fears millions more H.I.V. infection. Retrieved from http://www.nytimes.com/2010/11/20/world/ africa/20safrica.html CNN U.S. (2003). Major power outage hits New York, other large cities. Retrieved from http://articles.cnn. com/2003-08-14/us/power.outage_1_outage-powerplant-lightning-strike?_s=PM:US Democratic Underground. (n. d.). Chinese dam failure rate averages 68 per year. Retrieved from http:// www.democraticunderground.com/discuss/duboard. php?az=view_all&address=115x64070 Fackler, M. (2005). Take it from Japan: Bubbles hurt. Retrieved from http://www.nytimes. com/2005/12/25/business/yourmoney/25japan.html FEMA. (n. d.). Incident command system. Retrieved from http://www.fema.gov/emergency/nims/IncidentCommandSystem.shtm Galbraith, J. K. (1954). The great crash: 1929. Boston, MA: Houghton Mifflin. Global Security. (n. d.). World Trade Center - New York City 9-11 terrorist attacks. Retrieved from http:// www.globalsecurity.org/eye/wtc.htm Grant, S. E., & Powell, D. (2000). Crisis response and communication planning manual. Retrieved from http://www.foodsafetynetwork.ca/ Graumann, A., Houston, T., & Lawrimore, J. Levinson, D. Lott, N., McGown, S. et al. (2005). Hurricane Katrina: A climatological report (Tech. Rep. No. 2005-01). Washington, DC: NOAA’s National Climatic Data Center. Guardian News. (2009, August 21). Lockerbie bombing: The aftermath. Retrieved from http:// www.guardian.co.uk/uk/gallery/2008/dec/21/ lockerbie-terrorism Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 44-56, April-June 2011 55 Hinton, P. (2009). The start of the global financial crisis (2008). Retrieved from http://www.suite101. com/content/the-start-of-the-global-financial-crisis2008-a88065 Power, D. (2005b). Can DSS/IS/IT improve the incident command system? What needs can DSS meet? DSS News, 6(8). Hub Pages. (2010). Haiti earthquake facts. Retrieved from http://hubpages.com/hub/Haiti-EarthquakeFacts Reh, F. J. (2010). Survive the unthinkable through crisis planning. Retrieved from http://management. about.com/cs/communication/a/PlaceBlame1000. htm Institute for Crisis Management. (2008). The essence of crisis management. Retrieved from http://www. crisisexperts.com/essence_main.htm Rosenberg, J. (n. d.) Munich massacre. Retrieved from http://history1900s.about.com/od/ famouscrimesscandals/p/munichmassacre.htm Johnson, D. (2007). Ebola on the rise: Terrifying disease mystifies, concerns scientists. Retrieved from http://www.infoplease.com/spot/ebola1.html Singh, M. (2010). Bhopal and the BP oil spill: A tale of two disasters. Retrieved from http://www.time. com/time/world/article/0,8599,1995029,00.html McKay, B., Luhnow, D., & Goldstein, J. (2009, April 26). Swine flu is public health emergency, with new U.S. cases. Retrieved from http://online.wsj.com/ article/SB124069763075656299.html Tortorella, A. (2005). Expert view: What is a corporate crisis? Retrieved From http://www.ogilvypr. com/expert-views/corporate-crisis.cfm Money, C. N. N. (2000). Firestone tires recalled. Retrieved from http://money.cnn.com/2000/08/09/ news/firestone_recall/ Mt. St. Helens Information Resource Center. (n. d.). History-Mt. St. Helens. Retrieved from http://www. mountsthelens.com/history-1.html Muir, H., & Cowan, R. (2005). Four bombs in 50 minutes - Britain suffers its worst-ever terror attack. Retrieved from http://web.archive.org/ web/20071217222740/http://www.guardian.co.uk/ uk_news/story/0,1523819,00.html National Geographic. (2005). The deadliest tsunami in history? Retrieved from http://news.nationalgeographic.com/news/2004/12/1227_041226_tsunami. html OSHA. (n. d.). Incident command system. Retrieved from http://www.osha.gov/SLTC/etools/ics/index. html PBS. (n. d.). Stock market crash. Retrieved from http://www.pbs.org/fmc/timeline/estockmktcrash. htm Power, D. (2001). Can DSS and decision support technologies help reduce the threat of terrorism? DSS News, 2(20). Power, D. (2002). Decision support systems: Concepts and resources for managers. Westport, Ct: Quorum Books. Power, D. (2005a). How can DSS help in crisis planning, response and management? DSS News, 6(6). U. S. Fire Administration. (2007). I-35W bridge collapse and response (Tech. Rep. No. USFA-TR-166). Minneapolis, MN: FEMA. Walton, M. S., III. (2003). Rebuilding an emergency operations center for NYC following 9/11. Retrieved from http://dssresources.com/cases/eteam/index. html World Health Organization. (n. d.). The Summary (Indianapolis, Ind.). Retrieved from http://www. who.int/en/. World Nuclear Association. (2011). Chernobyl accident. Retrieved from http://www.world-nuclear. org/info/chernobyl/inf07.html Yahoo. News. (2007). The tri-state tornado of 1925. Retrieved from http://www.associatedcontent.com/ article/134954/the_tristate_tornado_of_1925.html Yarett, I., & Jones, K. (2010). Trouble on the horizon. Retrieved from http://www.newsweek.com/ photo/2010/05/22/oil-spill-timeline.html Yousef, H. (2009). Toyota: 3.8 million cars with risky floor mats. Retrieved from http://money.cnn. com/2009/09/29/news/companies/toyota_lexus_ floor_mats/ enDnote 1 An earlier version of this article was presented at the 2010 Annual Meeting of the Decision Sciences Institute and was published in the Proceedings; some material from this article first appeared in articles found at DSSResources.com. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 56 International Journal of Decision Support System Technology, 3(2), 44-56, April-June 2011 Daniel “Dan” J. Power is a Professor of Information Systems and Management at the College of Business Administration at the University of Northern Iowa, Cedar Falls, Iowa and the editor of DSSResources.COM, the Web-based knowledge repository about computerized systems that support decision making. He has authored three books on Decision Support Systems and his DSS Concepts book (2002) is a broad ranging handbook on the fundamentals of building decision support systems. In 1982, Professor Power received a Ph.D. in Business Administration from the University of Wisconsin-Madison. His Web home page is URL http://dssresources. com/vita/djphomepage.html. Roberta M. Roth is an Associate Professor of Management Information Systems at the College of Business Administration at the University of Northern Iowa. She received a Ph.D. in Management Information Systems from the University of Iowa. She currently teaches courses in business application development and systems analysis and design. She is co-author of a Systems Analysis and Design textbook and has authored numerous instructor manuals and teaching supplementary material. Her research has appeared in a number of publications and her research interests include IS pedagogy and distance education. Rex Karsten is an Associate Professor of Management Information Systems at the College of Business Administration at the University of Northern Iowa where he teaches information systems management and information systems development. He received his Ph.D. from the University of Nebraska-Lincoln. His research has appeared in a variety of publications including the Journal of Organizational and End User Computing, the Journal of Computer Information Systems, the Journal of Research in Educational Computing, Computers and Education, and Advances in Taxation. He currently serves on the Editorial Advisory Board for the Journal of Organizational and End User Computing. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 57 understanding organisational Decision Support Maturity: Case Studies of Irish organisations Mary Daly, University College Cork, Ireland Frederic Adam, University College Cork, Ireland AbStRACt Forty years after Gorry and Scott Morton’s seminal paper on DSS, supporting decisions in organisation is still a critical objective. Given the elapsed time since DSSs were irst introduced, it is important to gauge the scope and quality of decision support provided to managers. Using Executive MBA students as informants about decision making in their organisations, the authors carried out 10 case studies of Irish organisations to assess their maturity in terms of decision support usage. The indings indicate that, in the vast majority of irms, decision support is still not available to help manage in situations involving high levels of abstraction. As was the case at the beginning of the history of DSS, the operational level is still where DSSs are used most consistently across irms. Furthermore, this study illustrates that engaging with managers on the topic of decision making is dificult, given the possibility of bias and misrepresentation inherent in the reality of decision making. Keywords: Business Intelligence (BI), Case Studies, Cognitive Levels, Decision Makers, Decision Models, Decision Support Systems (DSS) IntRoDuCtIon Since Ackoff’s seminal and provocative paper (Ackoff 1967), researchers have sought to propose concepts, systems and methodologies to achieve the goal of providing managers with the information they need to make decisions. Throughout this time, it has remained true, however, that basic tools such as spreadsheets have formed the bulk of computer-based decision support (Fahy et al., 1996; Panko, 2006). Alter (2004) proposed that “decision support, provides a richer basis than DSS” for further DOI: 10.4018/jdsst.2011040105 research as well as for use in practice. The basis for his argument is that we must avoid the pitfalls that have at times plagued DSS research: techno-hype, domination of software vendors’ rhetoric and failure to understand the underlying problems which decision makers are facing (Arnott et al., 2008). Recently, new terms, such as Business Intelligence (BI), information cockpits or dashboards have been proposed (Dover, 2004; Gitlow, 2005) that leverage recent technologies – e.g., web technologies, multi-dimensional modelling tools – to deliver the silver bullet solutions to managerial decision making needs. However, it seems BI as a new tool is having a similar fate Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 58 International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 as previous installments of DSS technologies, with 40% of respondents to a recent study saying that the language used by vendors can often be ambiguous or confused, and a further 44% saying that vendors are creating an unhelpful mire of marketing speak around BI (Vile 2007). This is likely to be because, fundamentally, the problems raised by managerial decision making and the provision of information to support it – especially in situations involving high levels of uncertainty or equivocality (Earl et al., 1980) – are of an intractable nature. Decision making is inherently a human activity, as defining a human trait as language (Damasio, 1994). The role of the decision maker is to complete the model, as well as to control or to identify the gap in what has been programmed in the decision support systems (DSS) and the reality it is supposed to present (Levine et al., 1995). Situations involving high levels of uncertainty are those decision problems that have not been encountered in quite the same form and for which no predetermined and explicit set of ordered responses exists in the organisation (Mintzberg et al., 1976). The decision maker does not have a model, as they endeavour to understand the problem and provide an ordered response, long before a programmed system can be considered. In this paper, we use Humphreys’ framework of representation levels (Humphreys, 1989) to classify decision problems and Adam and Pomerol’s classification of decision support in terms of Reporting, Scrutinising and Discovering (Adam et al., 2008) to measure the extent of decision support provided by the portfolio of decision support tools in ten Irish firms. By tools we mean systems, routines, procedures and other forms of information dissemination (Simon, 1977). After eliciting the specific problems inherent in supporting managerial decision making and presenting the two frameworks used in our study, we describe the case studies on which our analysis is based. We then present our findings and conclusions with respect to the maturity of the decision problems encountered and the decision support capability of the firms we studied. 