IJISCRAM Editorial Board Editors-in-Chief: Murray E. Jennex, San Diego State U., USA Bartel Van de Walle, Tilburg U., The Netherlands Associate Editors: Tom De Groeve, EU Joint Research Center, Italy Dewald van Niekerk, African Center for Disaster Studies, South Africa Murray Turoff, New Jersey Institute of Technology, USA Song Yan, Harbin Engineering U., China IGI Editorial: Heather A. Probst, Director of Journal Publications Jamie M. Wilson, Journal Development Editor Ron Blair, Journal Editorial Assistant Chris Hrobak, Journal Production Lead Gregory Snader, Production Assistant Brittany Metzel, Production Assistant International Editorial Review Board: Walter Ammann, IDRC, Switzerland Einar Bjorgo, UNOSAT, Switzerland Marcos Borges, Federal U. of Rio de Janeiro, Brazil Tung Bui, U. of Hawaii, USA Paul Burghardt, DECIS, The Netherlands José H. Canós Cerdá, Universitat Politècnica de València, Spain Benny Carlé, Nuclear Research Center SCK-CEN, Belgium Paul Currion, humanitarian.info, USA Chamindra de Silva, Lanka Software Foundation, Sri Lanka Paloma Díaz, Universidad Carlos III de Madrid, Spain Julie Dugdale, Université Pierre Mendes France, France Frank Fiedrich, George Washington U., USA Simon French, Manchester Business School, UK Eric Frost, San Diego State U., USA Tim Grant, Netherlands Defense Academy, The Netherlands IGIP Jack Harrald, George Washington U., USA Mark Haselkorn, Washington U., USA Sanjana Hattotuwa, info-share.org, Sri Lanka Roxanne Hiltz, New Jersey Institute of Technology, USA Renato Iannella, National ICT Australia, Australia Jonas Landgren, Viktoria Institute, Sweden Greg Madey, U. of Notre Dame, USA David Mendonca, New Jersey Institute of Technology, USA Leysia Palen, U. of Colorado, USA Nigel Snoad, Microsoft, USA Suha Ulgen, UN OCHA, Switzerland Firoz Verjee, NOAA, USA Peter Wooders, Department of Internal Affairs, New Zealand Sisi Zlatanova, Delft U. of Technology, The Netherlands IGI PublIshInG www.igi-global.com CALL FOR ARTICLES International Journal of Information Systems for Crisis Response and Management An oficial publication of the Conference on Information Systems for Crisis Response and Management The International Journal of Information Systems for Crisis Response and Management (IJISCRAM), is an academically rigorous outlet for research into crisis response and management. It is focused on the design, development, implementation, use and evaluation of IS technologies and methodologies to support crisis response and management. It explores issues critical to the application of IS to crisis response and management. MISSION: The mission of the International Journal of Information Systems for Crisis Response and Management (IJISCRAM) is to provide an outlet for innovative research in the area of information systems for crisis response and management. Research is expected to be rigorous but can utilize any accepted methodology and may be qualitative or quantitative in nature. The journal will provide a comprehensive cross disciplinary forum for advancing the understanding of the organizational, technical, human, and cognitive issues associated with the use of information systems in responding and managing crises of all kinds. COVERAGE/MAJOR TOPICS: Topics include but are not limited to... • • • • • • • • • • • • • • • • • • • Case studies, research methods, and modeling approaches Collaborative and intelligent systems Command and control Communication technologies Crisis planning, training, exercising, and gaming Data fusion, representation, and visualization Decision making and judgment Disaster risk reduction, risk management, ad-hoc, and sensor networks Early warning systems Emergency response systems Geographical information systems Globalization and development issues Healthcare and health information systems Human-computer interaction Humanitarian operations Information systems strategy Knowledge management and systems Systems interoperability information systems infrastructures Virtual teams and organizations ISSN 1937-9390 eISSN 1937-9420 Published quarterly Please send all submissions to: Murray E. Jennex and Bartel Van de Walle IJISCRAM@iscram.org 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/IJISCRAM and click on the "Library Recommendation Form" link along the right margin. InternatIonal Journal of InformatIon SyStemS for CrISIS reSponSe and management July-September 2010, Vol. 2, No. 3 Table of Contents Research Articles 1 Lessons Learned on the Operation of the LoST Protocol for Mobile IP-based Emergency Calls Ana Goulart, Texas A&M University, USA Anna Zacchi, Texas A&M University, USA Bharath Chintapatla, Texas A&M University, USA Walt Magnussen, Texas A&M University, USA 25 Factors that Inluence Crisis Managers and their Decision-Making Ability during Extreme Events Connie White, Jacksonville State University, USA Murray Turof, New Jersey Institute of Technology, USA 36 Achieving Electric Restoration Logistical Eiciencies During Critical Infrastructure Crisis Response: A Knowledge Management Analysis Teresa Durbin, San Diego Gas and Electric, USA Murray E. Jennex, San Diego State University, USA Eric Frost, San Diego State University, USA Robert Judge, San Diego State University, USA 51 Curriculum Design and Development at the Nexus of International Crisis Management and Information Systems Keith Clement, California State University Fresno, USA InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 1 Lessons Learned on the Operation of the LoST Protocol for Mobile IPbased Emergency Calls Ana Goulart, Texas A&M University, USA Anna Zacchi, Texas A&M University, USA Bharath Chintapatla, Texas A&M University, USA Walt Magnussen, Texas A&M University, USA ABSTRACT The technology used in citizen-to-authority emergency calls is based on traditional telephony, that is, circuitswitched systems. However, new standards and protocols are being developed by the Internet Engineering Task Force (IETF) to allow emergency communications over packet switched networks, such as the Internet. This architecture is known as Next Generation-9-1-1 (NG-911). In this paper, the authors present lessons learned from experiments on the IETF standard called Location to Service Translation protocol (LoST). LoST maps the user’s location to the address of the emergency call center that serves that location. After implementing the standards in a test-bed with real-world systems, spatial databases, and communication networks, the authors observed performance issues that users may experience. Based on their observations, the authors propose practical ideas to improve the performance of the NG-911 system and LoST protocol operation for mobile users. Keywords: Location-Based Services, Location to Service Translation (LoST), NG-911, Session Initiation Protocol (SIP), Test-Bed, Voice over IP (VoIP) 1. INTRODUCTION Every minute, tens of thousands of emergency calls are placed in the United States. Thousands more are probably being made in other countries. Within each country or region, there is usually a common number to reach emergency services DOI: 10.4018/jiscrm.2010070101 (e.g., 9-1-1, 1-1-2, etc). In order to better explain the context of this paper, next we provide a brief history of emergency calls, followed by a description of this paper’s motivation. A. A Brief History of Emergency Call Systems It was in 1937, in Great Britain, that the idea of a single emergency number was developed. Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 2 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 By dialing 9-9-9, British citizens could contact the police, fire department, or medical services. Only later, in 1958 US legislators began to consider the use of a single emergency number. The number chosen in the US was 9-1-1. It is known that one of the reasons for this choice is that the numbers were located on opposite sides of the key-pad, which reduced the chances of accidental calls. In 1968, the first 9-1-1 call was made in Alabama. Later that year, the American Telephone and Telegraph Company (AT&T) announced the 9-1-1 service. This first 9-1-1 service is also known as “traditional 9-1-1”. It directed all emergency calls to a nearby Public Safety Answering Point (PSAP) which was directly connected, via dedicated trunks, to the telephone company’s central office (CO). A typical PSAP consists of a group of call takers who are responsible for answering emergency calls as well as dispatching the appropriate emergency service (such as ambulance, or police, or fire fighters). Because the plain old telephone system (POTS) network only established and terminated calls, call takers had few resources to determine the caller’s location. As a result, traditional 9-1-1 soon gave way to landline Enhanced 9-1-1 (E-911), which greatly increased the ability to locate the caller. E-911 employs an address database called Automatic Location Identification (ALI), which uses the telephone number of the caller to determine his/her identity and location. E-911 was implemented in 1978 and is still being used today (Figure 1). Landline E-911, however, only temporarily met the new technological need. Cell phones soon made it necessary to again upgrade the system. The Federal Communications Commission (FCC) mandated that all United States wireless carriers provide the location information of the caller (Reed, Krizman, Woerner, & Rappaport, 1998). Implementation of this mandate would enable wireless carriers to “pinpoint” the location of any 9-1-1 caller. The FCC mandate was divided in two phases: Phase I required the wireless carrier to send the PSAP the location of the antenna or cell site; Phase II required the wireless carrier to send an estimate of the geographic coordinates of the mobile user, typically using some type of network triangulation scheme (Feng & Law, 2002; Sayed, Tarighat, & Khajehnour, 2005). Not all wireless providers have implemented Phase II service as of 2009, based on the National Emergency Number Association (NENA)’s recent 9-1-1 Deployment Reports (NENA, 2010). In the late 1990s, with the onset of voice over Internet Protocol (VoIP) (Goode, 2002) and other Internet services, researchers and developers have realized that a complete overhaul of the current E-911 system is necessary – proposing a Next Generation 9-1-1 (NG-911) (Hixson, Cobb, & Halley, 2007; NENA, 2007a, 2007b; Schulzrinne & Arabshian, 2002) system that Figure 1. Evolution of North American 9-1-1 emergency calling system Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 3 enables both voice and data communications for emergency calls. Recently, an NG-911 proof-of-concept was developed (Kim, Song, & Schulzrinne, 2006; Song et al., 2008). This development represents a critical step and an important building block to validate IP-based emergency calling. The proof-of-concept allowed the testing of IPbased emergency calls to IP-based PSAPs on five main scenarios: (i) calls from traditional landline phones, (ii) calls from cellular phones, (iii) text messages (SMS), (iv) telematics service messages, and (v) multimedia calls from fixed/ enterprise VoIP devices. In addition, the NG-911 proof-of-concept tested the idea of Emergency Services Router Proxies (ESRPs) for policy-based routing. ESRPs improve the call routing capability of 9-1-1 calls: they can re-route emergency calls to any PSAP (e.g., PSAPs that are not overloaded, or PSAPs with special capabilities such as the ones with sign-language capable call takers). With IP-based emergency calling, a PSAP can now be anywhere, even in a mobile device such as a laptop. The NG-9-1-1 proof-of-concept and all the technology behind it mean that a PSAP does not have to be a centralized call center as before, but can be a much more flexible, portable, and less expensive system. The NG-911 proof-of-concept concludes our brief history of the 9-1-1 evolution. However, IP-based emergency calling still faces new challenges. Standards are being developed, security vulnerabilities must be addressed, and the transition from PSTN-based PSAPs to IPbased PSAPs is being investigated. B. Motivations for This Paper One important NG-911 research challenge is related to the location of the caller, which can be very dynamic. Enterprise VoIP phones have the flexibility to be connected to any network port. Home VoIP subscribers can be connected to any access network (e.g., an Internet service provider). Also, mobile devices such as dualmode cell phones (with 802.11 radio), laptops with broadband wireless coverage (using 3G, or 4G cellular data cards) can be constantly changing its location. Moreover, the architecture for IP-based emergency services will have different entities: the access infrastructure network and the application service provider (such as a voice service provider (VSP)) (Tschofenig, Schulzrinne, Shanmugan, & Newton, 2007). This poses unique challenges to current IP-based emergency systems. For instance, it is the subscriber’s responsibility to provide his/her location to the VoIP service provider; however, the subscriber may not update this information when the VoIP device is moved to a different location (i.e., a different access network). As a result, VoIP service providers may not have the most up to date civic or geographic address of the user. New Internet Engineering Task Force (IETF) standards have been created to allow the addition of location information (Peterson, 2005; Polk & Rosen, 2009) to the Session Initiation Protocol (SIP) call setup messages (known as SIP Invites) (Rosenberg et al., 2002). In addition, the Location to Service Translation (LoST) protocol (Hardie, Newton, Schulrzrinne, & Tschofenig, 2008; Schulzrinne, Tschofenig, Hardie, & Newton, 2007; Schulzrinne, 2008) has been created to help both end users or SIP proxies to obtain the SIP address of the PSAP where the emergency call must be routed to. This SIP address is known as SIP Uniform Resource Identifier (URI) (e.g., sip:psap1@domain). Initial experiments using the LoST protocol were performed in (Schulzrinne, Tschofenig, Hardie, & Newton, 2007), which presents the concept of the LoST protocol, introduces the idea of the Emergency Services IP Network (ESInet), and how the protocol queries are performed. The authors conclude the paper with a few performance results, mainly delays of the LoST queries. Additionally, the use of emergency services by mobile users is described by an IETF standard on best current practices (Rosen & Polk, 2009; Schulzrinne & Marshall, 2008). It provides a set of guidelines on obtaining the location of Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 4 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 mobile users, querying LoST servers, and storing the PSAP URI at a cache at the SIP/LoST client (Wu & Schulzrinne, 2004). In Chintapatla, Goulart, and Magnussen (2010), we presented preliminary results from an initial work with the NG-911 testbed, where preliminary performance results on the LoST queries and processing delay at the mobile client were presented. Now, we present in this paper more detailed experiments, lessons learned, and a complete overview of the operation and performance of the LoST protocol, including description of techniques in computational geometry for processing location information and service boundaries data (both at the LoST client and server). Contributions This paper provides a detailed tutorial on the routing of IP-based emergency calls; moreover, it presents experimental results that provide new insights on the operation and performance of the LoST protocol for mobile users. Using realworld service boundaries data, we performed a large set of test-bed experiments; the results helped us identify new algorithms to more efficiently support the caching mechanism (Rosen & Polk, 2009) of the LoST protocol for mobile users (i.e., clients with dynamic location information). This work is valuable to researchers and practitioners working not only on the transition of traditional emergency communications to IP-based emergency communications but also on applications that use location-based services for mobile users. This paper is organized as follows. Section 2 provides an overview of the operation of the LoST protocol. Then, Section 3 discusses the issue of detecting service boundary crossings at the mobile LoST client, which can be addressed by algorithms of computational geometry. Section 4 also discusses service boundaries, but from the viewpoint of the LoST server. Section 5 describes our testbed setup, followed by the experimental results which are presented in Section 6. Finally, this paper is concluded in Section 7. 2. A NEW ARCHITECTURE FOR ROUTING IP-BASED EMERGENCY CALLS In an emergency call, location information, or a reference to a server that has the location information, must be added to the body of the SIP Invite message. Location information can be obtained by the user or a SIP proxy (Barnes, 2008; NENA, 2006; Polk, Schnizlein, & Linsner, 2004; TIA, 2006). As in traditional E-9-1-1 services, location is needed for routing the call to the appropriate PSAP. In addition, location is needed for dispatching the necessary services. There are different accuracy needs between routing and dispatching. For routing, typically an approximate location will suffice, unless the mobile user is close to the border of a jurisdictional boundary which could lead to routing the call to the wrong PSAP. However, for dispatching, emergency responders really need an accurate location in order to quickly reach the user. Concerning the routing of IP-based emergency calls, the LoST protocol is the new protocol specified for call routing in the NG-911 architecture. It maps the caller’s location to the PSAP’s Uniform Resource Identifier (URI). In other words, it resolves a user’s location to the address of a specific PSAP (in the IP-based architecture, the address is an IP address). Or it could be an IP address of an Emergency Services Routing Proxy (ESRP). In case there is neither location nor mapping information in a SIP Invite, the call is routed to a default PSAP. A. Call Flow Figure 2 illustrates the basic operation of the LoST protocol in our NG-911 test-bed. The Geocoder machine shown in the box includes the SIP proxy, LoST server, and Domain Name System (DNS) server. As an example, here are the main steps for an emergency call going through the ESINet based on our current test-bed: Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 5 Figure 2. NG-911 testbed 1. 2. 3. 4. 