1. the PRobLeM wIth SuPPoRtIng MAnAgeRIAL DeCISIon MAkIng Information systems for top management raise specific problems which have primarily to do with the nature of managerial work itself (Dover, 2004; Fahy et al., 1996; Mintzberg, 1973), as they are intended to tackle the needs of users whose most important role is “to create a vision of the future of the company and to lead the company towards it” (King, 1985). However managers also spend considerable effort in their role of “go-between”, allocating work to subordinates and networking with internal and external peers (Kotter, 1982; Mintzberg, 1973). How computer systems can be used for these activities is largely unknown apart from the use of computer-mediated communication media – for instance email, which has a long history of successfully supporting managers (Crawford, 1982), sometimes with unintended consequences (Lee, 1994). Lest these observations be dismissed as outdated, they are in fact as accurate today as they were when they were printed. Evidently, information systems can help with decision making and information dissemination, but managers spend considerable time dealing with decision problems (Simon, 1977) and systems in this space still need improvement. Management decision making is based on much more than computer generated outputs and also rely on paper-based documents and back-of-the-envelope calculations. Despite the claims of software vendors, there is some evidence that the problems inherent in proposing effective decision support are of such a nature that modern GUIs, interfaces and the myriads of tool kits available from software vendors to develop advanced dashboards with minimal programming expertise are unlikely to solve them conclusively. It is the enlightened selection, and accurate capture of the organisation’s Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 59 currently available data sources, of the critical indicators most useful to the business managers that are problematic. Evidently, this requires collaboration between managers / users and IT specialists. This is an aged-old problem as far as Information Systems are concerned, which has been discussed in relation to Decision Support Systems, Executive Information Systems and generally any other type of systems that have been proposed for managerial support since the 1960’s (Ackoff, 1967; Keen et al., 1978; Rockart et al., 1988; Scott Morton, 1986; Watson et al., 1993), and more recently with the more sophisticated Business Intelligence offerings (Negash et al., 2008). 1.1. Classification of Inquiry type Different levels of management understanding of the problems they face must be treated differently when providing decision making support (Anthony, 1965). It has been proposed that managers can leverage the data provided by their support systems for three types of inquiry, (1) reporting, when managers ask questions that they understand well, and can monitor the answers to these questions over time with the use of tight aggregative models where previous decisions and ways to resolve them are embodied, (2) scrutinising, where managers ask questions which they understand in broad terms, but still find it difficult to ask precisely on the basis of incomplete models and (3) discovering, where managers are not even sure which questions they should be asking, sometimes in the complete absence of a model or even a specific problem to solve (Adam et al., 2008). These three decision support activities are practical from a developer’s viewpoint because they correspond to the level of knowledge that an analyst can gain a priori about an information system requirement they are about to tackle. 1.2. Classification of Decision Problems The three types of support can also be matched against the level of understanding which managers have of the decision problems they face. Humphreys et al. (1985) have usefully characterised this level of comprehension with their concept of representation levels. The five representation levels theorise on the evolution of managers’ thinking as they learn about the reality that surrounds them, based on: (1) the degree of abstraction of the representation they have of the problems to be tackled and (2) the degree of formalisation of the representations of the proposed solutions. The five representation levels can be illustrated with Humphreys and Berkeley’s description of the problem handling process, which is adapted in Table 1 (Humphreys et al., 1985). The process described by Humphreys is a top down process whereby the structuration of the concepts investigated is refined from one level to the next over time. Problem solving is viewed as a development process passing through the five representation levels, from more to less abstract. As noted by Lévine and Pomerol (1995), levels 5 and 4 are generally considered as strategic levels of reflection handled by top executives (problem defining), whereas the remaining three levels correspond to more operational and tactical levels (problem solving). Although, all levels of management span the 5 cognitive levels, it is clear that lower levels of management are more likely to be given problems already well formulated to work on, such that their thinking is mostly geared towards levels 1 and 2 (Levine et al., 1995). Level 5 in Table 1 is particularly important in that, at this early stage, the decision maker has total freedom to decide on a direction to follow. In the literature on human decision making, this initial step appears under the name “setting the agenda” (Simon, 1997) or “problem setting” (Checkland, 1981). This stage is also important because it conditions the outcome of the decision making process as avenues not considered at this stage are less likely to ever be considered. In addition, the natural progression across the levels of the framework is one that goes from 5 to 1, and rarely back to a previous stage unless a strong stimulus forces a change of mind about the situation. This representation of managers’ handling of deci- Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 60 International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 Table 1. Representation levels in managerial decision making (after Humphreys & Berkeley, 1985) Cognitive Level Representations of Managerial Thinking 5 Representations are mainly cultural and psychological; managers are more or less aware of what problems may involve, but their expression is beyond language. Problems are shaped at this level but are beyond modelling. 4 Representations become explicit and problems can be broken into sub-problems, some of them formalised. The structuration of problems is still partial and managers refer to ‘the marketing function’ or ‘the marketing process’. Data mining may be used to formalise ideas and test hypotheses but it is still hard for managers to discuss these. 3 Decision makers are able to clearly define the structure of the problems to be solved and develop scenarios for possible alternatives. They are able to put forward models for investigating alternatives solutions and to discuss these with analysts. 2 Decision makers perform sensitivity analysis using the defined models so as to determine suitable input values; saved searches and views created using scrutinising tools can become increasingly formalised and move from level 3 to level 2. 1 Managers decide upon the most suitable values and the representation of the problems is stable and fully operational. Report can be created and are regularly available. sion problems and information needs is a simplification in that it separates what is essentially a continuous cognitive process into discrete processes. However, from the point of view of the designer of management decision support, this framework has the merit of clarifying what design avenues can be pursued to support managers in situations that are more akin to stage 1, stage 5, or any other stage. 2. MeASuRIng the extent of DeCISIon SuPPoRt PRovIDeD Adam and Pomerol (2008) argue that, if managers can name specific performance indicators and know how these must be represented, the situation corresponds to the lowest representation level (level 1) in the Humphreys and Berkeley framework (especially if they are also able to calibrate performance level based on their own knowledge). This is essentially a reporting scenario where specific answers are given to specific questions. When, however, it is not exactly known how to measure or represent an indicator, this corresponds to levels 2 and 3 in the framework. This is more of a scrutinising situation where managers know they are Abstraction Level Maximum Minimum on to something, but they are not sure how to formally monitor it. Finally, when managers are not sure what indicator should be monitored to measure emergent changes in the activities of their organisations, or changes to market responses, this is more akin to a level 4 situation, or a level 5 situation if managers are still at the problem finding stage (Pounds, 1969). The development of the decision support capability of the firm thus becomes an iterative process where problems and their representations improve over time and where discovery turns into scrutiny and scrutiny turns into reporting. This theoretical proposition requires that the decision support capability of a firm is articulated around a complete portfolio of applications covering at least levels 1, 2 and 3, if not all levels. The completeness of the portfolio gives an idea of the level of maturity of a firm in terms of decision support. The notion of IT Maturity is not new and has been approached in a variety of ways by researchers (Earl, 1989; Galliers, 1993; Khandelwal & Ferguson, 1999) since the days of Nolan (1973, 1979), and his further work with other colleagues (Nolan et al., 1995). DSS maturity is not new either. Huerta Arribasa and Sánchez Inchustab (1999) used DSS maturity Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 61 – “IT to aid decision making” - as one of their factors in measuring IT maturity – explained as “the degree to which companies incorporate IT to pursue organisational aims” (p. 153). Adam et al. (1998) discussed DSS maturity in a sample of 18 organisations in terms of a DSS spread score, measuring the proportion of problems which are considered with the help of DSS applications in the firm, and a DSS complexity score, measuring the complexity of the problems on which DSS applications were applied in the organisation. The authors concluded that