5. A caller from College Station in Brazos County makes a LoST lookup on the statewide LoST server, with the result being that the caller should instead route to Brazos’ ESRP. The caller calls his/her local SIP proxy with the “Route” field of the SIP Invite being set to esrp-brazos.tamu.edu. The ESRP makes a LoST lookup on the Brazos County LoST server, with the result that the caller should be connected to College Station’s police department at psap-collegestation.tamu.edu. The call is routed to the PSAP server at College Station. The server at College Station’s police PSAP distributes the call among its call takers. It is expected that the end user makes a LoST query when it gets its location information. In this way, the PSAP address is stored in a cache, and updated when the location changes. When the need for an emergency call arises, a new LoST query can be made, or if a new LoST cannot be made or is not successful, the stored results can be used. Then, if an emergency arise, the new (or cached) PSAP URI is added to the header of the SIP message, while the actual location information is added to the body of the SIP message in an XML format known as Presence Information Data Format-Location Object (PIDF-LO) (Peterson, 2005). A practical hierarchy of LoST servers is illustrated in Figure 3, where different levels of information can be implemented at LoST databases at the city, council of governance Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 6 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 (COG), state, country, and even global levels (Schulzrinne, 2008). B. Service Boundaries Crossings Now consider mobile devices such as cell phones moving along a highway at high speeds. In this case, the LoST client would need to make a new LoST query every time its location changes based on its GPS updates, for instance. But according to the IETF’s draft on best current practices for IP-based emergency calls (Rosen & Polk, 2009), “mapping should be performed at boot time and whenever location changes beyond the service boundary”. This requires the service boundary to be sent to the mobile because the mobile device needs to decide whether its new location is inside or outside the current boundary. To allow the client to check service boundary crossings, the LoST protocol specifies a getServiceBoundary query. With this query, the client can get the coordinates of the service boundary, which is a polygon. Both findService and getServiceBoundary queries are illustrated in Figure 4. Both transactions can happen between the LoST client and LoST server several times prior to an emergency call, without the user even knowing about them. The findService query returns a key that identifies a certain polygon. If the polygon coordinates are saved in the client’s cache, then there is no need to get the coordinates of the polygon. This typically happens when the mobile user returns to a service boundary that he/she has visited before. But if the mobile user has never visited that service boundary, the key will not be in the cache. An additional LoST query needs to be made (getServiceBoundary query) to download the service boundary information. Thus, a new boundary will be saved at the mobile user. To illustrate the findService query, in Figure 5 we show a sample of a LoST findService query sent to our server, using service boundary as reference. The current user location is sent as geographic coordinate (in this example 30.627977 and -96.3344068). The response to this query provides the PSAP address information (as a URI) and the key to the service boundary for that PSAP. Note that the LoST messages are in Extensible Markup Language (XML) format, carried in HyperText Transfer Protocol (HTTP) and HTTP secure (HTTPS) protocol exchanges. In summary, each PSAP has a certain area or region it would serve. The area each PSAP serves is stored as corners of a polygon. Using Figure 3. Potential hierarchy of LoST databases Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 7 Figure 4. LoST queries: findService gives the PSAP USI and getServiceBoundary gives the service boundary Figure 5. A sample of a findService query (XML format) these points, we can determine which PSAP would serve a particular point. Thus, whenever the client makes a LoST query, along with the PSAP address it also gets the boundary information of the service are that the PSAP is serving. The rationale for doing the boundary checks at the mobile is that there may be thousands of mobile devices for only one LoST server. It reduces the load at the LoST server which otherwise would be overloaded with too many findService queries. As a drawback of the schemes that we have just described is the requirement for additional calculations at the client, which is the mobile users. This issue is addressed in the Section 3, where we describe in more detail techniques that can be implemented in the LoST client to determine if a location is inside the service boundary. This is usually called point inclusion problem. 3. CLIENT-BASED APPROACHES TO DETECT BOUNDARY CROSSINGS The calculations done at the client to determine if it has crossed the boundary can be solved using well-established point-inclusion techniques which are based on the Jordan Curve theorem (Preparata & Shamos, 1985; Shimrat, 1962). The main one is known as Ray-tracing algorithm. We also propose a new technique that uses the concept of pre-processing to reduce the calculation time. Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 8 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 A. Ray-Tracing Method A simple way to describe the Ray-Tracing method is that a horizontal line (i.e., a “ray”) is traced through the point which is our current location. Then we must check if each and every edge of the service boundary intersects this horizontal line. Depending on the number of intersections, we can determine if the point is inside or outside the service boundary. Let us consider the point of our location as z and the service boundary as a polygon P, as shown in Figure 6. For checking whether the edge intersects the horizontal line through z, we test if one vertex of the edge is above the horizontal line and the other vertex below it. Once we find such an edge, then using elementary geometry we find the x-coordinate of the point of intersection. Comparing the x-coordinates of the points of intersection, we can determine the number of intersections which are left of z and right of z. Based on the Jordan theorem, the number of points of intersection is always an even number, except in the case the horizontal line crosses a vertex of the polygon. If the number of points of intersection towards its left and right is an odd number, then that point is inside the polygon. It the number of intersections is even, the point is outside the polygon. Because the Ray-Tracing method’s implementation requires a loop over all the points in the polygon every time the location changes, the processing time whether a point z is internal to a simple N-gon polygon can be determined in O(N) time (Preparata & Shamos, 1985). Although the Ray-Tracing method is very efficient for checking if the point is inside the polygon for the first time, it is not efficient when we have a dynamic point. In other words, if repetitive calculations are needed, it does not use the information from previous checks to make a decision about the current check. This motivated us to investigate a new method which uses information from previous checks to make a decision. Note that there are other preprocessing calculations that are presented in the literature. For instance, in Preparata and Shamos (1985) a pre-processing approach requires the polygon to be divided into wedges that are then used in the point-inclusion problem. However, when discussing the practical implementation of that approach, we concluded that some of the pre-processing would require calculations at the LoST server. Our approach that uses preprocessing is done at the client side, which is described next. B. Minimum Distance Method In this subsection, a new method is proposed – we call it the Minimum Distance method. It is used with the Ray-Tracing method but it takes advantage of previous Ray-Tracing calculations The Minimum Distance method aims to determine if a point z is inside a polygon P. The first time there is a location change, the Ray-Tracing method is used to determine if the mobile client Figure 6. Point inclusion detection using the Ray-tracing method Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 9 crossed the boundary of the current polygon P. Using the points of intersections found using the Ray-Tracing method, we select the closest intersection p. Then using simple trigonometry, we calculate the distance between p and z, which is the minimum distance between the current location and the boundary of the polygon dmin. This can be illustrated as the process of drawing a circle around the current location with a radius equal to the minimum distance, and the point z at the center (Figure 7). When the mobile changes its location, the algorithm checks if the mobile has moved outside the polygon P by comparing the distance moved di with the radius of the circle dmin. As long as the total distance moved by the mobile client is less than dmin, we do not need to do any other calculation. Once the client moves a distance greater than the radius we need to do a new Ray-Tracing-based check to confirm if the client moved outside of the polygon. In summary, the advantage of the Minimum Distance method is that once the minimum distance is calculated it takes less processing for future calculations. However, compared to the Ray-Tracing method, the minimum distance method requires additional calculations to find the initial radius (i.e., minimum distance). Also, it may give false positives, when the mobile moves in different directions within the minimum distance circle, but the total distance moved becomes higher than the minimum distance, the process starts all over again: the minimum distance value will need to be updated. 4. SERVER-BASED APPROACH TO SIMPLIFY THE SERVICE BOUNDARY Continuing on the issue of determining service boundary crossing, we also researched potential improvements at the LoST server’s side. The new way refers to one sentence of the LoST standard (Hardie, Newton, Schulzrinne & Tschofenig, 2008) that states that a hint of the polygon may be sent as a response to getServiceBoundary. Therefore, we investigated changes in the LoST server to allow the use of simplified polygons. This server-based approach was motivated by two things: (i) to reduce the amount of information sent to the client, and (ii) to make the point-in-polygon calculations faster. A simplified polygon is an abbreviated version of a polygon. It has less number of corners compared to the original polygon. The example shown in Figure 8 shows two polygons: an external and an internal polygon. (We say that this polygon has a “hole”. Surprisingly, in our LoST database we observed several cases of polygons with holes in it, making the point inclusion determination even more complex.) In the simplified version of a polygon in Figure 8, the original corners of the inner and outer polygon are shown in darker color, while the simplified version has corners shown in a lighter color. We can see the reduced number of points in the lighter color. Figure 7. Point inclusion detection using the Minimum Distance method Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 10 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 Using the simplified polygon approach, our LoST server maintained two polygons for every service boundary: the original polygon is used by the server to map location to PSAP in findService query; the simplified polygon is sent as a response to getServiceBoundary. The algorithm we used is the Ramer-Douglas-Peucker algorithm (Douglas & Peucker, 1973). Given two initial points in the polygon, new points are added and a line segment is drawn through them. Then, we measure the distance between all other points and the line segment. The point that is farthest from the line segment is selected first. If the distance from this point to the line is greater than some constant (which is the accuracy required, e.g., 100m) we include the point in the polygon. Then recursively this is done with these two new line segments. In other words, all the points of the simplified polygon are actual points of the service boundary, but in a reduced amount (i.e., points in between are omitted). Additionally, the calculations of the simplified polygon can be done off-line and installed in the LoST server prior to its operation. Therefore, this method makes changes at the LoST server’s side. To ensure compatibility with the LoST protocol, servers that support the simplified polygon should not require major changes in the LoST queries. After presenting our testbed and results in the next sections, we will further discuss this topic of implementation. 5. TESTBED SETUP – LOST CLIENTS AND SERVER The NG-911 testbed and call flow has been introduced in Section 2 (Figure 2). We now provide additional details on the types of LoST clients, and the spatial database at the LoST server. A. LoST Clients We performed experiments with two different mobile clients: i. A laptop that implements the functions of a SIP client and a LoST client: we used the multi-agent SIP/LoST client implemented at Columbia University (Wu & Schulzrinne, 2004). To communicate with the servers in the test-bed, the laptop uses a wireless 3G data card (Sprint Sierra Wireless Aircard 597A). Embedded in the 3G data card, a GPS receiver obtains Figure 8. Simplified polygon of Brazos County (lighter dots) Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 11 the end-user’s location. The GPS location is then sent to a serial port. The SIP client continuously reads the GPS data from the serial port and compares if the location has changed. If it has changed, the LoST client verifies if this location is within the same service boundary (or jurisdictional boundary). ii. A cell-phone: In addition to the SIP client running on a laptop, we have implemented the functionality of the LoST protocol on a smart-phone (Figure 9). The cell phone’s LoST client was implemented in JavaME. It runs on the BlackBerry 8820 with the Blackberry 4.5 OS. This LoST client also has a GPS and can use the latitude and longitude readings to perform the LoST queries. To parse the results of the LoST queries we used the kxml parse (http:// kxml.sourceforge.net/). The kxml parser is a pull parser: it reads a little bit of a document at each time, while still downloading the data. The application drives the parser through the document by repeatedly requesting the next piece. The LoST client verifies the new GPS reading in the same way as the client on the laptop, by checking for every new location if it is within the current service boundary. Because the experiments required long trips to physically move the clients to a different service boundary, we have enabled both the LoST client at the laptop and at the smart-phone to read simulated GPS data so that we could test points in several boundaries. We have also tested these points at different cellular networks (e.g., rural and urban areas) (Zhan & Goulart, 2009). B. LoST Server/Database The LoST server was implemented using PostgreSQL with PostGIS for the mapping service. The computer running this service has a dual core processor at 1800 MHz, with 1GByte of memory. Columbia University provided the software implementation of the LoST protocol (Song et al., 2008), and we have populated the database with our local data. The data contained in our database is at the state level (as in Figure 3) with actual data from the service boundaries in our state (i.e.,Texas). There are around 500 service boundaries in Texas. Out of this large number. we have sampled 198 boundaries to verify their complexity in terms of number of corners. To clarify the concept of “number of corners”, the service boundaries consist of data files that contains the geographic coordinates (i.e., latitude and longitude pairs) of each corner. In this way, each data file describes the corners of a polygon. Figure 10 shows a histogram of our sampled data, in terms of number of corners. The first bar shows that there are 74 service boundaries with less than 1000 corners. Out of 198 samples, this accounts for 37 percent. These low complexity boundaries typically have a regular shape (e.g., rectangular-like shape). However, Figure 10 also shows that there are several boundaries with a few thousands of corners. It is of our special interest to verify the performance of the LoST protocol for such complex service boundaries, especially boundaries with more than 5000 corners, which accounted for 29 service boundaries in our sample, or 15 percent. We have observed that the complexity of the boundary depends on geography (borders with neighboring states, rivers, mountains, coastal lines) and population. On the latter, it is significant to mention the size metropolitan areas contribute to the complexity of the service boundary. For example, the four counties with largest population in Texas have service boundaries with more than 5000 corners. Note that the maximum number of corners that we observed in our database was 19278 corners. 