their framework was useful in categorising organisations with respect to DSS maturity, in a way that took into account the specificity of each firm. As observed in Nutt (1984) however, there is a trade-off between the level of detail which one can take into account about each of the cases analysed and the extent to which comparisons can be made across large samples. The complexity of the DSS maturity framework as developed in Adam et al. (1998) raises the risk that its application to large sample sizes could be problematic. To avoid such issue, we propose in this paper to measure DSS maturity in terms of the Humphreys and Berkeley (1985) framework. Specifically, we posit that the height of the footprint of DSS applications mapped against the portfolio of problems which an organisation faces across the categories of the framework can be used to read the relative level of DSS maturity of an organisation. If this footprint does not rise above level 3, then an organisation can be considered to be leveraging the concept of DSS. However, if the footprint rises to level 4 and even level 5 in tangible ways, an organisation can be termed to have reached some degree of DSS maturity. In order to operationalise our research objective and determine the decision problems encountered in organisations and the level of decision support provided we sought to answer the following questions: • What decision problems are on managers’ agenda and at what representation levels are they located? • • What is the scope and quality of decision support, as against decision support systems, provided in the firm, across representation levels? What conclusions can we draw on the maturity of the decision support provided in the firm? Meeting our research objective also requires collecting data about a substantial number of decision problems, observed across a broad sample of organisations. This means we needed to decided to elicit information from a large number of managers. The protocol we followed in seeking to achieve this, is the subject of the next section. 3. the ReSeARCh APPRoACh When considering the research questions as outlined above, and since they apply to an organizational level, we sought to carry out our research over a broad sample of firms. To achieve this, we enlisted the help of experienced managers in employment who were students in the Executive MBA program (EMBA) at University College Cork, and who were able to provide high levels of insight into a large sample of firms’ decision making and decision support capabilities. This follows a well establish tradition in business literature to use practitioners engaged in educational programmes for research purposes (Edmondson et al., 1988; Remus, 1986; Weick, 1967; Campbell & Stanley, 1963). The study allowed the researchers to collect information on manager’s decision activity in their organisations. This formed part of their marking for the MBA course and provided suitable motivation in contributing to the research. In preparation for their field work, all students were coached by the researchers in the application of the frameworks in Figure 1. The study took place across two successive EMBA cycles, which afforded the possibility of fine tuning the research protocol, as discussed below. The students were in their third semester of a four semester program, and had covered Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 62 International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 “fundamentals” of information systems in a previous semester. Groups were formed and each group selected two target organisations where at least one of the team members were employees and were engaged in decision making at management level. The other team members in the group provided critical validation for the decision level classification. The groups then presented their analysis to the researchers in extensive presentations and a detailed written report. These reports and presentations were used as research instruments for data collection and led to the analysis of the portfolio of decision levels and decision support in each of the case study organisation. The assignment question set for the groups was as follows: “Identify decisions made in your organisation, and identify the DSS which facilitate decision making for these decisions”. Implicit in the question was to also identify the gaps in decision support”. Based on the analysis of the portfolio of decision support tools as presented in phase one of the research, the question posed for the second group was amended to place a greater emphasis on decision support and all sources of information, as distinct from technology based information systems only, as follows: “Identify the decisions made in your organization. Identify the decision support which facilitates the decision making for these decisions. Consider all sources of information taken into account in the decision making process”. After the presentations, the researchers selected the most rigorously produced student reports and focused their analysis on the 10 case studies presented thereafter. 4. PReSentAtIon of CASeS AnD DISCuSSIon of fInDIngS Table 2 shows the key demographical data for the 10 companies in our sample. It indicates the spread of our observations across a range of industries and a range of sizes from medium Figure 1. Matching decision support tool contents to managerial needs (after Adam & Pomerol, 2008) Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 63 to very large. Our sample covers indigenous Irish firms and multinational companies, where the Irish subsidiaries were studied. The 10 companies also feature different domains of expertise from engineering to health. This reflects our attempts to cover many different types of organisational settings and to present a broad spectrum of observations. Table 2 also provides a brief account of the context of the firms and the challenges faced by their managers. This shows the general context of managerial decision making in these firms and allows their classification in terms of the dynamism of the environment in which they operated and the pace of change which they faced. It is useful as a backdrop against which to evaluate the extent to which DSS applications are being used to support managers in these firms in the crucial aspects of their jobs. 4.1. Presentation of the Data Table 3 shows the detail of the data we have assembled about the 10 firms. For each firm, decision problems and formal or informal decision support available to managers at each level of the framework have been recorded and classified based on the degree of abstraction of the managers’ representation of the decision problems presented, and the level of understanding of the decision problem solution. Prima facia, Table 3 reveals that managers in some firms do not tackle problems at the higher levels of the framework. When no decision problems are identified at a certain level, this level is omitted in the table. Thus, no firm has cells corresponding to level 5, 5 firms have no cells corresponding to level 4 and 1 firm only has cells at level 1. Table 3 also shows that decision problems classified at level one of the framework –those characterised by little ambiguity and low levels of abstraction – were well covered by information systems and DSS were used extensively for operational control and performance monitoring across all organisations. In other words, the decisions identified at level 1 of the framework were supported by well developed reporting tools based on ERPtype systems of data recording, augmented by industry-standard report generators / BI tool. Level 2 decision problems were also well covered, with “what-if” and “drill-down” type support widely used across most of the cases. The sophistication of such tools varied across the cases, with MS Excel being most favoured, and a number of organisations have well established BI type tools implemented. Conversely, very few decision problems were identified with complex and semi-formed ideas, where outcomes were unclear and ambiguous, even in organisations that are operating in very challenging, highly competitive and uncertain environments. The problems presented as “abstract” were in fact clearly stated. In a number of situations potential solutions were more uncertain, but possible scenarios were entertained. Finally, where no formal decision support was available to managers to support them in the search for solutions at a certain level, cells coded in italics. These preliminary observations confirm the results of earlier studies that a basic level of DSS proficiency, as against maturity, has been achieved by most firms at this point in time. However, as outlined in the earlier sections of this paper, we perceive that it is in the upper categories of the framework that DSS maturity must be measured. Therefore a more detailed analysis of the data is required to explicate the contribution of this study to our understanding of DSS usage in organisation and derive specific results on the measurement of DSS maturity in the firms in the sample. 4.2. Analysing the Case Data 4.2.1. Company A Company A is a private healthcare provider, with operations in five locations in Ireland. While individual patient admissions can be in the region of 60,000 per year, the primary source of revenue for the group is the private health insurers. Traditionally IT has been deployed Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 64 International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 Table 2. Demographics of the 10 companies in the sample Firm Activity Turnover Ownership Main Business Factors A Private Healthcare. €144 m. Private ndependent Primary source of revenue comes from private health insurers. 60,000 patient admissions per year. Changes to funding model for private healthcare in Ireland. B Energy supply. €1.1 bn. State body 2 main businesses – Gas Transportation and Energy Supply. Some deregulation of the energy supply market. However operating in a regulated market, with government approval required for price charged in all customer categories. C Milk Products €200 m. Irish cooperative Cheese, food ingredients and flavours manufacturer. Produces 25% of all cheese manufactured in Ireland. Quality of raw materials and securing reliable suppliers are key issues. D Medical Device manufacture $4 bn. worldwide. Private US multinational 7 manufacturing sites worldwide, with Cork plant accounting for 40% of total production. New plant in China will be a source of increased competition for product allocation. Extremely price sensitive market. Key Performance Indicators (KPI) oriented culture, but goals are handed down from headquarters to local site for each functional area and converted into strict targets for managers. E Hi-Tech manufacture €6 bn. worldwide. Private US multinational World leader in information management and data storage products, services and solutions. Has had to enlarge its product portfolio towards cheaper lower end products and also include software and consultancy products quite different from its traditional hardware products F Medical Device Manufacture €144 m. Private US multinational Large portfolio of innovative products, technologies and services that advance the practice of less-invasive medicine in a wide range of medical areas. Faces key changes in how healthcare is provided and funded in its core markets in the future. G Bioscience energy generation €10 m. Irish private company Electrical power generation from sustainable fuel sources or “green energy“. Start-up company, in a very immature industry segment, with few customers and suppliers. H Spirit distiller €7 bn worldwide Irish co-operative Part of largest wine and spirit company in the world. Extensive use of external market research data inc. data on the key drinks companies, their brands, sales volumes etc. I Food and Beverage $ 60 bn worldwide Private US multinational The primary focus of the organisation had become the creation of healthier products and reducing the organisation’s negative impact on the environment. Specific goals are handed down from headquarters to local sites for each functional area. J Supply Chain management €120 m. Private Irish international Irish supply chain management company with a product portfolio across consumer electronics, personal computers, medical devices and telecommunications. Rolled out KPIs across the organisation. The entrepreneurial founder continues as managing director, operations director and decision maker. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 65 Table 3. Decision problems and decision support activities for the 10 companies Firm Level A B C Cognitive Level Decision Problems Decision Support Activity 4 Understand how medical and technology advances and government decisions will change patient care provision and revenue model Some information for contract negotiation with health care purchases in Discovery mode is available, but managers should run scenarios to understand the impact of bottom line and operations 3 Optimising resource utilisation with improved financial performance enabling benchmarking between hospitals is a critical activity Resource utilisation modelling is available in areas such as outpatient metrics, theatres and bed management across the hospitals. Information derived from level 2 is used to make predictions for changes in health sector. 2 Accurate analysis and tuning of local company performance across complex indicators Ad-hoc assessment of key business metrics in financial and clinical areas across all hospitals – bed occupancy, theatre utilisation etc. is available 1 Managers seek to measure all aspects of operational and financial performance to improve services delivered, and patient and financial outcomes. Reporting activity is well developed. A Hospital Information System (HIS) enables the management of scheduled admissions, theatre scheduling and staff/consultant workload. A data warehouse has been developed as well. 4 More competition has been introduced in the residential gas market – must lose market share down to a set level. In the new single wholesale Electricity market, company B is a new entrant - How will it operate in this market? The effect of global warming on energy demand is also a key uncertainty Regression analysis assesses the relationship between gas demand and degree days, price change and customer segmentation. The dataset represent 60% of the residential and small temperature sensitive Industrial and Commercial customers. The purpose is to discover what the operational environment may be like and the implications for the energy trading business, especially in terms of pricing going forward. 3 The decisions made based on the projected price of electricity are of material value to the business. In-depth knowledge of the workings of the market is required. An informed view of where the SMP (System marginal price) will be for each half hour of the day is a key strategic asset as well as an operational asset as it helps to determine what contracts should be entered into, and to manage capacity on a day to day basis. Portfolio modelling applications are used to support the identification/ prioritisation of gas and electricity commercial activities The organisation has invested in 2 market modelling applications to help in its forecasting of the SMP price. SMP price together with the business hedging strategy for the following 12 months determines what contracts are entered into and for what prices and quantities. 2 The organisation recognises the importance of analytics where optimisation and efficiency are key components to operating in a new energy trading environment There are a number of systems in use which allow a level of scrutiny. Market-to-market reporting is used to predict the future benefit derived from entering into forward transactions enabling management to optimise purchase contracts, and allowing corrective action should the firm’s hedging strategy require amendment. 1 All aspects of the ‘claims management’ area must be monitored in near real time Recent systems developments have replaced Excel spreadsheet reporting, and has enabled the capability of data analysis based on data warehouses 4 The raw material of cheese is milk, ie 90% water but managers do not know how to address the issue of yield and efficiency n/a 3 Dry hot summers mean poor milk yield and low milk quality which increases the cost of cheese but the reasons for these variations are unclear Available systems compute these variations but cannot help with diagnosis or corrective actions. continued on following page Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 66 International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 Table 3. continued D E F 2 Controlling fixed costs and managing the milk throughput are critical. Understand the reasons for spoilage, analysis of the relationship between milk quality and cheese recipe used. Critical KPIs at scrutinising level are all produced manually based of various SCADA and forecasting systems. Excel spreadsheets are prepared and hand delivered to management weekly, 2 working days after each weekend. 1 Company C produces cheese more efficiently than any of its competitors. Maintaining that efficiency is a core competency which drives a sustained competitive advantage. Relevant CSFs are based on a system of budget vs actual variances Company C excel in dashboard technology to control and monitor all aspects of the production process. KPIs are reported upon in dashboard format and include: Milk cost per tonne of cheese, Direct wages cost per tonne of cheese, Direct energy cost per tonne of cheese for instance. 3 Competition both internally and externally is forcing the Cork site to consider its cost structure n/a 2 From the CSF’s monitored at level 1, a core set of key performance indicators (KPI’s) are produced and reviewed, with the frequency of review being determined both by the criticality of the operation and the availability of information. Little drilldown capability is available to managers to facilitate scrutinising. Reports are mostly static. 1 The Cork site has a number of critical success factors (CSF’s) that if managed effectively can ensure the site is a success. Current reporting systems monitor day-to-day operations and the ERP provides some data. However manual systems generate most of the weekly reports prepared by Finance. An “equipment effectiveness” dashboard allows drilldown in each machine’s downtime but is not integrated with any other system 3 When increased resolution times are apparent, management can predict the potential impact on service levels based on the volume of service calls, the number of staff, and the introduction of new products and the quality of training. Each business unit has visibility of specific hardware products dashboards, with defective attributes flagged. This in turn allows the Global Services unit to flag product issues to the engineering organisation, or to roll out further training where appropriate. 2 Improving management ability to investigate the reasons for the outcomes at level 1, where the cause and effect relationship is not as factual or evident is a critical factor Scrutinising the performance of the business units and their ability to meet SLA’s can highlight problems – for newly released products for example. This information is derived from level 1 systems and further manipulated manually. 1 Improving management ability at problem solving, maintaining customer SLA agreements, and tracking compliance of documented processes is essential. This is presented in Dashboard format with colour coding to indicate if SLA levels are not met. 4 What disease are emerging and how to support them. How will US government decisions on healthcare insurance bill influence the product portfolio n/a 3 Effects of corporate / market changes on Cork Plant Manual collation and manipulation of data from external market research continued on following page Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 67 Table 3. continued G H I J 2 Plant specific strategy aligned to Corporate. Monthly ranking analysis across 23 plants, poor performance trend analysis at plant level, Customer complaint analysis at plant level. Excel based Ranked League table generated (at corporate level) based on information derived at levels 1 and 2, and analysing performance across the 23 plants – then fed back to Cork. Long term trending difficult to achieve and requires considerable manual manipulation – using Excel. Extensive analytics skills in place across many departments 1 Analysis of performance based data based on 9 panel Balanced Scorecard for operational problem solving and quality metrics - resource allocation, project start, project cancellation, issue escalation, customer complaint analysis. SAP and SAP BW implemented. Very strong on production data capture, but reporting is siloed and manually collated. Quality data captured on Excel showing weekly trends based on complaints 3 How to increase market share and profitability? How to fine tune the set up of the UK operations? All scrutinising is based on external information – published reports, governments and European “green” strategy policy, waste industry specialists and market analysts, grid connection regulations. Excel is the only tool but staff analytical skills are high. 2 Contract negotiation – ie analysis of what contracts to sign Excel used by engineering staff to monitor waste tonnage and price charged by waste operators, type of waste gas yields – fed into excel-generated financial models for sensitivity analysis.. 1 Day to day operational effectiveness. SAGE and Excel are the main systems . 3 How to increase market share and profitability e.g. launch of product to new market? Excel the main scrutinising tool using data from the data warehouse plus external market research data and tacit information from marketing specialists. 2 Weekly review by CEO on all products with emphasis on contribution to bottom line Business analyst uses Cognos/PowerPlay for weekly report based on Data warehouse updated in Level 1. 1 Monitoring of production and sales targets and KPIs SAP used daily to record all transaction. Data warehouse updated once daily. Cognos/PowerPlay BI toll available across enterprise for reporting and drilldown capability. 3 Declining demand for carbonated soft drinks Move towards healthier products and lifestyles, n/a 2 Implementation of modern manufacturing tools – lean, six sigma. Increase capacity without extra resources. The only IT system is MS Excel at this level, as inquiry and reporting from MAPICS and LMS are transactional and not integrated. 