6. EXPERIMENTAL RESULTS This section presents experimental results that illustrate the operation of the LoST protocol in a NG-911 test-bed (Figure 2) The operation of Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 12 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 Figure 9. LoST client implemented in a smart-phone, returning PSAP URI, boundary key, and delay to get response the SIP and LoST protocol together with the computational geometry algorithms of Section 4 are tested in terms of message sizes, service boundaries’ complexity, delays for findService and getServiceBoundary queries, and processing delays for the point inclusion problem. A. SIP Call Flow – Message Size and Delay at the Access Network The SIP call setup transaction in Figure 11 follows a standard SIP call flow, with mobile’s location information added in the initial SIP Invite message. The two intermediate servers are the SIP proxy (geocoder) which is at the side of the SIP client’s domain, and the emergency services routing proxy (ESRP) which is at the side of the PSAP, in the Emergency Services IP network (ESInet). Note that all SIP messages in an emergency call are transported over Transport Control Protocol (TCP), which provides reliable transport to IP packets. From our example (Figure 11), the number of bytes exchanged between the SIP client and SIP proxy to complete the call setup transaction was 4744 bytes. The largest message is the initial SIP Invite message: 2230 bytes. This large size is due to the location information which is in the body of the SIP Invite message, in an Extensible Markup Language (XML) format known as Presence Information Data Format – Location Object (PIDF-LO). In our experiments we are observing that the SIP Invite messages are being fragmented (or divided) in two different packets, whereas regular nonemergency SIP Invites are usually smaller than 1000 bytes and are not fragmented. This was an important finding, and raised some questions on the issue of one of the fragments of the SIP Invite being lost. TCP will take care of retrans- Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 13 Figure 10. Distribution of the number of corners from a sample of in 198 service boundaries in the state of Texas mitting the message, but the added delay to retransmit the complete SIP Invite to the PSAP can be very critical. We also measured the average time to connect the emergency call to a PSAP using the 3G data card: it took an average of 1.5 sec using the laptop. There were no other calls to the PSAP. We concluded that the 3G access network adds a considerable delay in the SIP call setup process. For instance, we measured the round trip times (RTT) for wireless connections such as IEEE 802.11g (54 Mbps) and a 3G card for packet sizes from 128 to 1024 bytes (Table 1). From the call flow shown in Figure 4, we can observe that the wireless network we use will affect only those packets which are sent or received by SIP client. However, the wireless access network is the major source of delay in this lightly loaded scenario. B. SIP LoST “findService” Query Delays In this experiment, we measured the delay from the time the LoST client sends a findService query to the time it gets and parses the findService response, which contains the PSAP URI and the key to the service boundary. This is the first two messages shown in Figure 11. We have used as the LoST client the smart-phone, connected to a commercial 3G network. As mentioned in the previous sub-section, the delay at the 3G access network is relevant, and we will see in the next results that this delay depends on the mobile’s location. The data used in the tests correspond to simulated GPS data representing points in five different service boundaries: Stonewall, College Station, Harris, Fort Worth, and Brazos jurisdictional boundaries. These boundaries are of varying complexity, with a minimum number of corners of 322 and a maximum of 10718. For each test, all the five coordinates were sent to the LoST server and the corresponding service PSAP URI was sent back to the client. For each coordinate, three measurements were taken. Each test was done at different places (rural, urban), adding to a total of 22 different locations. The results in Figure 12 show the Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 14 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 Figure 11. Emegency call flow (SIP and LoST transations), with message sizes and fragmentation average delay of findService queries. The data points are grouped by the place where they were collected, where Figure 12 shows the results in 10 locations. From this graph, we conclude that the location of the mobile user was an important factor in the performance of the LoST client to receive a response from the server. Location 7, for instance. had the worst delays for most of the boundaries. Additionally, Table 2 shows the average delay of all our measurements, in all locations with the correspondent standard deviation. Note that for the service boundary with the smallest number of corners (323 corners) the delays were smaller than all the others, about 200 msec less. This was interesting because we expected that the findService delays by reference were independent of the size of the service boundary. The size in bytes of the queries and responses are about the same (around 600 bytes for the findService query and 800 bytes for the findService response, as shown in Figure 11) no matter which service boundary. Therefore, the network delays are approximately the same, but we suspect that there is an additional processing delay component at the LoST server when it accesses the database. To verify if there are differences in the processing delays when a server receives a findService query, we have used a packet analyzer (e.g., Wireshark) at the server to to measure the time between the arrival of a findSer- Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 15 Table 1. Wireless access networks delay comparison Packet Size 802.11g - RTT 3G card – RTT 128 bytes 6 ms 156 ms 256 bytes 11 ms 160 ms 512 bytes 10 ms 216 ms 1024 bytes 25 ms 270 ms Figure 12. FindService query delays for a mobile user at different locations, simulating points from 5 different service boundaries vice query and the time the server issues the response. Figure 13 shows the average results for four different service boundaries, ranging from 323 to 13,528 corners. A total of 10 measurements were taken for each boundary. For instance, a 10,700-corner-boundary required approximately 0.5 seconds of processing time at the server. Although the delays at the server had a large variance for a couple of boundary sizes (the ones with 2,045 and 13,528 corners), we conclude that the processing delay increases with the complexity of the boundary, i.e., complexity in terms of number of corners and number of holes in the boundary (see third column in Table 2). The high processing delay was unexpected, and this could be due to the limited memory size of our server (1GByte), but further experiments are needed to confirm that. In short, the findService delays had two main components: the network delay and the processing time at the server. Our results showed that the network delays are very dependent on the physical location of the mobile user (e.g., near a cell tower or the number of users in the cell site), whereas the processing delay at the servers depend on the service boundary’s complexity. Also, the difference in processing delays was clearly observed for the polygons with a very small number of corners (such as Stonewall with 323 corners) than the ones with thousands of corners. Next, we will see how the service boundary complexity affects the getServiceBoundary query. Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 16 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 Table 2. FindServiceResponse Delays Service boundary name Number of corners Number of holes in the boundary Mean delay for findServiceResponse (ms) Standard deviation for findservice (ms) Stonewall 323 0 1946.5 244.0 College Station 2045 0 2151.9 602.7 Harris 5343 17 2291.3 985.1 Fort Worth 10606 7 2160.3 611.4 Brazos 10719 1 2181.0 759.6 C. LoST getServiceBoundary Query Delays boundary. This processing delay is discussed next. When a getServiceBoundary is issued (normally after the findService transaction), the server sends the mobile all the coordinates of the polygon (i.e., all the corners). For instance, Figure 11 shows an example in which the size of the getServiceBoundaryResponse message was 250Kbytes, which resulted in 199 packets sent from the LoST server to the client. To have a better idea of the delays for a mobile user to obtain the service boundaries, we tested several service boundary downloads using the LoST client at the smart-phone (Figure 13). Large delays, on the order of 1 minute are observed for the service boundaries with more than 10000 corners. There is a clear dependency on the number of corners and the delay to obtain the service boundary, since the response messages contain all the corners of the polygon. In a lesser degree, the delays depend on the location of the user and the type of network access, as discussed previously. The delays also depend on the type of mobile device (e.g., smart-phones had larger delays than laptops). Although boundary downloads occur only when a user changes to a new service boundary, it is clear that such delays are very large. Users can definitely perceive the delay for a LoST client to download complex service boundaries, i.e., with more than 5000 corners (Figure 14). Once the boundary has been downloaded and stored in cache, the mobile will use this data to decide if a new location is within the D. Processing Time to Detect Service Boundary Crossings In this experiment we measured the time the client spends checking whether a given point is inside the polygon. We compare the processing using the Ray-Tracing method and the Minimum Distance method (Section 3). The tests were performed using polygon of different complexities. The simplest polygon has 323 corners, while the most complex has 14,389 corners. To be able to simulate these different service boundaries, we simulated GPS data as inputs to the LoST client, with a new location update every 5 seconds. In this experiment we used the combined SIP/LoST client installed in a laptop. The results are shown in Figure 15. The x-axis represents the different service boundaries and their number of corners. The y-axis shows the processing delay that each method (Ray-Tracing or Minimum Distance) needed to check if the new location is inside or outside the boundary. To measure this processing delay, we used a timer that is triggered every time a program function named “contains” is called. Note that we have two different programs, each with a different version of the “contains” function: one that implements the pure Ray-Tracing method, and one that implements the Minimum Distance Method. At the end of the “contains” function, the timer stops. Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 17 Figure 13. Processing delay at the LoST server before a findServiceResponse The results show that as the number of corners of the polygon increases, the time taken to check whether a given point is inside the polygon increases proportionally with the number of corners in the Ray-Tracing method (O(n) where n is the number of corners). This does not happen with the Minimum Distance method, which had small delays that varied independently of the complexity of the polygons. The difference in delays between RayTracing and Minimum Distance methods is significant for the last three polygons in Figure 15, which have above 5000 corners. However, with the Minimum Distance method we have an overhead of calculating the minimum distance value for the first time. In the worst case scenario, i.e., for polygons with more than 10000 corners, we measured the overhead as approximately 400ms. Given this overhead, the Minimum Distance method would be advantageous in certain scenarios: • • • When the service boundaries has a large number of corners (5000 or more), When the mobile is far from the boundary (i.e., the minimum distance will be large). When the mobile is stationary or travels at low speed, so the traveled distance would remain smaller than the minimum distance for a long period of time. Thus, new calculations of the Minimum Distance radius and its implicit overhead would not be needed very often. E. Experiments with Simplified Polygons The previous experiments have shown that small polygons such as the Stonewal service boundary with 323 corners have smaller findService delays, getServiceBoundary delays, and smaller processing delay to verify if a point is inside Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 18 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 Figure 14. GetserviceBoundary Response delay from server for a mobile LoST client (smart-phone) its boundary than all the other larger polygons. What if all the polygons were as simple as this one? In this section, we show that by using the algorithm described in Section 4 to simplify polygons can considerably reduce the complexity of the polygons. For instance, using the Ramer-DouglasPeucker, with two different accuracy values (0.1 miles and 0.03 miles), Figure 16 shows the percentage reduction in the number of corners of the simplified polygon for 75 service boundaries within the state of Texas. All polygons that were simplified had more than 2000 corners in its original version. Another way to illustrate this dramatic change in the size of the polygons, the average size of all the original polygons was 4998 corners. When we applied the simplification algorithm with accuracy of 0.1 miles, the average size of all polygons became 336 corners. With 0.03 miles, the reduction is less but still considerable: the average size of all polygons was 748 corners. We observed that the number of points in the simplified polygon does not directly depend on the number of points in the original polygon; however, they depend on the shape of the original polygon, as shown in the simplified polygon of Figure 17, which illustrates the boundary at a coastal area. And a polygon with many curves will have more points in its simplified version. But in all the cases the decrease is very significant. For an accuracy of 0.1 mile and for the polygons shown in Table 3, the reduction was in average 92 percent. Implementing a Database with Simplified Polygons Once we have the simplified polygons, we updated our LoST database to add a new column named “simplified polygon”. This column will Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 19 Figure 15. Client processing time to detect if point is inside service boundary have the simplified version of the polygon for each PSAP boundary. Our LoST server uses this polygon to reply to a GetServiceBoundary query. This decreases the GetServiceBoundary response time significantly. For instance, we performed experiments at the border of a service boundary with 2045 corners. In this case we used the client implemented in the laptop with the 3G data card. As shown in Figure 18 when we queried the LoST server without the simplified polygon (from a location inside the boundary “30.622 -96.34934”) it took an average 4.6 seconds to reply, with the worst case being 8.27 seconds. But when we queried a LoST server with Simplified Polygon from the same location it took only 0.4 seconds on an average with worst case being 0.511 seconds. In our test-bed, we are using the original polygon to respond to the findService query. In other words, the mobile will always get an accurate PSAP URI. But when the client requests the service boundary data, the server sends the simplified polygon along with a warning message. Either the Ray Tracing or the Minimum Distance algorithms may be used to process the point inclusion problem within that simplified boundary. Of course, there will be a loss in the accuracy on the decisions at the mobile. The approach of using different polygons for findService and getServiceBoundary queries is only for testing purposes. In a real system, we recommend that if a small accuracy loss is acceptable (e.g., 0.03 miles), the same service boundary should be used for all queries. The reason is that in very rare situations when our client is near the border of a service boundary, we observed a potential deadlock situation: the mobile’s location is inside the original polygon but outside the simplified version, it would lead to a new findService query all the time, which may lead to a performance problem. 7. CONCLUSION AND FUTURE WORK In this paper we described the new architecture for routing calls using the Location to Service Translation (LoST) protocol. Since the number of mobile broadband users is expanding rapidly, we focused on mobile LoST clients, such as laptops and smart-phones using cellular access networks. After implementing and testing the operation of the LoST protocol in a testbed with real-world service boundary information, we observed that the complexity of the service boundaries (i.e., number of corners and also number of holes) contribute to the delays in the LoST queries (both findService and getServiceBoundary), in addition to the processing delay at the mobile. Considerable delays were observed for boundaries with more than 5000 corners. Therefore, we explored new approaches Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 20 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 Figure 16. Percentage reduction in the number of points in the boundary polygon using RamerDouglas-Peucker Method Figure 17. The complexity and the simplification of a coastal service boundary to improve the performance of the LoST transactions with the caching mechanism: the minimum distance method (at the client side) and the simplified polygons (at the server side). They do not require changes in the standard protocol and the results are promising, although they do have some limitations such as false positives for the minimum distance method, and the performance issue when the mobile is outside the simplified polygon but inside the original polygon. As one can observe, this is an on-going work and we are still researching further improvements to the operation of the LoST transactions and the approaches here discussed. Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 21 Table 3. Reduction in number of points with a 0.1 mile error Polygon Name Number of Points in Original Number of Points in Simplified % Reduction Bryan 10717 379 96.46356 College Station 2045 94 95.40342 Harris 14388 2007 86.05088 Stonewall 322 31 90.37267 Anderson 4979 423 91.50432 Houston 9355 1334 85.74025 Fortbend 9680 653 93.25413 Forworth 10620 831 92.17514 Dickens 746 8 98.92761 Eastland 450 10 97.77778 Austin 8070 866 89.2689 Ellis 9337 769 91.76395 Figure 18. GetServiceBoundary query delays using simplified and original polygon Moreover, future work includes tests on IETF’s proposed security schemes for the LoST transactions, to assure the confidentiality of both location information and routing information. Also, as fourth generation (4G) cellular technologies are being deployed, we will perform experiments on the performance of the LoST queries and re-evaluate the impact of the complex service boundaries on delays and processing time. ACKNOWLEDGMENT This study was supported by a grant from the National Science Foundation (NSF) award no. 0751118. Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 22 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 REFERENCES Barnes, M. (2008). HTTP Enabled Location Delivery (HELD) (IETF draft-ietf-geopriv-http-locationdelivery-07). Internet Engineering Task Force. IETF. Chintapatla, B., Goulart, A., & Magnussen, W. (2010). Testbed Experiments on the Location to Service Translation (LoST) Protocol for Mobile Users. In Proceedings of the IEEE Consumer Communications and Networking Conference (CCNC). Douglas, D., & Peucker, T. (1973). Algorithms for the reduction of the number of points required to represent a digitized line or its caricature. The Canadian Cartographer, 10(2), 112–122. Peterson, J. (2005). A Presence-based GEOPRIV Location Object Format (IETF RFC 4119). Internet Engineering Task Force. IETF. Polk, J., & Rosen, B. (2009). Location conveyance for the session initiation protocol (IETF draft-ietfsip-location-conveyance-13). Internet Engineering Task Force. IETF. Polk, J., Schnizlein, J., & Linsner, M. (2004). Dynamic host configuration protocol option for coordinate-based location configuration information (IETF RFC 3825). Internet Engineering Task Force. IETF. Preparata, F., & Shamos, M. (1985). Computational Geometry – An Introduction. New York: Springer. Feng, S., & Law, C. (2002). Assisted GPS and its Impact on Navigation in Intelligent Transportation Systems. In Proceedings of the 5th IEEE International Conference on Intelligent Transportation Systems (pp. 926-993). Reed, J., Krizman, K., Woerner, B., & Rappaport, T. (1998). An overview of the challenges and progress in meeting the E-911 requirements for location service. IEEE Communications Magazine, 30–37. doi:10.1109/35.667410 Goode, B. (2002). Voice over Internet Protocol (VoIP). Proceedings of the IEEE, 90(9), 1495–1517. doi:10.1109/JPROC.2002.802005 Rosen, B., & Polk, J. (2009). Best current practices for communication services in support of emergency calling (IETF draft-ietf-ecrit-phonebcp-08). Internet Engineering Task Force. IETF. Hardie, T., Newton, A., Schulzrinne, H., & Tschofenig, H. (2008). LoST: A Location-to-Service Translation Protocol (IETF RFC 5222). Internet Engineering Task Force. IETF. Hixson, R., Cobb, B., & Halley, P. (2007). 9-1-1: The Next Generation. 9-1-1 Magazine, 18-21. Kim, J. Y., Song, W., & Schulzrinne, H. (2006). An Enhanced VoIP Emergency Services Prototype. Paper presented at the 3rd International Information Systems for Crisis Response and Management (ISCRAM) Conference. NENA. (2006). NENA recommended method(s) for location determination to support IP-based emergency services (No. 08-505 v1). Retrieved from http:// www.nena.org/standards/technical/voip/locationdetermination-ip-based-emergency-services NENA. (2007a). NENA Functional and Interface Standards for Next Generation 9-1-1 Version 1.0 (i3) (No. 08-002). Retrieved from http://www.nena. org/standards/technical/voip/functional-interfaceNG911-i3 NENA. (2007b). IP-capable PSAP Minimum Operational Requirements Standard (No. 58-001). Retrieved from http://www.nena.org/standards/ operations/IP-PSAP-minimum-requirements NENA. (2010). NENA 9-1-1 Deployment Reports & Maps. Retrieved from http://nena.ddti.net Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston, A., Peterson, J., & Sparks, R. (2002). SIP: Session Initiation Protocol (IETF RFC 3261). Internet Engineering Task Force. IETF. Sayed, A., Tarighat, A., & Khajehnour, N. (2005). Network-based wireless location. IEEE Signal Processing Magazine, 22(4), 24–40. doi:10.1109/ MSP.2005.1458275 Schulzrinne, H. (2008). Synchronizing Location to Service Translation (LoST) Servers (IETF draftietf-ecrit-lost-sync-00). Internet Engineering Task Force. IETF. Schulzrinne, H., & Arabshian, K. (2002). Providing emergency services in Internet telephony. IEEE Internet Computing, 6(3), 39–47. doi:10.1109/ MIC.2002.1003130 Schulzrinne, H., & Marshall, R. (2008). Requirements for emergency context resolution with Internet technologies (IETF RFC 5012). Internet Engineering Task Force. IETF. Schulzrinne, H., Tschofenig, H., Hardie, T., & Newton, A. (2007). LoST: A Protocol for Mapping Geographic Locations to Public Safety Answering Points. In Proceedings of the IEEE International Performance, Computing, and Communications Conference (IPCCC) (pp. 606-611). Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 23 Shimrat, M. (1962). Algorithm 112: Position of point relative to polygon. Communications of the ACM Archive, 5(8), 434. doi:10.1145/368637.368653 Song, W., et al. (2008). Next Generation 9-1-1 Proof-of-Concept System. Paper presented at SIGCOMM ’08. Telecommunications Industry Association. (2006). Link layer discovery protocol for media endpoint devices (LLDP-MED) (ANSI-TIA-1057). Arlington, VA: Author. Tschofenig, H., Schulzrinne, H., Shanmugan, M., & Newton, A. (2007). Protecting First-Level Responder Resources in an IP-based Emergency Services Architecture. In Proceedings of the IEEE International Performance, Computing, and Communications Conference (IPCCC) (pp. 626-631). Wu, X., & Schulzrinne, H. (2004). SIPc, a multifunction SIP user agent. In Proceedings of the IFIP/IEEE International Conference, Management of Multimedia Networks and Services (MMNS’04) (pp. 269-281). Zhan, W., & Goulart, A. (2009). Statistical Analysis of Broadband Wireless Links in Rural Areas. The Journal of Communication, 4(5), 320–328. Ana E. Goulart received a bachelor’s degree in electrical engineering from the Federal School of Engineering of Itajuba (EFEI), in Brazil. While working in the industry, she received a M. Sc. degree in Information Systems Management from the Pontificial Catholic University of Campinas, in 1997. She moved to the United States in 1997 where she earned a M. Sc. in Computer Engineering at North Carolina State University, Raleigh, NC; followed by a Ph.D. in Electrical and Computer Engineering at Georgia Tech, Atlanta, GA, in 2005. She is currently an Assistant Professor in the Electronics and Telecommunications Engineering Technology program at Texas A&M University, in College Station, TX. Her research interests include wireless communications, protocols for real-time voice and video communications, IP-based emergency communications, networked robotics, and rural telecommunications. Anna Zacchi received a Laurea cum laude in Information Science from the Universita' degli Studi di Milano, Italy, and her M. Sc. in Computer Science from Texas A&M University. She is currently a Ph.D. student in the department of Computer Science and Engineering at Texas A&M University. From 2007 to 2009 she collaborated in the Next Generation 911 (NG-911), project sponsored by US Department of Transportation (DoT); the purpose of the project was to develop a proof-of-concept for NG911. In 2009-10 she participated in the NSF funded research project about NG-911 security. She is now working on a research project in the field of Human Computer Interaction. Bharath Chintapatla is a masters student in Computer Science at Texas A&M University. He received his BTech in Computer Science & Engineering from Indian Institute of Technology, Guwahati in August 2008. His research interests include network protocols, wireless technologies, and network security. Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 24 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),1-24,July-September2010 Walt Magnussen is the Director for the Texas A&M University Telecommunications Office, and the Director for the Texas A&M Internet2 Evaluation Center. As Telecommunications Director, Dr. Magnussen manages all aspects of telecommunications for the third largest university in the United States. As the Director of the ITEC, he leads the Voice-over-Internet Protocol (VoIP) and Internet Protocol Television (IPTV) initiatives for Internet2, a consortium of the 209 leading research universities in the United States. Dr. Magnussen has been the lead for many other projects completed within the ITEC, including lead in the development of the Next Generation 9-1-1 (NG9-1-1) Proof of Concept projects for the United States’ Departments of Commerce and Transportation. Dr. Magnussen acquired his bachelor’s and master’s degrees from the University of Minnesota, and his doctoral degree from Texas A&M University. Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),25-35,July-September2010 25 Factors that Inluence Crisis Managers and their Decision-Making Ability during Extreme Events Connie White, Jacksonville State University, USA Murray Turoff, New Jersey Institute of Technology, USA ABSTRACT This paper reviews crisis literature, identifying factors that most challenge decision makers during extreme events. The objectives are to understand the environment in which the emergency manager is working; isolate factors that hinder the decision maker’s ability to implement optimum solutions; and identify structures that best it the problem type. These objectives are important because extreme events are not well managed. Extreme events are best characterized as wicked problems. Stress, information overload, bias, and uncertainty create an environment that challenges even the best decision makers. Factors must be better understood so that policies, systems, and technologies can be created to better it the needs of the decision maker. The authors discuss ongoing research efforts and describe systems being designed and implemented that provide a variety of web based collaborative tools, as well as solutions to these wicked problems. Keywords: Crisis Management, Crisis Managers, Decision Makers, Emergency Response, Web Based Collaborative Tools INTRODUCTION The objective of this article is to examine the emergency and disaster literature in order to identify the factors that challenge crisis managers over the course of a catastrophic event. Characteristics of an extreme event are identified and found to match the same characteristics that define wicked problems. “An emergency is by definition a unique and unpredictable event, DOI: 10.4018/jiscrm.2010070102 and it is seldom possible, even in retrospect, to assess what the outcome of an emergency response would have been if alternative measures had been followed” (Danielsson & Ohlsson, 1999, p. 92). Responding to the needs of an emergency depends on the severity of the event that has occurred. On one extreme of the scale, there is the routine every day emergency. This includes responding to a car wreck, a heart attack, or to a local flood. These types of problems are structured, occur frequently and protocols exist to Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 26 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),25-35,July-September2010 mitigate the situation. More severe emergencies, like 100 year floods, earthquakes, and tornadoes, occur less frequently and pose a greater challenge to crisis managers. These are defined as ill-structured problems and can eventually be tamed into a series of structured sub-problems that can be managed (Turoff & Hiltz, 1982). The most severe crisis are catastrophic events like the 1974 Super Outbreak where 148 Tornadoes touched down over a two day period across the United States, the 1980 eruption of Mount St. Helens, the 2004 Asian Tsunami and in 2005, Hurricane Katrina. Extreme events occur rarely, have not been experienced before by crisis management and have no clear solutions to problems. Extreme events are best characterized as wicked structures problems. The literature reveals that extreme events do have particular characteristics with common themes reappearing throughout scientific literature. Carl Von Clausewitz offers a cohesive observation of the characteristics plaguing extreme events. He wrote: “A commander must continually face situations involving uncertainties, questionable or incomplete data or several possible alternatives. As the primary decision maker, he, with the assistance of his staff, must not only decide what to do and how to do it, but he must also recognize if and when he must make a decision” (Clausewitz, 1976, p. 383). This research is important because the needs of crisis managers must be identified from the literature found within the emergency domain. Too often, systems are designed, policies are created and plans are written that don’t consider the ground level decision maker. For disasters to be managed so that the solutions are a good fit to the problem, the systems, policies and plans need to be created by asking: ➢ ➢ Who is the decision maker? Under what conditions are these decisions being made? ➢ ➢ What factors pose the greatest threat to making optimal decisions? How can a system be designed to overcome the most challenging factors that hinder decision making? It is important for the results of studies confirming the task type, needs and considerations of the practitioners themselves to be observed so that technology, systems, policy and procedures can be developed to support the needs of decision makers to facilitate a rapid response and recovery given a catastrophic event has occurred. The literature clearly revealed a number of influences that can be managed better if systems are designed to meet those needs. BACKGROUND Stressors Decisions are time critical and can be emotionally stressful when life and death triage type decisions must be made. Crisis managers are confronted with additional stressors that challenge the best decision makers. Simple creature comforts are nonexistent and that affects a human’s decision making ability without taking into account, the numerous other stressors that are ongoing. A lack of sleep, personal issues (family, home), proper nutrition, hygiene, and transportation are characteristics of the environment in which the decisions are being made. After the great Deluge of New Orleans when the levees broke flooding 80% of the city, many first responders’ had lost their own homes, their own family members, had no food, no place to sleep and were every bit as much victims as the people they were to help rescue and protect. Responders reacted in a number of ways; most notorious were a small percentage of the police force who abandoned their roles, and a handful who abused their authority (Kushma, 2007). This contradicted the expected roles they represented. These environmental, mental and physical influences were beyond their control and clearly affected their rational decision mak- Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),25-35,July-September2010 27 ing ability. In the face of adversity, most these men and women reported for duty. However, reports of police brutality resulting in murder charges and convictions are the end result of those responders who could not handle stress on a personal level (Brinkley, 2006). Stress is an understandable emotion felt by emergency managers. They must make life and death decisions especially where such tragedies requiring triage may have to be decided in the selection criterion between groups of people (Kowalski-Trakofler, Vaught, & Scharf, 2003). Another source of stress arises when decisions must be made under severe time constraints (Rodriquez, 1997; Kowalski-Trakofler et al., 2003). Decision makers have to forecast and make predictions given the uncertainty in expectations of future events (Rodriguez, 1997). Time is precious, and accurate decisions must be made along a time line at particular points in time over the duration of the event as a disaster evolves (Brehmer, 1987; Danielsson & Ohlsson, 1999), “The operational commander continually faces an uncertain environment” (Rodriguez, 1997, p. 5). Information: Too Much, Too Little, Too Late Initially, getting enough accurate and timely data is a challenge, but afterwards information overload can create bias conducive to poor decisions being implemented. This is all confounded by the lack of experience managing similar events of the same magnitude by teams and leadership. If a decision is made one way or the other, no one is sure what the outcome will be so there’s a roll of the dice with every alternative. Critical judgments must be made where large amounts of information are available for consideration creating information overload. To make matters worse, this information can be wrong or incomplete (Kowalski-Trakofler et al., 2003) or sufficient time may be lacking to gain the perfect and complete information needed before the decision is made (Rodriquez, 1997). “In dealing with the uncertainty of a continually changing environment, the decision maker must achieve a trade-off between the cost of action and the risk of non-action” (Kowalski-Trakofler & Vaught, 2003, p. 283). Sometimes these decisions are made on the decision maker’s (DM) assumptions and intuition when information is not attainable (Rodriquez, 1997). One person is in charge of making the final decision for action, but this is a collaborative effort of numerous stakeholders sharing numerous overlapping tasks. “As complexity increases, it becomes impossible for a single individual with the limited information processing capacity to gain control” (Danielsson & Ohlsson, 1999, p. 93). A dynamic decision making approach is a much needed method due to the inherent nature of the chaos characteristic of extreme events (Danielsson & Ohlsson 1999). Extreme events need to be managed using structure with flexibility to improvise or adapt where necessary to achieve agility (Harrald, 2009). Lack of Experience Small events occur frequently, and catastrophic events occur rarely (Hyndman and Hyndman, 2006). Protocols or heuristics can be used for the emergencies that are smaller and occur more frequently. However, management is posed with the problem of not having any or little prior experience to larger events where national boundaries are ignored and the demands of the resources needed far exceed the availability of supply. Research reveals that extreme events have different characteristics from smaller disasters (Skertchly & Skertchly, 2001). This calls for a dynamic approach to decision making to fit the task due to the overwhelming nature of these extreme events considered with the limitations of a human’s mental capacity and ability to manage a large set of ongoing problems at any one given time. A major problem exists in a decision maker’s ability to effectively manage all of the ongoing events simultaneously during an extreme event (Danielsson & Ohlsson, 1999; Kerstholt, 1996). In the remainder of this chapter, these areas will be explored further probing deeper Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 28 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),25-35,July-September2010 into the needs of emergency managers. First, how extreme events are different from small emergencies and must be approached as a different task type is covered. Second, extreme events are defined as wicked problems. These characteristics are laid out and mapped against extreme events. Good versus bad characteristics in crisis decision making from the literature are listed in attempts to steer design. Third, types of bias that are specific in emergency situations and decision making are covered. Next, literature findings concerning time, stress and information overload are provided. Methods describing how crisis managers handle information presently are discussed and related to other research concepts already explored in this research effort. Next, research indicating how feedback and expert intuition are used to manage uncertainty is presented. We close by providing some ongoing research endeavors that are being designed to help solve these problems. Extreme Events Large scale extreme events are not like small emergencies. Small emergencies occur regularly where most decisions are rule based due to the experience of the event (Rasmussen, 1983). This is referred to as procedural expertise (Adams & Ericsson, 2000). In the event that a small emergency should occur, the EM may not even be notified because firefighters, police and emergency medical attendants already know how to proceed (Danielsson & Ohlsson, 1999). On the other hand, extreme events present a different set of characteristics due to the problem type and task structure (Campbell, 1999; Mitchell, 1999; McLellan et al., 2003). In large-scale operations, the cognitive demands on the EM are severe (Danielsson & Ohlsson, 1999). Team coordination strategies will evolve from explicit coordination under low workload conditions to implicit coordination as work load increases. Large-scale emergency operations imply distributed decision making in that decisions are disseminated among many stakeholders, of which no single individual has complete knowledge of the current situation (Danielsson & Ohlsson, 1999; Mitchell, 1999; Kowalski-Trakofler & Vaught, 2003). Wicked Problems Extreme events possess characteristics, are problem types and have task structures that are categorized as wicked. Wicked problems are volatile and of a very dynamic nature with considerable uncertainty and ambiguity (Horn, 2005). Wicked problems are ongoing and have no stopping rule (Rittel & Webber, 1973, Digh, 2000). They are never resolved and change over time (Conklin, 1998). Wicked problems are solved per se when they no longer are of interest to the stakeholders, when resources are depleted or when the political agenda changes (Horst & Webber, 1973). Many stakeholders with multiple value conflicts redefine what the problem is repeatedly, reconsider what the causal factors are and have multiple views of how to approach and hopefully deal with the problem (Rittel & Webber, 1973; Conklin, 1998; Digh, 2000). Getting and maintaining agreement amongst the stakeholders is most difficult because each has their own perception and, thus, opinion of what is best (Rittel & Webber, 1973). Extreme events possess the characteristics of those found within the definitions of wicked problems. “Each dysfunctional event has its own unique characteristics, impacts, and legacies” (Skertchly & Skertchly, 2001, p. 23). For example, catastrophic disasters have the following attributes and dimensions many of which are the same as those described in wicked problems: ➢ ➢ ➢ ➢ ➢ ➢ *They don’t have any rules. Often, emergency services are insufficient to cope with the demands given the limited amount of available resources. Vital resources are damaged and nonfunctional. *Procedures for dealing with the situation are inadequate. *No solutions for resolution exist on a short-term basis. *Events continue to escalate. Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),25-35,July-September2010 29 *Serious differences of opinion arise about how things should be managed. ➢ The government of the day and the bureaucracy becomes seriously involved. ➢ The public takes an armchair position and is fed by the media. ➢ *The number of authorities and officials involved are growing. ➢ *Sometimes simply trying to identify which of the emergency services and investigative bodies is doing what results in complete chaos. ➢ The need to know who is in charge is urgent (Campbell, 1999, p. 52). *are characteristic of wicked problems ➢ Crisis managers’ tasks differ from control task types in that, no two events are the same so different decision processes are required to be implemented. Interacting variables are many, and the domain is ill defined and unknown at times (Danielsson & Ohlsson, 1999). An emergency manager cannot project any future decisions with any degree of accuracy due to all of the variables that are involved and all of the different scenarios that can exist due to the great amount of uncertainty involved and lack of experience of the unknown (U.S. Naval War College, 1996). Decision Making in Emergency Management Decision tasks are perceived to be difficult by the EM where issues involving life saving operations such as evacuations or triage have the potential to have devastating results if not conducted accurately (Danielsson & Ohlsson, 1999). Studies show an EMs most difficult aspects of work are: ➢ ➢ Lack of routine and practice–refers to the infrequency of major accidents, making it difficult to get experiences of the command and control proper. Communicational shortcomings. ➢ ➢ ➢ ➢ ➢ Information overload is salient during the initial phase of an emergency response and is seen as especially severe if no staff members are available to perform communication duties. Technical equipment inadequacy. Lack of skills in handling communication equipment. Feelings of isolation. Lack of peers with whom to discuss common problems (Danielsson & Ohlsson 1999, p. 94). Other psychological processes are associated with decisions made by crisis decision makers. Effective decision makers must take many factors of the environment into consideration to understand that these are complex, dynamic, time-pressured, high-stakes, multi-person task environments (McLellan et al., 2003). Some hazard conceptualization and management problems developed from Mitchell (1999) are presented: ➢ ➢ ➢ ➢ ➢ ➢ ➢ ➢ ➢ ➢ ➢ ➢ *Lack of agreement about definition and identification of problems *Lack of awareness of natural and unnatural (human-made) hazards *Lack of future forecasting capabilities *Misperception of misjudgment of risks associated with hazards Deliberate misrepresentation of hazards and risks *Lack of awareness of appropriate responses *Lack of expertise to make use of responses Lack of money or resources to pay for responses *Lack of coordination among institutions and organizations Lack of attention to relationship between ‘disasters’ and ‘development’ Failure to treat hazards as contextual problem whose components require simultaneous attention Lack of access by affected populations to decision making Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 30 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),25-35,July-September2010 Lack of public confidence in scientific knowledge ➢ Lack of capable and enlightened political leadership ➢ *Conflicting goals among populations at risk ➢ *Fluctuating salience of hazards ➢ Public opposition by negatively affected individuals and groups. *wicked characteristics ➢ Many of these are also characteristic of the wicked problem types defined earlier and have characteristics in common with those of extreme events (Rittel & Webber, 1973; Campbell, 1999). Other commonalities between wicked problems and extreme events are that the event is continuous; the operation is twenty-four hours a day. The decision maker has to trust others to make his or her decisions when they are not awake. It is unpredictable what the problems will be, who will be involved or who needs to be involved so sharing of the current situation by all involved is critical. Time Time lost is always a disadvantage that is bound in some way to weaken he who loses it (Clauswitz, 1976, p. 383). Time is a critical factor that further complicates the decision making process. In extreme events, an emergency manager must consider an enormous number of factors quickly (KowalskiTrakofler et al., 2003). Decisions must be made, sometimes forced due to time constraints. “The faster a decision has to be made, the less time the information processing system has to convert or gather enough accurate information to convert assumptions to facts” (Rodriquez, 1997, pp. 7-8). This means that decisions are made under uncertainty and without full consideration. Crisis managers must weigh delaying the decision making against the negative consequences that may occur while waiting for more requested information (Kowalski-Trakofler & Vaught, 2003). Once time has passed, alternative actions are no longer possible and perhaps the best decision has been bypassed leaving only less optimal conditions from which to choose. Kowalski-Trakofler and Vaught conducted a study of good decision making characteristics under life threatening situations. They found that, during any phase of the decision making process, a set of factors could significantly impact one’s ability to deal with complex problems under time critical situations. These factors are: ➢ ➢ ➢ ➢ Psychomotor skills, knowledge and attitude Information quality and completeness Stress–generated both by the problem at hand and any existing background problem The complexity of elements that must be attended (2003, p. 285). One research finding indicates that performance can be maintained under time pressure if the communication changes from explicit to implicit (Serfaty & Entin, 1993). They found that Implicit coordination patterns, anticipatory behavior, and redirection of the team communication strategy are evident under conditions of increased time-pressure. The authors conclude that effective changes in communication patterns may involve updating team members, regularly anticipating the needs of others by offering unrequested information, minimizing interruptions, and articulating plans at a high level in order to allow flexibility in the role of front-line emergency responders (Serfaty & Entin, 1993). Stress Stress is defined as “a process by which certain work demands evoke an appraisal process in which perceived demands exceed resources and result in undesirable physiological, emotional, cognitive and social changes” (Salas, Driskell, & Hughs, 1996, p.6). Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),25-35,July-September2010 31 Studies show that during an emergency, information quality varies on three dimensions: reliability, availability and relevance (Danielsson & Ohlsson, 1999). The decision to use information at any given time and the weight of the usage of the information is based on these dimensions. Information during an emergency can be the source of stress in many ways (KowalskiTrakofler, et al., 2003). First, due to technical malfunctions or just poor implementation, the initial warnings can be ambiguous and create a greater need for clarity in a situation. This causes the situation to be interpreted differently and leads to misinterpretations in how people perceive the information and hence, respond. Stressors due to information mismanagement are formed when people do not fully understand what is going on which creates disagreement between stakeholders; the right information is not gathered. This wastes time and causes more stress and frustration. Poor leadership is a stressor experienced too often. An enormous amount of stress was expressed by the Gulf Coast emergency community during the 2010 British Petroleum Oil crisis. Not being able to protect the wildlife and shorelines were most frustrating to local officials who watched helplessly as the oil decimated the marshlands and white sandy beaches. Poor leadership was exhibited in BP Oil crisis, the President of the United States was weak, the President of BP was weak, both were inexperienced and out of touch. Weak leadership exacerbates a problem, usually delaying resources creating a much worse human disaster due to mismanagement. Last, when technology or other apparatus fails, this leaves people without information and the inability to keep current with response efforts and will add more stress (Kowalski-Trakofler et al., 2003). Stress is a major factor in decision making especially during life critical situations (Kowalski-Trakofler et al., 2003). One of the primary stressors is the lack of information immediately after the event during the early phase of the emergency response where it concerns determining scale and the characteristics of damage (Danielsson & Ohlsson, 1999). A major problem occurs when people are making decisions under stress that leads to poor decision making. Research shows all of the feasible choices are not considered, and a decision is likely to be made prematurely (Keinan, Friedland, & Ben-Porath, 1987). This is not good because no matter how experienced an emergency manager may be, he/she will be confronted with situations they have not experienced previously (Harrald, 2009). Information Overload Good incident commanders function as if they have a good practical understanding of the limitations of their information processing system, and the corresponding limitations of others (McLellan et al., 2003). In particular, they operated in such a manner that (a) their effective working memory capacity was not exceeded, (b) they monitored and regulated their emotions and their arousal level, and (c) they communicated with subordinates in ways that took into account subordinates’ working memory capacity limitations. The foundation of their ability to manage their own information load effectively seems to be prior learning from past experience. Bias Many forms of bias exist when it comes to decision making, but emergency management has a set that is associated with disastrous leadership. Research indicates that this is from a lack of self awareness which is a normal reaction concerning information processing. Table 1 lists the bias types along with a brief description derived by Adams and Ericsson (2000). Muddling Through A large amount of information must be considered in a very small amount of time. Time to Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 32 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),25-35,July-September2010 Table 1. Bias in emergency management decision making Bias Type Description Sunk-costs Persisting with a tactic, which to the dispassionate observer is demonstrably ineffectual, simply because time and resources have already been invested in the tactic. Optimism Choosing a course of action which necessitates nothing whatsoever going wrong if it is to succeed. For example, positioning a crew on steep sloping terrain with high levels of burnable material above and below them. Need for Action Good incident commanders frequently report having to deliberately exercise self-restraint so as not to precipitately commit resources to a course of action before completing a thorough situation assessment Linear Rate of Change Associated with disastrous incident command at wildland fires; human beings seem to be incapable of accurately predicting non-linear rates of change. fully explore all alternatives is lacking not to mention, stress has a tendency to make decision makers focus narrowly on the list of available alternatives. Studies found that good crisis managers only focus on the most feasible and reliable solutions and eliminate the nonessential information (Kowalski-Trakofler et al., 2003). This does not compromise the decision making ability to make good decisions, but rather, simplifies the process allowing them to focus on the critical issues. This same approach was validated by other research studying decision processes of good decision making (McLellan et al., 2003). The study indicated that all of the information was scanned but focus was only considered on a ‘need to know’ basis and only on the relevant factors which needed to be considered. This decision making strategy is described by Charles Lindblom as Muddling Through (Lindblom, 1959, 1979). This employs methods that help a decision maker focus on the most relevant subgroup, given a list of alternatives from which to choose for any given task. Muddling through a problem guides decision makers to direct their focus into selecting incremental changes. The critical aspect of Lindlom is that everyone needs to be involved who is at any level in the organization if there is a problem to be solved which means the open sharing of information from any part of the organization to any other (Turoff, White, & Plotnick, 2009). Feedback Timely and reliable feedback is one means to help crisis managers make good judgments. One type of uncertainty is from the lack of feedback or reported information from the initial assessment from affected areas. Particularly bothersome to emergency management can be in the lack of feedback where the next decision cannot be made without the present information acquired especially when the damage cannot be visualized (Danielsson & Ohlsson, 1999). This can have detrimental effects on the outcome of the event, because the emergency manager’s performance is diminished. Uncertainty The demands on emergency management are described by The Catastrophic Annex to the National Response Plan (NRP) (DHS, 2004): “A detailed and credible common operating picture may not be achievable for 24 to 48 hours (or longer).” “As a result, response activities must begin without the benefit of a detailed or complete situation and critical needs assessment” (Harrald, 2006, p. 258). Due to the nature of an extreme event, many judgments must be made with information that is often ambiguous, wrong and incomplete (Kowalski-Trakofler et al., 2003). The operational activities involve “hierarchical teams of trained individuals, using specialized equipment, whose efforts must be coordinated via command, control, and com- Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),25-35,July-September2010 33 munication processes to achieve specified objectives under conditions of threat, uncertainty, and limited resources, both human and material” (McLellan et al., 2003, p. 2). Not only are the decisions made presently under questionable information, in addition, forecasting future events poses a challenge due to the uncertainty in the future events as they play out over the duration of the extreme event (Rodriguez, 1997). “To make decisions about an uncertain future, the commander must make many assumptions. Intuitive thinking is an important skill in the ability to make a sound assumption” (Rodriguez, 1997, p. 1). Disconnected bits of data exist where gaps in the logic providing useful information are missing. This is where the experts are using intuition to fill gaps in information needs. Expert Intuition Assumptions are used by crisis managers to fill in gaps where uncertainty exists (Rodriguez, 1997). Intuition plays a large role in filling in these gaps and can have good consequences from those with experience. “For experienced commanders, intuition fills in the decision making processes where imperfect information leaves off” (Battle Command, 1994, p. 25). A study conducted on a large group of top executives supports the concept that intuition was used to guide critical decision making situations. The situations and environments in which intuition was mostly used and helpful were found to be where: ➢ ➢ ➢ ➢ ➢ ➢ ➢ A high level of uncertainty exists The event has little previous precedent Variables are often not scientifically predictable “Facts are limited” Facts do not clearly point the way to go Time is limited and the pressure is to be right Several plausible alternative solutions are available to choose from, with good arguments for each (Argot, 1986, p 18). When considering the issue of analytical versus intuition judgment, the National Institute for Occupational Safety and Health (NIOSH) reported: The point here is that research which focuses on judgment must include scrutiny not only of decisions that are made, but also of real-world variables that influence them. The quality of any decision may have little or no direct relationship to the eventual outcome of its execution in a given situation. This is because a decisionmaker is constrained not only by the stress of the situation or personal knowledge and attitudes, but also because he or she can only weigh information that is available (KowalskiTrakofler et al., 2003, p. 286). Normal decision making techniques do not suffice in such complex situations as extreme events. Characteristics were identified as: ➢ ➢ ➢ Novelty—the officer had never encountered such a situation before, Opacity—needed information was not available, Resource inadequacy—the resources currently available were not sufficient to permit an optimal response (McLellan et al., 2003, p. 3). The emergency manager is continually facing an uncertain environment. There