1 Plant manager responsible for operations and quality. Local area management (not reporting to plant manager) responsible for engineering and supply chain. Manual (by supply chain based on customer orders) input to MAPICS generates production requirement weekly. Interfaced to Oracle, which is system of record for production. Quality data based on manually extracting Lab results. 1 Matching supply with customer demand through tight Inventory management, control over purchasing etc. Effective cash-flow management. Better control of operations through managing a core set of KPIs available to staff, customers and third party vendors ERP system provides fast and reliable financial reporting and analysis. Excel used for further selective reporting. KPI portal provides data in report and dashboard format on predefined KPIs. Excel remains the tool for any scrutinising activity, but staff analytics skills are poor. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 68 International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 in a standalone fashion, with each hospital implementing different IT systems. This created difficulties with preparing routine management and financial reports and in operational and strategic planning. Since 2006 a Business Intelligence Data Warehouse (BIDW) is being implemented. Whilst the BIDW project was clearly focused on providing robust and comprehensive visibility on operations, it has become the platform for the full spectrum of managerial decision support from reporting to scrutinising to discovering. Whilst level 2 is well covered by the implementation of a benchmarking concept, levels 3 still presents specific design difficulties as managers seek to understand how they can use the data warehouse to face up to the challenges of the future. The lack of a model to capture the essence of decisions in this domain remains a problem. Furthermore there is no evidence that specific systems for discovery use for problems identified at levels 4 and 5 have been considered. 4.2.2. Company B Company B is a commercial State Body operating in the energy industry. The company is wholly owned by the Irish Government and consists of 2 main businesses – Gas transportation and Energy Supply. The residential gas market is the primary area of business. A new wholesale electricity market has come into operation in Ireland since November 2007. The effects of global warming and improved housing insulation standards will affect demand for energy in the future. Company B entered the retail electricity market in 2006, and currently holds 12% of the electricity market in Ireland. Company B is an interesting site from a decision support viewpoint, as outlined in Table 3. The first observation that can be made is that the engineering vocation of the firm has helped the creation of an “all-knowing” dashboard for reporting in real time on all security elements of the network. Flow, pressure, consumption etc. are monitored in real time. The reporting on maintenance and accidents is also very ad- vanced. On the commercial side, company B is extremely mature in its development of highly complex models for planning for consumption and justifying the price per cubic meter charged to the different categories of customers (which the Department of Finance must approve once a year). This has been largely based on spreadsheets of a highly complex nature, developed by specialists in econometrics and business modelling. Based on the generic scenarios, managers in the transportation department run simulations which are then used for price setting or also for justifying capital expenditure. Altogether, this portfolio of applications adds up to a complex set of decision support covering the reporting and scrutinising side very comprehensively, and making a definitive contribution at the discovery level. 4.2.3. Company C Company C is a major international cheese manufacturer and also manufactures food ingredients and flavors. Headquartered in Cork, Ireland, it produces 25% of the total cheese manufactured in Ireland, and has been the largest manufacturer of cheese in Ireland for the last 20 years. Company C do not have any decision support systems to support upper level management decision making. All management reports are prepared on spreadsheets, with input from disparate transactional systems and SCADA-type (supervisory control and data acquisition) process control systems. In this site, the failure to support higher level decision activities is very evident and we could not identify any significant attempt to cover any decision need at levels 3, 4 or 5. This, however, was in sharp contrast with our findings at level 1 and 2, which clearly showed intense reporting and some limited scrutinising activities. A substantial body of mature DSS applications was developed over a number of years in the shape of dashboard type applications and a substantial body of manual preparation of data used for scrutinising operations was also undertaken. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 69 4.2.4. Company D Company D is a medical device manufacturer, and is part of a US multinational. This company has seven manufacturing sites around the world, with a new facility currently being built in China. The Cork site is the largest manufacturing facility, accounting for approximately 40% of total production. For products in this market, gaining additional market share is largely dependent on price competiveness, with significant competition in the market. Although a large US multinational firm, Company D seems remarkably close to Company C in decision support terms, despite having a totally different profile in general terms. This is more than likely due to our examination of a local manufacturing site, rather than the corporation overall. Managers are very well equipped at level 1 and 2, where KPIs are clearly identified, but decision making tools for scrutinising in general terms and for discovering are totally absent. This reflects the KPI-oriented culture of many MNCs where specific goals are handed down from headquarters to local sites for each functional area and converted into strict targets by each manager. This culture means that the incentive and the time to develop specific DSSs at the higher levels of decision making are low because local managers have little autonomy of action. 4.2.5. Company E Company E is a world leader in information management and data storage products, services and solutions. For the purposes of this study, the Global Services (GS) division was the focus. Global Services is Company E’s customer support organisation, with almost 10,000 technical/ field experts located in 35 locations globally and delivering “follow-the-sun” support in over 75 countries worldwide. An Oracle CRM and workflow system provides key operational data, including install base data, time tracking and parts usage. Business objects and Crystal reporting software is used for querying and reporting as required. Company E presents a profile that is similar to that of company D, as a part of a US MNC, with the difference that, in this case, our access in the case allowed us to study a global unit, rather than a local manufacturing unit. This results in a more complete landscape of support tools, all the way up to level 3, where production problems and training needs can be anticipated before anyone has considered training to be a problem. This illustrates the natural progression of all decision problems up the levels of the framework over time, from the stage where managers cannot even express them properly, to the stage where they become part of the normal scrutiny activity of the firm, and, given time, fall into the general reporting area, based on well-defined models that capture the essence of the decision problem. 4.2.6. Company F Company F is a worldwide developer and manufacturer of medical devices, and is part of a US multinational. It has advanced the practice of less-invasive medicine by providing a broad and deep portfolio of innovative products, technologies and services across a wide range of medical specialities. Current US Government initiatives in regard to the US healthcare bill present a challenge for the organisation, and the move towards universal health insurance may mean substantial changes as to how healthcare is provided and funded in the US in the future. The Cork plant was set up in 1998 as a manufacturing site for all Neurovascular products worldwide. Each site is measured on a nine panel Balanced Scorecard metric based on operational and quality metrics which are integrated with the corporate goals and objectives. A league table generated by corporate, ranks the twenty three plants in terms of performance and alignment with corporate goals. This could be classified as level 3 decision problem classification, as senior management can clearly state the problem/requirements, and know what options are available to them in order to execute the requirements. This researcher found little evidence of level 4 or 5 decision problems being considered with a view towards their resolution. All other decision making is at levels one and Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 70 International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 two and is of the nature of resource allocation, resource hiring, project cancellation and new project starts, and escalation of issues to external (outside of Cork) resources depending on severity. Most of the information provided to managers for decision making is in the form of dashboards of balanced scorecards. The data is manually extracted though specific inquiries, onto MS Excel spreadsheets, and manipulated and formatted as required. This inquiry and transformation is completed by financial analysts and administration personnel who have specific expertise and training in the process. management use this information for the full spectrum of managerial decision support from reporting to scrutinising to discovering. The metric reporting interval varies and can be hourly, daily or even weekly. An SAP ERP / BW (Business Warehouse) platform provides key operational data, including all production and quality data. Business objects reporting software is used for querying and reporting as required. There is very little integration between systems, particularly between the main transactional SAP system, and other bespoke in-house developed systems, which also record business transactions. Managers are aware of decision problems which would be classified at levels 4 and 5, where the uncertainty of the business environment would impact the day to day business decisions. However, at a local manufacturing site level, environmental uncertainty has been removed, and achieving a high performance ranking relative to the other plants, is essential for the overall and continued success of the local plant. on spreadsheets of a highly complex nature, developed by specialists in econometrics and business modelling. Expansion to new markets, preferably where higher electricity prices are available, is a key decision in the pursuit of increased shareholder value. In a start-up environment, senior managers are continually scanning for opportunities, and this is the only organisation within the study where the organisation agenda has not been completely set. One of the most interesting findings during the study was how decision making in this case revisits higher decision problem levels as the decision making progresses, i.e. decisions made at level 3 progressed for sensitivity analysis at level 2, but went back to level 3 or even 4 for further refinement. This is in marked contrast to all the other companies in the study, who followed a top-down type progression. As a start-up company, Company G has minimal traditional type IS systems, and use personal type applications for transaction recording. Extensive use is made of external information from sources such as published reports and government strategy reports on green policy and the regulatory framework for electricity price setting and for grid connection. With just thirty employees, manual scrutinising of both qualitative and quantitative information is completed by highly skilled engineers who perform sensitivity analysis of models and information. There is considerable evidence of discovery type activity in place in this case, but there are no tools available. This is a company where understanding and knowledge of the company’s strategy as well as the industry and environmental factors is key. 4.2.7. Company G 4.2.8. Company H Company G is an Irish-based bio-science company which focuses on electrical power generation in Ireland and the UK from sustainable fuel sources or “green energy” using technologies such as gasification and dry fermentation. The company has recently set up operation in the UK. Company G is an engineering firm and has a highly skilled analytics ethos in place, based Company H is part of a French wine and spirit company since 1988, which is one of the largest wine and spirit companies worldwide. However the Irish company was formed in 1966 when three distilleries amalgamated, and all whiskey production transferred to one site. It intends to continue its international development, strengthened by an enriched portfolio of Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 71 brands, an increased global presence and an efficient decentralised organisation. Decentralised decision-making constitutes a key principle of the parent organisation. The Group’s Holding company defines the Group’s strategy and its main policies, but local management adapts this strategy to their local markets. Thus, a culture of decentralisation allows the Cork site CEO some level of autonomy and level 3 decision problems are identified and resolved. Lever 1 and 2 decision problems focus on efficiency and product competitiveness issues. Company H has a very sophisticated portfolio of systems which are utilized extensively in the organisation. A SAP ERP system and COGNOS PowerPlay BI tools are implemented enterprise wide. Use of external market research data is also made, including the International Wine and Spirit Records (IWSR), which maintains data on the majority of drinks companies, such as their brands, sales volumes by brand and other relevant information. Company H employs some very skilled analysts who combine internal and external data for scrutinising and reporting. However there was no evidence of higher level discovery type analytics available. Company H operate in a very mature market, with an abundance of external information providers based of trading and export figures, consumer surveys and other market intelligence. It has access to, and makes extensive use of this intelligence. Moreover, the information available is downloadable to MS Excel which can be merged with internal data. Therefore the quality of decision support can be classified in a very positive way. However, the decision making requirements are once more indicative of operating in a subsidiary, where the primary objective is to ensure the adherence to corporate strategy as well as to meet the objective as set out by corporate Company H is a company where decision support more than adequately matches the requirements of the manager’s decision making problems. The light shading reflects the delivery of support at levels 4 and 5 was still largely an aspiration at the time of the study. Whilst levels 2 and 3 are well covered by reports produced by the skilled analysts employed in the company, these are not automated, and are reflected by the mid level shading on table 3. The lack of a model to capture the essence of decisions at level 3 remains a problem. Furthermore there is no evidence that specific systems for discovery use for problems identified at levels 4 and 5 have been considered. 4.2.9. Company I Company I is the fourth largest food and beverage business in the world, and is a world leader in convenient foods and beverages, with revenues of over $27 billion. In recent times, it has changed its strategy in response to a changing market, and since 2006 the primary focus of the organisation had been the creation of healthier products and reducing the organisations negative impact on the environment. The business activities located in Cork include the manufacture of concentrate (exported to 105 countries worldwide), laboratories, financial shared services (supporting 65 countries), IT providing support to global operations and R&D. Company I operate within a very hierarchical structure, and with some autonomy in day-to-day activities. The main site objective is to manufacture concentrate product of high quality, made safely with minimum net impact on the environment and shipped on time to the customer. However a long term goal set by HQ is to develop the Ireland site as a centre of excellence within the group. A structured problem identification and analysis approach to decision making had been adopted in company I, known as DMAIC, (Define problem clearly, Measure impact of alternatives, Analyse root cause, Improve and Control). This approach serves level 2 and 3 decision problems well, and examples would include corrective action where necessary to ensure the quality of the product, or managing within budgetary constraints, but also decisions considering the implementation of modern manufacturing tools including lean and six sigma. Company I has a number of systems available. The production plant has a very low level Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 72 International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 of automation. The hierarchical structure of the organisation highlights the lack of systems integration, and results is a disconnect between the plant manager and the functional local area managers. The local area manager for supply chain compiles the production schedule weekly, based on customer orders and raw material availability which are maintained on a MAPICS inventory system. The production schedule reports are “delivered” to operations, who then schedule the following week’s production. Even though all the data is MAPICS based, the report is formatted on MS Excel for production. Timein-motion studies are all manually recorded. Reporting on product quality is also based on manually extracting Laboratory information systems data. However any scrutinising activity is based on manually captured and recorded data used as the basis for MS Excel models for scenario testing. Managers are well equipped at level 1 reporting only. Even at level 2, the plant manager seems to have his own reporting mechanisms through MS Excel and his own knowledge and gut feeling for what is happening in the plant, which facilitates scenario testing or root cause analysis. As discussed Level 1 day-to-day decisions are well covered, reflected by the deep shading. Any reporting and scrutinising at levels 2 and 3 are based on individual managers own reporting capability created using MS Excel. This reflects the non-KPI-oriented culture, and while specific goals are handed down from headquarters to local sites for each functional area, some managers manage in their own style. 4.2.10. Company J Company J is a supply chain management company with a product portfolio across consumer electronics, personal computers, medical devices and telecommunications. It offers a wide range of services from product design through to fulfillment direct to the hub or to the end client. The company was founded in 1996 in Ireland, and maintains its company headquarters there, while it operations headquarters is in Shenzhen, China. At an operational level, management have recently introduced a core set of KPIs across all functions in the organisation to ensure better control of operations, and more effective cash-flow management. Matching supply with customer demand is an integral part of the business, and requires tight inventory management and control over purchasing. Other than operational decision problems which can be categorised at levels 1 and 2 of the Humpreys and Berkley framework, there was no evidence of any other decision problems being discussed within the organisation. Company J has a very sophisticated portfolio of enterprise systems in place. The value of information and the ubiquity of the internet have been leveraged to provide customers with up-to-date and accurate information on their customers’ orders at all stages of the delivery process. From the company’s inception, it has had an in-house development team based in South Africa, and information systems providing online order status over a robust network were the first to be prioritised. However the lack of systems ensuring an accurate and comprehensive system of record for the company proved a major drawback when the managing director needed fully consolidated accounts for a joint venture due diligence exercise. Following this, an organisation-wide fully integrated ERP system was implemented in 2009. However none of the historical data was migrated. The emphasis has been on providing their customers and third party suppliers traceability on their (customers) orders from a logistics perspective. The ERP implementation has coincided with the roll out of Key Performance Indicators (KPI’s) across the organisation. Currently the ERP system is perceived to provide fast and reliable financial reporting and analysis. MS Excel used for further selective reporting and scrutinising, but staff analytics skills are poor. Managers are well equipped at level 1 reflected in Table 3. Levels 2 and 3 rely on MS Excel for reporting and any scrutinising activity which relate to the KPI’s are as a result of MS Excel extracts for the ERP system, and subsequent manipulation of the data into required formats. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 73 4.3. Measuring Decision Support Maturity: the Cross Case Analysis The data presented in section 4.2 provides a picture of the usage of DSS applications in the 10 case studies in our study. In some cases, the case data is factual and outlines specific applications used by managers in the case, whereas in other cases, it is largely aspirational in that little is known about how to design the support tools, although the agenda for decision making has been set (Mintzberg, 1994). Table 4 presents a quick summary of our observations in terms of the levels of decision problems we observed in the ten companies and where formalized decision support is in place. The cross case table indicates that the broad spectrum of firms we included in our sample is matched by a broad spectrum of findings with respect to the maturity of the use of decision support tools. Cells coded in black typeface indicate that managers were able to identify decision problems at the corresponding representation level and that formal decision support was available to them. In the scenario where decision problems existed but no matching formalised decision support was available (the cells coded in italics in Table 3), we distinguished two different categories: (1) Heavy grey typeface indicates that managers could indicate decision problems, but that decision support was more an aspiration than a reality – for instance, systems may be able to compute variances, but may provide no help in investigating them; (2) light grey typeface indicates that no decision support of any kind was available to managers for the problems they identified at a particular representation level. All in all, the decision support maturity of firms is revealed in Table 4 in the extent to which all levels of the framework are covered and Table 4 is ordered from left to right in terms of decreasing decision support maturity. For instance, firm E is ranked third because formalised decision support is available up to level 3, ahead of firms C and F where problems are identified up to level 4, but no formalised systems are available to provide decision support beyond level 2. Our primary observation that no company has formalised support for level 5 problems is compounded by the observation that many managers could not describe level 5 problems that they were facing. This is a limitation of the study, probably reflecting the level of managers we spoke to, insofar as, given the organisational context described in Table 2, it was legitimate to expect that some of these firms definitely faced level 5 pressures. Furthermore, only 40% of the companies in our sample have conceptualised problems they face at level 4. Ninety percent have considered decision problems at level 3, whilst 1 company appears to be concentrated on lower level problems. In this case, it may be that the nature of the firm’s business has a bearing on the problems managers face. Clearly finding new contracts and new customers would rank at level 3 and 4 in the table (Table 3, Company J), but the managers engaged in this activity where not available for interview. The managers interviewed were focused on execution rather than analysis. Thus, measuring DSS maturity using our framework requires careful sampling Table 4. Summary of the observations in the 10 companies of the sample Firms B A Level 4 X X Level 3 X X E C F X X X X G H I D X X X X J Level 5 X Level 2 X X X X X X X X X X Level 1 X X X X X X X X X X Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 74 International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 of respondents within each case study, which is another limitation of this study, insofar as the data collection mechanism did not lend itself to a well controlled selection of informants. In future iterations of this research projects, we will give more definitive instructions to each student group in terms of sampling and access to higher managerial level, though this will remain more difficult to achieve in MNCs than in smaller indigenous firms. This observation may be reflected in the fact that the MNCs in our sample did not rank high in DSS maturity terms. This discussion about MNCs leads to another key finding of the study. The data reveals that companies can display a given level of DSS maturity for different reasons, notably lack of expertise as in company C where some variances could be computed in available systems but little skill was there to do any more, or lack of incentive as in company D where managers did not seem to be empowered to conduct inquiries into problems which they could see, which is quite different. Thus, the existence or absence of decision support at the scrutinising and discovery levels is about more than just the abilities of the managers and DSS developers of the firm to properly model the issues facing them. Managers must also recognise the need to perform such activities and feel that the amount of autonomy that they have warrants the significant efforts required in conceptualising the problems. Given limited attention (Simon, 1977), managers may prefer to concentrate on level 1 or 2 which allows them to manage the narrowly-focused KPIs handed down to them by top management, where there is little or no discretion in the choice of procedures used to structure the decision problem and formulate a policy for action. In firms where the context facing managers provides clear incentives to (1) attempt to formalise level 3 and level 4 problems and (2) to seek the help of developers in taking their decision support tools beyond simple end-user developed spreadsheets, organisations may display very complete portfolio of decision support applications spanning three levels (companies A, B, and E). However, even in these firms where clear incentives are present, it will remain that, few organisations ever achieve a complete portfolio spanning 4 levels, let alone 5 levels on a permanent basis. In other words, reaching level 5 is not like reaching a threshold at which one is certain to remain. Quite the opposite, it is a matter of reaching a certain level of understanding of the problems facing the firm, at a particular point in time, where the environment is presenting a new, identifiable pattern of competition, regulation etc… until Nature’s next move changes the state of play again and managers shift their focus onto other, newer ideas, as they become aware of new challenges facing them. Yesterday’s level 5 problems become level 4 or 3 problems, or drop off the agenda altogether. Tomorrow’s level 5 problem, of course, will take time to crystallise. A final interesting finding from a pedagogical viewpoint was that the best presentations of the decisions levels and how the decision maker was facilitated were from managers in organisations where a comprehensive range of decision support systems are in place, i.e. organisations where an almost complete portfolio of information systems have been developed, and are used extensively by managers. In that sense, DSS maturity may be a component of the broader concept of IT maturity as previously defined in the literature. These observations point to an important finding, which is that the DSS maturity of an organisation is a complex indicator which captures two distinct elements at least. On the one hand, managers must understand and be motivated by the notion that Decision Support is one activity which can be undertaken as part of decision making. Asking questions and seeking answers is something that all managers do as part of their job, but it is the nature of the questions they ask themselves, as measured by the Humphreys and Berkeley framework which matters. Some managers are satisfied with asking operational questions and seeking operational answers, whereas others are motivated by higher level problems, and the context in which they operate. We can say tentatively at least, that level 3 is a critical level in the framework, which is in keeping with Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 75 the theoretical presentation of the Humphreys and Berkeley framework, but had never before received specific empirical validation. 4.4. Issues with the Study When analysing the feedback from the complete research data set, including the cases we excluded because we were not satisfied with the quality of the data collection carried out by the groups, it becomes evident that the classification of decisions resulting from the analysis as we carried it out can become distorted in a number of ways: (1) The manager’s perception of their own position in an organisation influenced their perception of the level of the decision, and most of the managers overstated the level of decisions made. This was especially true in organisations where ‘strategic goal alignment’is part of the day-to-day organisational culture and managers mistakenly equated their strategic role with decision levels. (2) The degree of discretion available to the manager influenced the determination of decision level. Where discretion levels were high, the managers presented a higher decision level classification. (3) Some managers were swayed by the terminology ‘strategic, tactical and operational’ and an alignment with the abstraction levels, and reverted to their own interpretation of these terms within their own organisations. This is a related but different bias to the first outlined. (4) Finally, many managers identified decisions by the information systems or decision support systems which provided the decision maker with the required information to make the decision. Moreover, the classification of decision level based on the concepts of reporting, scrutinising and discovery was far more accurate that through any other mechanism. During the feedback sessions we organised, the students agreed that, in general, decision level classifications were overstated by at least one level. Thus, managers find it difficult to measure the degree of abstraction of an idea in conjunction with the degree of formalization of the solutions they apply to it. This is an interesting observation on the concept of representation level as proposed by Humphreys and Berkeley: it is not spontaneously understood by managers. The feedback session discussions facilitated the realignment of the decision level classification in the 10 cases, such that the data in Tables 3 and 4 was corrected and is accurate as presented. 5. ConCLuSIon This research study has been successful in applying existing frameworks to develop a method for evaluating the maturity of organizations with respect to the use of decision support tools. It confirmed the usefulness of our adaptation of the Humphreys and Berkeley framework in measuring DSS maturity and provides some up to date empirical validation of this framework which had not been used in this direct fashion up to now. Our observations across 10 case studies of Irish firms confirms and updates the findings of previous research in DSS, that the higher levels of abstraction in decision making are not covered by decision support, either formal decision support by DSS or decision support by other softer mechanisms. The use of the framework, however, allows us to measure the “DSS maturity gap” with some accuracy. Only one of our 10 firms has any concrete decision support above level 3 in the Humphreys framework and only 4 have conclusively considered what issues could be supported at level 4. Despite 40 years of research on DSS since the seminal Gorry and Scott Morton paper, little is still known about how managers could be supported in that part of their work that requires the most vision. We feel that it is difficult to argue that this is not a disappointing observation for the decision support area, but it is one which it is important to validate with up to date empirical data. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 76 International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 Finally, another important finding of our research is that it is difficult to engage with managers on the topic of decision making and decision support. Even in the relatively controlled environment of the class room, discussing real life organizations and the problems they face, on the basis of a well explained grammar (the framework in Figure 1), discussions with managers on the topic still reveal the possibility of important bias and misrepresentation. We are committed to continue our research towards a grammar and set of questions that will allow us to scale up our data collection at the level of a survey. Clearly, we have not achieved this stage and our investigations must continue. This is a condition for being able to deliver concrete and tangible prescriptions for managers and decision support staff in organisations. RefeRenCeS Ackoff, R. L. (1967). Management misinformation systems. Management Science, 14(4), 147–156. doi:10.1287/mnsc.14.4.B147 Adam, F., & Pomerol, J.-C. (2008). Developing practical decision support tools using dashboards of information . In Burbstein, F., & Holsale, C. W. (Eds.), Handbook on decision support systems (pp. 151–173). New York, NY: Springer. doi:10.1007/9783-540-48716-6_8 Damasio, A. R. (1994). Descartes’ error: Emotion, reason and the human brain. New York, NY: Grosset/Putnam. Dover, C. (2004). How dashboards can change your culture. Strategic Finance, 86(4), 43–48. Earl, M. J. (1989). Management strategies for information technology. Upper Saddle River, NJ: Prentice Hall. Earl, M. J., & Hopwood, A. G. (1980). From management information to information management . In Langefors, B., Verijn-Stuart, A. A., & Bracchi, G. (Eds.), Trends in information systems (pp. 315–325). Amsterdam, The Netherlands: North-Holland Publishing. Fahy, M., & Murphy, C. (1996). From end-user computing to management developed systems. In Proceedings of the 4th European Conference on Information Systems, Lisbon, Portugal, (pp. 127-142). Galliers, R. D. (1993). IT strategies: Beyond competitive advantage. The Journal of Strategic Information Systems, 2(4), 283–292. doi:10.1016/09638687(93)90007-W Gitlow, H. (2005). Organizational dashboards: Steering an organization towards its mission. Quality Engineering, 17(3), 345–357. doi:10.1081/QEN200059847 Humphreys, P. (1989). Intelligence in decision support - a process model. In Doukidis, G. I., Land, F., & Miller, G. (Eds.), Knowledge-based management support systems. Chichester, UK: Hellis Hovwood. Alter, S. (2004). A work system view of DSS in its fourth decade. Decision Support Systems, 38(3), 319–327. doi:10.1016/j.dss.2003.04.001 Humphreys, P., & Berkeley, D. (1985). Handling uncertainty: Levels of analysis of decision problems . In Wright, G. (Ed.), Behavioral decision making. New York, NY: Springer. Anthony, R. N. (1965). Planning & control systems: A framework for analysis. Boston, MA: Harvard University Press. Keen, P., & Morton, M. (1978). Decision support systems: an organizational perspective. Reading, MA: Addison-Wesley. Arnott, D., & Pervan, G. (2008). Eight key issues for the decision support systems discipline. Decision Support Systems, 44(3), 657–672. doi:10.1016/j. dss.2007.09.003 Khandelwal, V. K., & Ferguson, J. R. (1999). Critical success factors (CSFs) & the growth of IT in selected geographic regions. In Proceedings of the 32nd Annual Hawaii International Conference on System Science (Vol. 7, p. 13). Checkland, P. (1981). Systems thinking - systems practice. Chichester, UK: John Wiley & Sons. Crawford, A. B. Jr. (1982). Corporate electronic mail - a communication-intensive application of information technology. Management Information Systems Quarterly, 6(3), 1–13. doi:10.2307/248652 King, W. R. (1985). CEOs & Their PCs. Management Information Systems Quarterly, 9(3), xi–xii. Kotter, J. P. (1982). What effective general managers really do. Harvard Business Review, 60(6), 156. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 77 Lee, A. S. (1994). Electronic mail as a medium for rich communication: An empirical investigation using hermeneutic interpretation. Management Information Systems Quarterly, 18(2), 143–157. doi:10.2307/249762 Levine, P., & Pomerol, J.-C. (1995). The role of the decision maker in DSSs and representation levels. In Proceedings of the 29th Hawaii International Conference on System Sciences (Vol. 3, pp. 42-51). Mintzberg, H. (1973). Nature of managerial work. New York, NY: Harper & Row. Mintzberg, H. (1994). The rise & fall of strategic planning. Upper Saddle River, NJ: Prentice Hall. Mintzberg, H., Raisinghani, D., & Theoret, A. (1976). The structure of ‘unstructured’ decision processes. Administrative Science Quarterly, 21(2), 246–275. doi:10.2307/2392045 Negash, S., & Gray, P. (2008). Business intelligence . In Burstein, F., & Holsale, C. W. (Eds.), Handbook on decision support systems 2 (pp. 175–193). New York, NY: Springer. doi:10.1007/978-3-540-48716-6_9 Nolan, R. L. (1973). Managing the computer resource: A stage hypothesis. Communications of the ACM, 7, 399–405. doi:10.1145/362280.362284 Nolan, R. L. (1979). Managing the crisis in data processing. Harvard Business Review, 2, 115–126. Panko, R. (2006). Spreadsheets & Sarbanes–Oxley: Regulations, risks, & control frameworks. Communications of the Association for Information Systems, 17(1), 29. Pounds, W. F. (1969). The process of problem finding. Industrial Management Review, 11(1), 1–19. Rockart, J., & DeLong, D. (1988). Executive support systems: The emergence of top management computer use. Homewood, IL: Business One Irwin. Scott Morton, M. (1986). The rise of management computing . In Irwin, D. J. (Ed.), The state of the art of research in management information systems (pp. 325–353). Cambridge, MA: MIT Press. Simon, H. A. (1977). The new science of management decision. Upper Saddle River, NJ: Prentice Hall. Simon, H. A. (1997). Administrative behavior. New York, NY: Free Press. Vile, D. (2007). Vendors causing confusion on business intelligence - is the marketing getting out of hand? Retrieved from http://www.theregister. co.uk/2007/04/09/bi_ws_wk2/ Watson, H. J., & Frolick, M. N. (1993). Determining information requirements for an EIS. Management Information Systems Quarterly, 17(3), 255–269. doi:10.2307/249771 Nolan, R. L., & Crosson, D. C. (1995). Creative destruction: A six stage process for transforming the organisation. Boston, MA: Harvard Business School Press. Mary Daly is a lecturer in Business Information Systems at University College Cork (UCC) in Ireland. Her research interests include executive decision-making and decision support. Prior to joining UCC, she has gained considerable experience in industry in senior IT Management positions. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 78 International Journal of Decision Support System Technology, 3(2), 57-78, April-June 2011 Frédéric Adam is Associate Professor in Business Information Systems at University College Cork (UCC) in Ireland and Visiting Research Fellow in the School of Economics and Management at Lund University (Sweden). He holds PhDs from the National University of Ireland and Université Paris VI (France). His research interests are in the area of decision-making and decision support and in the area of ERP. He has authored and edited 7 books and has over 20 journal papers published in international journals including Information and Management, the Journal of Strategic Information Systems, Decision Support Systems, and the Journal of Information Technology. He is the Editor-in-Chief of the Journal of Decision Systems and the Chair of the Working Group 8.3 on DSS of the International Federation for Information processing (IFIP). In 2010, he was awarded the IFIP outstanding Service Award for his on-going work within the working group. He has recently been appointed Head of the Graduate School of the College of Business and Law at UCC. Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Decision Support System Technology An oficial publication of the Information Resources Management Association Mission The primary objective of the International Journal of Decision Support System Technology (IJDSST) is to provide comprehensive coverage for DMSS technology issues. The issues can involve, among other things, new hardware and software for DMSS, new models to deliver decision making support, dialog management between the user and system, data and model base management within the system, output display and presentation, DMSS operations, and DMSS technology management. Since the technology’s purpose is to improve decision making, the articles are expected to link DMSS technology to improvements in the process and outcomes of the decision making process. This link can be established theoretically, mathematically, or empirically in a systematic and scientiic manner. Subscription Information IJDSST is published quarterly: January-March; April-June; July-September; October-December by IGI Global. Full subscription information may be found at www.igi-global.com/IJDSST. The journal is available in print and electronic formats. Institutions may also purchase a site license providing access to the full IGI Global journal collection featuring more than 100 topical journals in information/computer science and technology applied to business & public administration, engineering, education, medical & healthcare, and social science. For information visit www.infosci-journals.com or contact IGI at eresources@igi-global.com. Copies of Articles Copies of individual articles are available for purchase at InfoSci-On-Demand (www.infosci-on-demand. com), an aggregated database consisting of more than 31,000 research articles written by prominent experts and scholars in the ield of information/computer science and technology applied to business & public administration, engineering, education, medical & healthcare, and social science. Copyright The International Journal of Decision Support System Technology (ISSN 1941-6296; eISSN 1941-630X). Copyright © 2011 IGI Global. All rights, including translation into other languages reserved by the publisher. No part of this journal may be reproduced or used in any form or by any means without written permission from the publisher, except for noncommercial, educational use including classroom teaching purposes. Product or company names used in this journal are for identiication purposes only. Inclusion of the names of the products or companies does not indicate a claim of ownership by IGI Global of the trademark or registered trademark. The views expressed in this journal are those of the authors but not necessarily of IGI Global. Correspondence and questions: Editorial: Pascale Zaraté Editor-in-Chief IJDSST E-mail: zarate@irit.fr Subscriber Info: IGI Global Customer Service 701 E Chocolate Avenue Hershey PA 17033-1240, USA Tel: 717/533-8845 x100 E-mail: cust@igi-global.com IJDSST is currently listed or indexed in: Cabell’s Directory; DBLP; GetCited; Google Scholar; The Index of Information Systems Journal; INSPE; MediaFinder; Norwegian Social Science Data Services (NSD); The Standard Periodical Directory; Ulrich’s Periodicals Directory Information Resources Management Association Research Community! www.irma-international.org Discover thousands of open access research papers, as well as journal articles and book chapters on information resources management concepts and applications. Information Resources Management Association 701 E. Chocolate Avenue, Hershey, PA 17033, USA Tel: 717.533.8845; Fax: 717.533.8661 E-mail: member@irma-international.org www.irma-international.org