is insufficient time for the crisis managers to get the correct information they need and this must be weighed against the need to make a decision at a particular time, so he/she must rely on assumptions and intuition. Intuition fills the gap between disconnected bits of information helping to create a full picture from which problems can be best defined and solution sets available to fit the needs of the emergency manager. SUMMARY Ongoing research efforts are addressing ways to solve these problems. It is important to have dis- Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 34 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),25-35,July-September2010 tributed command and control operations which allow the participation of anyone at any time and any place in both the planning and ongoing operations (Turoff, Chumar, de Walle, & Yao, 2003). Current research explores highly efficient ways for emergent problems to be supported by concerned subgroups addressing complex problems by proposing solution components and comparing via voting their agreement where participating have the ability to supply comments on disagreements (White, Turoff, & Hiltz, 2010). Collaborative systems are being designed to support all phases of emergency management where participants distributed can collaborate on evolving scenarios to support planning (Yao, Turoff, & Chumar, 2009). Clearly, the literature demonstrates that there is a common theme identifying characteristics that influence a crisis manager’s ability to make optimal decisions during and after an extreme event has occurred. The decision maker should be the core focus upon which policies, protocols and systems are designed. Danielsson, M., & Ohlsson, K. (1999). Decision Making in Emergency Management: A Survey Study. International Journal of Cognitive Ergonomics, 3(2), 91–99. doi:10.1207/s15327566ijce0302_2 REFERENCES Kushma, J. (2007). Role Abandonment: Should we leave this myth behind. Natural Hazards Observer, XXXI(5). Adams, R. J., & Ericsson, A. E. (2000). Introduction to the cognitive processes of expert pilots. Journal of Human Performance in Extreme Environments, 5(1), 44–62. Agor, W. H. (1986). The Logic of Intuitive Decision Making. New York: Quorom Books. Battle Command. (1994). Leadership and Decision Making for War and Operations Other than War. Fort Leavenworth, KS: Battle Command Battle Laboratory. Baumgart, L., Bass, E., Philips, B., & Kloesel, K. (2008). Emergency Management Decision-Making During Severe Weather. Weather and Forecasting, 23(6), 1268–1279. doi:10.1175/2008WAF2007092.1 Campbell, R. (1999). Controlling Crisis Chaos. Journal of Emergency Management Australia, 14(3), 51–54. Clausewitz, C. V. (1984). On War. Princeton, NJ: Princeton University Press. Harrald, J. (2006). Agilitly and Discipline: Critical Success Factors for Disaster Response. AAPSS Annals, 604(1), 256–272. Harrald, J. (2009). Achieving Agility in Disaster Management. International Journal of Information Systems and Crisis Management, 1(1). Keinan, G., Friedland, N., & Ben-Porath, Y. (1987). Decision-making under stress: Scanning of alternatives under physical threat. Acta Psychologica, 64, 219–228. doi:10.1016/0001-6918(87)90008-4 Kerstholt, J. (1996). Dynamic Decision Making. Soesterberg, The Netherlands: TNO Human Factors. Kontogiannis, T., & Kossiavelou, Z. (1999). Stress and team performance: principles and challenges for intelligent decision aids, Safety Science, December, Vol.33, Issue 3, pp. 103 -128,. Kowalski-Trakofler, K., & Vaught, T. (2003). Judgment and decision making under stress: an overview for emergency managers. International Journal of Emergency Management, 1(3), 278–289. doi:10.1504/IJEM.2003.003297 McLennan, J., Holgate, A., & Wearing, A. (2003, September). Human Information Processing aspects of Effective Emergency Incident Management Decision Making. Paper presented at the Human Factors of Decision Making in Complex Systems Conference, Dunblane, Scotland. Mitchell, J. K. (1999). Crucibles of Hazard: MegaCities and Disasters in Transition. Tokyo: United Nations University Press. Rodriguez, D. M. (1997). Dominating Time in the Operational Decision Making Process. Newport, RI: U.S. Naval War College. Salas, E., Driskell, E., & Hughs, S. (1996). The study of stress and human performance . In Driskell, J. E., & Salas, E. (Eds.), Stress and Human Performance (pp. 1–45). Mahwah, NJ: Lawrence Erlbaum Associates. Skertchly, A., & Skertcly, K. (2001). Catastrophe management: coping with totally unexpected extreme disasters. The Australian Journal of Emergency Management, 16(1). Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),25-35,July-September2010 35 Turoff, M., Chumer, M., Van de Walle, B., & Yao, X. (2003). The Design of a Dynamic Emergency Response Management Information System (DERMIS). Journal of Information Technology Theory and Application, 5(4). Turoff, M., & Hiltz, S. R. (1982). Computer Support for Group versus Individual Decision Support. IEEE Transactions on Communications, 30(1), 82–91. doi:10.1109/TCOM.1982.1095370 Turoff, M., White, C., & Plotnick, L. (2009). Dynamic Emergency Response Management For Large Scale Extreme Events. International Journal for Information Systems and Crisis Response Management. U.S. Naval War College. (1996). Operational Decision Making. Instructional PPer NWC 4108. Newport, RI: Joint Military Operations Department, U.S. Naval War College. White, C., Hiltz, S. R., & Turoff, M. (2008). United We Respond: One Community, One Voice. Paper presented at the Information Systems for Crisis Response and Management Conference 2008, Washington, DC. White, C., Turoff, M., & de Walle, B. V. (2007). A Dynamic Delphi Process Utilizing a Modified Thurstone Scaling Method: Collaborative Judgment in Emergency Response. Paper presented at the 4th Annual Information Systems on Crisis and Response Management, Delft, The Netherlands. Yao, X., Turoff, M., & Chumar, M. (2009). Designing a Group Support System to Review and Practice Emergency Plans in Virtual Teams. Paper presented at the 6th Annual Information Systems on Crisis and Response Management, Washington, DC. Connie White is an Assistant Professor with the Institute of Emergency Preparedness at Jacksonville State University. Her research interests include designing and implementing web based Delphi systems and leveraging social media to support crisis communications. Dr. White’s background with a list of her publications can be found on her homepage: http://sites.google. com/site/conniemwhite/ Murray Turoff is a Distinguished Professor Emeritus at the New Jersey Institute of Technology. He is a co editor of a recent book on Emergency Management Information Systems (M.E. Sharp 2010). Besides his early and continuing work with the Delphi Method, he spent most of academic research career in the design and evaluation of Computer Mediated Communication systems. After 9/11 he turned his attention back to his early work in Emergency Management and related work in Planning and Foresight and Delphi Design. In 2004 he was a cofounder of the international organization ISCRAM (Information Systems for Crisis Response and Management). Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 36 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),36-50,July-September2010 Achieving Electric Restoration Logistical Eficiencies During Critical Infrastructure Crisis Response: A Knowledge Management Analysis Teresa Durbin, San Diego Gas and Electric, USA Murray E. Jennex, San Diego State University, USA Eric Frost, San Diego State University, USA Robert Judge, San Diego State University, USA ABSTRACT After the 2007 Southern California wildire events, event-assessment of the eficacy of spreadsheets and paper forms raised the question of whether alternative tools could have achieved greater eficiencies in the logistical support of command centers, the sites from which the local utility’s electric restoration personnel were deployed. In this paper, the authors examine what approach would have enabled personnel working on the logistics of the command center effort to have easier-to-use, faster-to-access, command center data stored in, and provided via, a catastrophe resilient platform other than the traditional company computer network. Additionally, the capability to store basic command center requirements from previous emergency responses, thereby saving time during the next emergency, was examined. Keywords: Command Center Logistics, Crisis Response, Crisis Response Logistics, Critical Infrastructure, Emergency Management, Knowledge Management INTRODUCTION Gas and electric services in the County of San Diego are provided by San Diego Gas & Electric Company (SDG&E), an investor-owned public utility. Many useful issues can be analyzed from the wildfires that damaged SDG&E’s electric system in 2003 and 2007 and certain activities that supported efficient service restoration can be examined and used for comparative analysis for a terrorist attack scenario, earthquakes, wildfires, or any other event that might impact service delivery. DOI: 10.4018/jiscrm.2010070103 Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),36-50,July-September2010 37 This paper analyzes the logistics support provided during the wild fires to determine what could be improved. This is an important topic as there is little research literature addressing crisis response logistics. Whybark et al. (2010) discuss disaster relief supply chains but only propose a research agenda. We agree there needs to be research in this area and start with this case study. Logistics/supply chains ensure that responders are able to sustain disaster relief. Their importance is illustrated by the response to the 2009 Haiti earthquake where ships and supplies stood empty or unused off the coast because of an inability to deliver supplies and evacuate injured. Keeping responders in the field is critical for long term disaster relief. Providing relief is critical for affected persons and societies to recover. This paper hopes to start this important research area by exploring alternatives to the spreadsheet approach used in the 2007 wild fires. Finally, the research question for this paper was what alternatives could be used in lieu of paper forms and spreadsheets to achieve more accurate, timely logistics data and solutions? METHODOLOGY This paper is a case study using action research to assess whether future activities by the command center support teams can be influenced to convert from paper forms and spreadsheets to something more real-time, as a new solution for crisis response. The working definition of action research (Zuber-Skerrit & Fletcher, 2007, p. 413) incorporated situations where people reflect and improve (or develop) their own work and their own situations by tightly interlinking their reflection and action; and also making their experience public not only to other participants but also to other persons interested in and concerned about the work and the situation, i.e. their public theories and practices of the work and the situation, and in which the situation is increasingly: data-gathering by participants themselves (or with the help of others) in rela- tion to their own questions; participation (in problem-posing and in answering questions) in decision-making, self-reflection, self-evaluation and self-management by autonomous and responsible persons and groups. This study is action research as the lead author is a team lead of the SDG&E supply chain systems team tasked with assessing logistics performance. Reflection for this study was done using knowledge management, KM as the reflective lens. Jennex (2005) summarized KM definitions to conclude that KM is about capturing knowledge created in an organization and making it available to those who need it to make decisions and improve organizational performance. Jennex (2007) discussed the role of KM in crisis response and included the role of post event evaluation of lessons learned as a way of capturing knowledge generated during an event and ensuring that knowledge is shared and incorporated into crisis response activities. This implies that KM is a good reflective lens for crisis response research. The ability to apply KM to the 2007 command center logistical effort and how it can benefit future emergency responders corresponds to KM in support of crisis response as expanded upon by Jennex and Raman (2009). The underlying KM principle upon which this study is constructed are that experience gained from one emergency can be applied as knowledge when shared with others to improve performance during a subsequent emergency. Additionally, improved performance will help an organization meet its goals or mandated objectives acceptably. Comparing how the basic computer tools used in the 2007 command center set-ups performed against how an alternative process would perform should demonstrate the need to move towards tools that are multi-user, require no consolidation of data, and which can be used with little to no training in the fast-paced environment of an emergency. Improving an existing business process draws justification from the theory of action research as discussed in the Technology Acceptance Model (TAM) of perceived usefulness Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 38 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),36-50,July-September2010 (Venkatesh et al., 2003). TAM addresses the realm of computer technology and validation of the usefulness of employees’ outputs. Davis (1989) verifies the usefulness of the employees’ outputs based on their belief of how they performed in that activity. Additionally, Bandura (1982) reiterated the general concept with his self-efficacy model of people’s response and reaction being based on their perception of their competency in face of new technologies resulting from the computer and internet evolution. Bandura (1982, p. 122) stated, “A capability is only as good as its execution.” Based on this, converting the paper forms and spreadsheets used for the command center logistical effort in 2007 to an alternative tool might be indicated. The action research of this paper is using lessons learned to convert the information tracked in the spreadsheets to a more technological format for easier and faster access in emergency response. This serves as the impetus for improving the existing emergency response process via a technology tool (Bandura, 1982; Davis, 1989; Venkatesh et al., 2003). The data used in this paper is from the 2007 Firestorm responses of San Diego Gas & Electric’s command center support. Their spreadsheets, emails, phone calls, including the personal observations of the lead author, who participated in the effort, provided the data for this study. The action research format is appropriate for this paper because the research process involved manipulating outputs containing the command center logistical details and their role in the activities in the organization’s response to emergency situations. The data that were analyzed related solely to the support of the command center camps and what it took to sustain their mission as bases from which to deploy the electric restoration personnel during the power outages. The details surrounding the ordering, receiving, inventorying, handling, staging, and deployment of construction materials, such as poles and wires, were not the purview of BSLs’ work in support of the command centers. Material handling was the responsibility of the Logistics and Warehousing groups. The spreadsheet listed 25 categories of information that included the following column headings: Region, Need for Site: Yes/Pending, Site Name, Command Center/Staging Area, Site Coordinator & Phone #, Thomas Brothers page, Address, Start Date, Estimated End Date, Total # of People, Food Counts, #B, #L, #D, water buffalo Y/N or 5 gal H2o, # of Porta Potties, Trash bin type: trash/cable/metal/treated wood/ recyclables, # of hand-wash stands, Portable Office Trailer, Copier, Lights, Generator(s), Parking Space Reqmnts (Requirements), Material Laydown Space Reqmnts (Requirements), Security, # of Vehicles, Phone /Data/Radio/ Printer, BMP Materials, Notes/Comments. The data that were examined were three versions of the spreadsheet, Firestorm 2007 Command Centers and Staging Areas, during the time frame of October 24, 2007 through November 10, 2007, to demonstrate the progression of the command centers’ expansion and contraction over time. The spreadsheet also displayed the types of items being tracked at a consolidated level. WILD FIRE LOGISTICS During the week of October 21, 2007, winddriven wildfires raged across San Diego County burning more than 360,000 acres and destroying at least 1,700 homes. As soon as the fires had moved through an area, SDG&E crews were on-site to assess damage to the utility’s assets. As a result, 750 SDG&E employees, along with 203 neighbor-state mutual-aid personnel and, at peak periods, 78 contract electric crews and 129 digging crews, were deployed to perform the service restoration efforts. Service was restored for approximately 83,000 customers affected by the fire and SDG&E eventually replaced more than 2,170 utility poles, 338 transformers, and at least 35 miles of overhead electrical wire (Geier, 2009). Four years earlier, after San Diego County’s 2003 wildfires, SDG&E had refined a model for supporting the logistical components of establishing and supporting command centers. Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),36-50,July-September2010 39 This model was ready for implementation during the October 21, 2007. SDG&E’s electric restoration response to the 2007 fires was supported by teams of employees from the Fleet, Facilities, Environmental, Supply Management, Logistics & Warehousing, Fleet Services, Facilities, Real Estate/Land, Environmental, and Safety departments (Dulgeroff, 2009). These teams were tasked with supporting the logistical needs for electric construction crews allowing the construction foremen and supervisors to be available to perform their core competency of repairing or rebuilding the electric system. Working in teams of approximately three people per shift, the support team employees were consumed for 10-15 hours a day for the first 10 days of the emergency. They had to respond to central points near the fire-damaged areas where they set-up command center camps to serve three meals a day. The command center sites included restrooms, shaded rest areas, water, ice, and snacks to all the crews who were working 16-hour shifts (Dulgeroff, 2009). In addition, the command center sites needed mobile offices with internet connectivity (often via satellite due to the remote locations and/or damage to on-site power and telecommunications connections). Also required were helicopter landing areas for hard-to-access, power pole replacements, water trucks for dust mitigation in the rural settings, mobile radio repeaters, and dedicated space for all the materials to be staged for the crews’ use. Preventative environmental checklists had to be followed at the rural sites. Straw wattle to prevent run-off, steel shaker plates to keep the vehicles from tracking loose dirt out onto the streets, and traffic control where the extra traffic related to the command center made an impact to normal traffic flow at the entrance were all required. Sites that were designated just as material lay-down locations, still needed restrooms and trash containers (Dulgeroff, 2009). The region’s basic communications infrastructure was not impacted by the fires, therefore alternate communications systems were not needed in the 2007 firestorm. It is timely to note however that emerging technologies leveraging communications devices already being used are being tested for response to emergencies impacting infrastructure. This new type of infrastructure-less wireless network, such as Georgia Tech’s LifeNet model (Wilson, 2005), which is formed out of consumer electronic devices such as laptops or smart phones could have great utility in disaster scenarios impacting the traditional means of communications. Using consumer devices such as smart phones and laptops to remove dependence on a company’s local network infrastructure and data center could also be addressed by cloud computing. In SDG&E’s 2007 response, the team members used cell phones and laptops with air cards to track all the support activities and items for command centers, on spreadsheets that were emailed back and forth to the manager leading the command center efforts. This enabled them to track the logistics of all command center sites in a consolidated manner. Each day the team members participated in two conference calls with their counterparts, during which they were given updates from other SDG&E departments supporting the logistical efforts. OBSERVED ISSUES The SDG&E team members who were serving as business solutions liaisons (BSLs) between the construction supervisors leading the restoration and rebuilding efforts of the crews, and the departments that procure and oversee delivery of the material and services used to set-up the command centers, reported occasions when their requests for items needed for the command centers were not conveyed exactly, or delivery was late. Other times requests were duplicated, or second and third calls were received from the providers to confirm if materials or meals were still needed, when the items had already been received. The BSLs’ daily submittal of the tracking spreadsheet to the manager’s administrative support required a significant consolidation chore and lacked real-time status. For example, an impact of the spreadsheet’s lack of timeliness was that if the BSLs needed to track something Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 40 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),36-50,July-September2010 more timely than information from the afternoon before, they would have to participate in both conference calls each day, despite other tasks they might have been performing. Additionally, the BSLs needed to take their own notes as to what each command center team was reporting during the calls. Then, the spreadsheet combined with their handwritten notes provided them with more detailed information, but resulted in an abundance of paper. Further, the daily consolidation of spreadsheets from 17 different command centers contributed to the chance of errors being made. The possibility of achieving efficiencies through a more real-time method, rather than paper forms, repeated consolidation of emailed spreadsheets, and email requests first arose during the type of miscommunication reported above. This suggested an opportunity for investigation of a more real-time approach to handling the logistics related to the emergency response. The rush to procure water, meals, tents, lights, portable bathrooms, and mobile offices is all performed in competition with other responding agencies in the region, public and private-sector. Any lag in response time could present the loss of the opportunity to obtain what is essential. Because of that competition for scarce resources, the first five days of a response are the most time-sensitive. Consequently, the potential for errors in the spreadsheets could have represented a loss of needed resources. Research discusses the range of spreadsheet errors from calculation errors, to one that is more germane to this study, errors in data quality. Caulkins et al. (2005) cite these errors in data quality as one of three typical reasons that contribute to undetected errors in an alarming “91 percent of spreadsheets” (p. 22). These authors contend that a simple task such as a “bad sort can destroy the integrity of a row, or a mismatch of units” (p. 23). This is supported by Panko’s research in which he estimated 94 percent of spreadsheets contain errors (Panko, 2007). Therefore, it is clear that any solution must need no manual manipulation of the data such as is needed when using spreadsheets for database functions. An alternate solution must also have virtually no training-time requirements, since the employees assigned to the task in subsequent emergencies will likely have no experience with the solution, nor would they have the luxury of time to learn it. The possibility of SDG&E changing its approach to command center logistics tracking as Commonwealth Edison and Duke Energy were able to do, merits further study. After the 2007 fires, event-assessment and debrief of the command center support personnel included the subject being raised of the efficacy of spreadsheets and paper forms and the question of whether an alternate approach could have achieved greater efficiencies, especially in the areas where duplications and problems were encountered was posed. This study addresses the following research questions: • • • What alternative approach would have enabled the teams working on the command center effort of emergency response to have easier-to-use, faster-to-access data? What solution would include the capability to store basic command center requirements from previous emergency responses to save time in addressing the first or most crucial items needed during the next emergency? What resiliency could be built into command center logistics tracking in case the company’s network was rendered inaccessible at the same time as command centers were needed for service restoration activities? Therefore, if SDG&E’s electric or gas system was damaged by terrorist activity or another natural disaster with equal or greater severity than the firestorms of 2003 and 2007, tracking the command center logistics could become exponentially more difficult proportionate to the severity of the disaster. An alternative solution designed for easy access by users in multiple areas could enhance the support of the logistical efforts to bring the region’s critical infrastructure back to normalcy in the most effective manner possible. Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),36-50,July-September2010 41 This study is designed to investigate the possibility of an alternate, multi-user approach, in order to improve accessibility or timeliness of information in a further refinement of disaster response tools. Results could aid decisionmakers in their planning for subsequent emergencies. The present body of knowledge will be expanded by the study of converting from spreadsheets and paper forms currently used for some emergency management activities to more effective tools of response. to make requests of many areas for their needs (Dulgeroff, 2009). The BSLs are not examined in this paper as a population. Their data output, however, is the sample examined. The BSL program would be activated under certain conditions including: • • • • POST WILD FIRE ANALYSIS After the Southern California fires of 2003, the Sempra Energy utilities’ vice president of Business Solutions, and Directors leading the support-type departments of Supply Management/Logistics, Fleet Services, Facilities, Real Estate/Land, Environmental, and Safety, formulated a program to further enhance their support to field operations areas in their response to emergencies. That new program was enacted for the first time during the 2007 Southern California fires. Under a dotted line reporting relationship to a manager responsible for all command center and staging area sites, “Business Solutions Liaisons” (BSLs) fulfilled a strategy that was designed to provide them as central points of contact. They were to interface between the operating centers’ field operations and SDG&E’s high-level central coordination and communications Emergency Operations Center (EOC), during emergencies that necessitated the set-up of staging areas and/or crew deployment areas. The BSLs were selected from a qualified pool of employees from the support departments, and based on their availability to respond to remote locations for ten-hour shifts. The BSLs were to be a visible presence in the field to facilitate proper two-way communications between field operations groups, the support departments, and the EOC, where requests would be received and dispatched to the proper responding support departments. The BSLs were intended to be the “one-stop-shop” for all requests instead of field supervisors having The utility experiences a major event The EOC is activated One or more off-property staging areas are needed It is requested by EOC staff The BSLs had to be ready to mobilize at the beginning of an event through response and recovery. They were to engage with operating groups and coordinate with existing emergency response systems. In addition, the BSLs were to be familiar with field operations activities, be prepared for inclement weather, and be field-ready. Teams of two BSLs were assigned to each location to cover two shifts per day. Sites were determined by identifying areas where the utility’s system had sustained significant damage, for economies of adjacency in location decisions. If there was suitable land, and the area met strategic placement guidelines, a command center was located there (Dulgeroff, 2009). In the first three days of the emergency, the Facilities department was activating its baseline set of plans, to achieve the earliest response to the expected need for some type of field command centers or staging areas. Therefore, they already had contingency plans in place for rapid procurement of several large tents to house the resting and dining functions for the emergency response, generators, lighting, drinking water, catering, and portable sanitation services. As soon as they reported to their assigned command center locations with their laptops and air cards, cell phones, and personal protective equipment, BSLs checked-in with the construction supervisors to receive any logistical requests and support them in their core mission of directing system repair for service restoration of the utility system. Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 42 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),36-50,July-September2010 As the emergency unfolded, appropriate numbers of utility personnel, mutual assistance crews, and contract construction company crews were deployed to repair damaged segments of the system (Dulgeroff, 2009) The BSLs were apprised by the construction supervisors, of the crew size increases or decreases, as work progressed across the damaged areas, and as mutual assistance crews arrived. The command center needs were adjusted depending on that crew movement (Dulgeroff, 2009). In order of priority, meals, water, ice, and sanitation facilities were the most important of the many things needed at the command centers. Shelter (tents), electricity, portable offices, internet connectivity, and garbage containers and garbage collection were next in order of priority. Following in priority were a fuel tanker to refuel the crew trucks and a pad for a helicopter to land, including a water truck to mitigate dust from the helicopter and the vehicle traffic. As soon as the BSLs reported to their assignments, they were to review the template documents in their binders, time permitting. There were often requests and/or people waiting for the BSLs as soon as they got to their locations, and the BSLs provided rapid response and follow-up on all requests. The One-StopShop was an exceptionally successful response model because the construction supervisors’ needs were continuous and the BSLs execution of the ordering and tracking was a better use of company resources. It quickly became apparent that the paper request forms in the binder were unworkable, for two main reasons. The forms were meant to be handwritten, then faxed to the EOC representative for that request category, however, there were no fax machines available, at least in the first several days of the emergency. Further, the forms required name, date, and time information. The natural inclination of the BSLs was to call the EOC representative and initiate the request via cell phone call, then complete an email note with the same information as a written confirmation, and for ease of follow-up. The email notes automatically recorded name, date, and time information, and the BSL added the other shift’s BSL in the carbon copy line. Early in the process, the paper request forms served as helpful visual cues for the details that needed to be included in the phone and email requests, but beyond that were hardly used. Similarly, the paper BSL Service Request Log was viewed as inconvenient given the Sent Items feature which saves sent-email in a folder in the sender’s Outlook program. Not all of the BSLs used this paper Log knowing their Outlook Sent Items function captured and archived them automatically and more permanently in case the information was needed at a later time. Given that the other shift’s BSL had been copied on any emails that alternate shift BSL did not need to review the paper Service Request Log to ensure smooth transition of information. FINDINGS AND DATA ANALYSIS What alternative approach would have enabled the teams working on the command center effort for emergency response to have easier-to-use, faster-to-access data? What solution would include the capability to store basic command center requirements from previous emergency responses to save time in addressing the first or most crucial items needed during the next emergency? What resiliency could be built into command center logistics tracking in case the company’s network was rendered inaccessible at the same time as command centers were needed for service restoration activities? The 2007 Southern California fires damaged a large amount of San Diego County’s territory including, in some areas, SDG&E’s means of delivery of electric service, its poles, wires, cable, and transformers (Dulgeroff, 2009). SDG&E coordinated a full-scale effort to restore service for its customers as soon as possible. “To provide forward support closer to the actual field locations where the bulk of the repair and restoration work was occurring, SDG&E established ‘command centers’ and ‘staging areas’ in strategic locations” (Geier, 2009, p. 14). Establishing and maintaining Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),36-50,July-September2010 43 the command centers and staging areas to support the crews in their restoration efforts, and ultimately closing them down, involved approximately 56 employees from supporting departments, for approximately three weeks in a significant logistical exercise (Dulgeroff, 2009). The process began with: (1) baseline requirements needed for a command center or staging area, (2) personnel assigned to the location to be the points of contact, called BSLs, and (3) the BSLs’ laptop computers with air cards for internet connectivity, the BSLs’ cell phones, a master spreadsheet to track all the details of all the command centers and consolidate them into one view, and paper forms on which BSLs could make requests for items needed at the sites. The spreadsheets did not look at the causes of the emergency, but helped manage the responses. The lead author, a team lead of the SDG&E supply chain systems team, served in the BSL capacity. The people who received the requests and information from the BSLs and fulfilled them were support department representatives manning their specific EOC positions. Each EOC position was staffed around the clock at the beginning of the emergency, and as appropriate later in the response, was reduced to 16 hours per day of coverage. The BSLs received all the requests associated with taking the command centers from their baseline configuration to fully-functioning field operations support, by working as the interface between the construction supervisors, the EOC representatives, and on occasion, directly with the vendors. The defined method for initiating a request for materials or services was in the library of paper forms. In practice however, instead of using the paper forms to initiate requests, there were two preferred methods by which BSLs sent requests to the EOC representatives. The first method was via cell phone call. The second method was via company email which was available via air card internet connectivity and VPN computer network access. These emails were often written follow-up to the initiating phone call. Also, since they were accessing the network via VPN, all internal drives and intranet pages and systems normally available to BSLs were accessible. The EOC representatives were on-site at SDG&E’s company headquarters location, and had the full company network available to them and received BSL email communications in that manner. The main tool used to handle the command center logistics effort was a spreadsheet. The spreadsheets used for this study are three samples of the daily iterations of the 2007 Firestorm command center and staging area consolidated information. All of the categories represented essential items needed to support response to the emergency. Many of the BSLs’ actual practices did not include filling-in the paper forms, which meant the paper forms provided in the binder went unused. The service request log in the binder was also not used by the BSLs as they preferred the daily-updated spreadsheets that conveyed the entire command center landscape. The spreadsheet in its consolidated form was the most significant tracking output of that effort. The main benefit of the spreadsheet was the ability to track, albeit a day later, the logistical details of the command center, which included upward reporting to and by the manager. This information was kept current by the BSLs completing and emailing the spreadsheets back to the manager after the afternoon conference call, for his administrative support’s updating. The inherent inefficiency of its manual consolidation prompted the research questions. A consideration of efficiencies to be gained and opportunities to track more categories of information, if needed, might reveal a new structure for the process, using an alternative approach to the paper forms and the spreadsheet. Consequently, the analysis of efficiency of the spreadsheets compared to real-time access to data, might include a new technology or a combination of several. An examination of the details of forms on smart phones, dashboards, blogs and wikis, and cloud computing would be timely, as SDG&E continuously strives for improvement of its crisis response. Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 44 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),36-50,July-September2010 DISCUSSION Based on the data resulting from SDG&E’s 2007 Firestorm command center support response analysis, two hours a day of conference call attendance, paper form requirements, and editing of the spreadsheet by each BSL team, indicates opportunities to improve efficiency exist. Several possibilities for alternative approaches to command center logistics tracking are growing in use and popularity. To address the first research question on what alternative approach to emergency response logistics tracking would be easier-to-use and provide faster access to data, rendering forms on smart phones in replacement of paper forms, blogs for BSLs to convey new situations or information they encounter, via a posting on a webpage, making it readily searchable for the next crisis, wikis, which can capture blog information of a more formal or permanent nature to document processes or procedures, and dashboards as a means to display the real-time, consolidated, command center information, links, and metrics are suggested. The use of a database as the backend data repository supporting what is displayed in dashboards related to the second question, and the concept of cloud computing related to the third research question will be discussed later. This combination of alternatives incorporates the concept of building resiliency into a company’s activities supporting critical infrastructure response work, as well as reducing duplicative or outdated methods. It also can be viewed through the lens of the expanded crisis response system model (Jennex, 2004). The expanded system encompasses more than the basic components of database, data analysis, normative models, and interface. Enhancing the model includes the addition of trained users (where users are personnel using the system to respond to or communicate about the emergency and consist of first responders, long term responders, the emergency response team, and experts), dynamic, integrated, and collaborative (yet possibly physically distributed) methods to communicate between users and data sources, protocols to facilitate communication, and processes and procedures used to guide the response to and improve decision making during the crisis. This expanded crisis response system model is applicable to the discussion of combining alternatives in order to better respond to the logistics tracking activities of SDG&E’s emergency response. Specifically, the dashboards fulfill the model’s condition for a dynamic method of communication, based on the realtime nature of the information received and displayed. The potentially constantly updated information supports the optimum display to users of what is residing in the underlying data sources and, further, to non-users such as executive teams, and decision-makers. The dashboards also fulfill the model’s component of integrated and collaborative protocols by their attribute of displaying multiple types of information from multiple sources or contributors that would not normally be displayed together in a single view, or viewable by so many different parties. Additionally, the nature of the field operations and the EOC functions being physically separated conforms to the model’s almost-certain distributed teams, sites, and means of communications. The sometimes duplicative activities associated with the BSLs’ use of paper forms and follow-up phone calls, and the time-consuming consolidation of that information in a spreadsheet with limited visual display characteristics, implied one or all of these alternatives may have efficacy in improving BSL-related data output in subsequent emergency efforts, as supported by the Jennex expanded crisis response system model. The result of analyzing enhanced concepts to the command center tracking needs, suggests further study of forms on smart phones, and dashboards connected to a robust back-end database already populated with baseline data as a starting template to save time, residing on remote servers in a private cloud. A site linked to the dashboard on which any participant in the command center effort could blog about any special experience for sharing with the other team members would be timely as well as easy to access. Problems or issues discussed Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),36-50,July-September2010 45 in the blogs and successfully addressed, could be elaborated upon or documented in a more formal and thoroughly developed way and posted as a wiki, as a more permanent and searchable archive. The need for resiliency in case of unforeseen inaccessibility, suggests the use of private cloud computing, which would result in a virtual, secure, remote location for SDG&E-related command center activities to be developed, housed, and delivered. That need for secure software and data is important because of the proprietary nature of the information. Therefore, a practical solution to help the utility in case of a catastrophic loss of its computer network might include private cloud computing. Possibilities of technological alternatives other than these may be suited for transforming the spreadsheets and the paper forms, and displaying command center logistics however, the intricacies of the related technology need to be addressed as further suggestions. The viability of these technologies in support of alternative systems for crisis response is derived from Jennex and Raman (2009). The forms on smart phones, blogs and wikis, cloud computing, and the dashboards similar to those developed by SDSU’s Visualization Center for flu tracking in San Diego County when implemented together, can all be considered components in the fusion of KM systems (Jennex & Raman, 2009). Filling out paper forms and completing the spreadsheets was actually just the manual assembling of a list of data in columns and rows. It was not a meaningful transfer of ideas, needs, and information. When ideas, needs, and information are assembled without further requirement for consolidation or user adaption, and then easily rendered visually, it results in expedient crisis response through impactful visual cues and a more efficient transfer of knowledge. The implication is the tacit knowledge from BSLs’ previous experiences can not be appropriately conveyed in a single file when the information is simply housed in a spreadsheet’s columns and rows format. However, the multi-dimensional, graphical layering, and photographic images displayed in a dashboard format heralds the emergence of knowledge transfer that would benefit the evolution of SDG&E’s emergency response model. Blogs to capture BSLs special experiences or issues, and wikis to more permanently archive the blog posts that lend themselves to formal processes and procedures, provide a searchable, easy to access forum for command center participants’ specific information, and should be included as links in the dashboard. Handling the knowledge in support of crisis response in this manner substantiates the Jennex and Raman (2009) assertion that, decision makers, when under stress, need systems that do more than just provide data, they need systems that can quickly find and display knowledge relevant to the situation in a format that facilitates the decision maker in making decisions. IMPLICATIONS AND RECOMMENDATIONS The research related to the spreadsheets revealed that they performed in an adequate manner based on the fact that the command center logistics tracking was generally accurate and the consensus among the interfacing groups was that the command center model performed very well. However, the lag time in displaying updated information on the distribution of the crews across the command center landscape during a day, and how that impacted meal counts which, potentially dramatically affected the crews’ perception of well being, can be surmised due to the manual consolidation effort necessary to keep spreadsheet data current. The literature validates the significance of the crews’ perceived well being as having substantial importance. Relative to the discussion of the crews’ well being, a tangent factor of the BSLs’ efforts in the command center support was their stated feelings of satisfaction related to their accomplishments, as discussed in a BSL debrief meeting in November 2007. The perceived usefulness of their efforts and their feelings of self-efficacy gave an indication of how likely BSLs would be Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 46 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),36-50,July-September2010 to embrace alternative approaches to the work if it meant further improvement in their response to a crisis. While this relates to implementing knowledge management systems in support of crisis response (Jennex & Raman, 2009) it must be acknowledged that the responsibility for efficient and timely restoration of essential services is still a mandate from the federal government as codified in HSPD7. Despite SDG&E’s culture of dedicated restoration activities, the BSLs’ satisfaction in support of the company mission, or the opportunity to leverage computer technology and knowledge management systems to enhance the logistical support, the restoration activities would still be required by the federal government even in the absence of those company requirements and employee satisfaction drivers. It also should be recognized that there could be a scenario when the normal network systems for achieving compliance with HSPD 7’s requirements would be completely interrupted. In the event of a loss of SDG&E’s computer network infrastructure, an alternative to the company network would be needed to handle the on-going restoration tracking efforts. One alternative to the traditional and local company network would be via cloud computing. In the event SDG&E’s computer network was rendered unusable, whatever applications were needed for the command center response could be housed on virtual servers in the cloud, and accessed via the internet, likely through wireless connectivity such as the BSLs’air cards. The implication is SDG&E could continuously access its disaster response support tools if they were housed and delivered outside the company network. The security aspect would be addressed by using the cloud computing model known as the private cloud, in which processing, storage, networking, and the application, are via an intranet, allowing for user authentication, and encryption of the company’s proprietary data. Computing via the cloud also allows for applications to be used on a pay-as-you-go pricing schedule (Vanmechelen et al., 2006). This would allow the cost of an important emergency response tool, after its initial development costs, to be borne only at the time the company is experiencing an emergency. Responding to emergencies often carries great costs, some of which can be recovered in rates if SDG&E’s governing regulatory agency grants such a request. Matching the costs of a pay-when-used type of emergency response support tool to the infrequent occasions of emergencies impacting SDG&E’s critical infrastructure, supports a position of careful stewardship of costs the company will eventually request to recover. Further, using cloud-based tools to house an emergency response logistics tracking application securely outside SDG&E’s territory and having it be accessible via any internet-based connection along with VPN, provides a resilience to catastrophic wildfire- or earthquakecaused network destruction that would interfere with tracking of, although not halt, SDG&E’s restoration activities in the field. Analysis of the manual, and not-well-used paper forms for BSLs to send requests to the EOC, demonstrated the need for an alternate request method. As was demonstrated by actual practice, phone calls and follow-up emails were preferred to the paper forms. However, the email was duplicative to the phone call, which suggests a less-than-optimal use of the BSLs’ time. The rendering of a paper form for display and user input on a smart phone has potential utility as the technology to replace the paper forms for command center logistics requests. Data initiated by the request is needed to execute the activity and provides the EOC recipient with record of the exact details needed for accurate fulfillment, almost always better than a phone call request, but usually not faster. Yet the BSLs all had cell phones and continuously made requests initiated by phone calls then followed-up with emails for solid confirmation. An alternative is using a smart phone with a form accessed from the phone’s embedded memory. The efficiency of using the hand-held portability and instant communications of a cell phone, while having the accuracy and full detail inherent in a form, can both be achieved with this technology. The electronic forms housed in the smart phones’ memory could also be avail- Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),36-50,July-September2010 47 able in a library of request forms linked on the dashboard, as a secondary and back-up location to the smart phones. The practices of the BSLs in the field imply this alternative merits further investigation and possible piloting during future SDG&E emergency drills. Another implication for future research addresses the need to experiment with the dashboard as a more informative alternative for the current spreadsheet data and to house links to other BSL-needed information. A test model would include a dashboard being provided to the BSLs, their decision-makers, and related support departments for further determination of the efficacy of the technology. The user requirements of both dashboard and cloud computing would be structured by the technical staff at SDG&E to determine whether cloud computing and dashboards are viable alternatives to replace the current emailed spreadsheet. The spreadsheet as the most significant tracking output of the BSLs’ 2007 efforts is important when considering how such a single, relatively simple file was the source for tracking a great deal of valuable SDG&E-owned or rented assets, providing the manager with the ability to report to the EOC and the Executive team, the extent of the command center support of field operations and his decision-making related to consolidations of sites, or needed geographical changes, particularly while the fires were still burning. Considering those substantive uses of the spreadsheet, the lag time in the data’s accuracy due to the need for consolidation, indicates the audiences could be better served as more real-time technologies become achievable for this specific purpose. Furthermore, the recognition of the potential need of other categories and/or sources of information implies the spreadsheet format was already at the limit of its utility. The need for a tool with the ability to display data in a visual, real-time format, with links to additional helpful or related sources of information such as BSL-related blogs and wikis, seems a likely area for future study, since dashboards provides internet-browser display of many categories of data from potentially many sources. Dashboards also provide for understanding of performance indicators of importance to a specific audience, by summarizing them and often displaying them by graphical icon. The summarized data displayed in a dashboard can be the starting point to drill down to their detail. Conveying the data in a summarized, graphical way represents a significant, behind-the-scenes effort of collection, consolidation, and presentation of the data that starts to transfer tacit knowledge when the consolidation allows for decision-making based on the whole picture that is drawn by the dashboard. Dashboards support two important points made by Turoff, supplying the best possible up-to-date information is critical to those whose action may risk lives and resources and that an emergency response information system must be an integrated electronic library of external data and information sources (Turoff, 2002). The data in SDG&E’s command center tracking spreadsheet has the potential to become enhanced decision-making information and therefore knowledge that can be applied by its users through the use of baseline data from past command center experiences to respond to a present emergency. Using a dashboard displaying data taken from the spreadsheet categories in a more actionable manner could enhance decision-making or at the least, readability and timeliness. The summarization and display of all of these types of information sources in a single dashboard enables viewers to drill down to access other websites, company intranet pages, participant blogs, related wikis, detailed data from maps and databases, and have it all accessible realtime. It would be a significant enhancement to the ways this information was available to SDG&E’s emergency response support teams in previous emergencies. This practice of selectively applying knowledge from previous experiences during turbulent moments of decision making, to current and future decision making activities with the express purpose of improving the organization’s effectiveness, would be possible via a KM system (Jennex & Raman, 2009) such as a dashboard along with its underlying means of delivery. Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 48 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),36-50,July-September2010 CONCLUSION This paper proposes using knowledge to drive an alternative to SDG&E’s paper request forms for command center needs, and suggests the use of request forms rendered on smart phones assigned to the BSLs for the duration of the effort. In addition, the development of dashboards, including links to command center participant blogs and wikis, is suggested as an effective alternative to spreadsheets to track the command center logistics, and provide easy to access information by means of links. Finally, the use of cloud computing as the development platform and host of an SDG&E dashboard application is recommended as a timely evolution to a KM system approach to achieve logistical tracking efficiencies. REFERENCES Jennex, M. E. (2004). Emergency Response Systems: The Utility Y2K Experience. Journal of Information Technology Theory and Application, 6(3), 85–102. Jennex, M. E. (2005). What is knowledge management? International Journal of Knowledge Management, 1(4), i–iv. Jennex, M. E. (2007, August 25). Knowledge Management in Support of Crisis Response. Paper presented at the ISCRAM China Workshop. Jennex, M. E., & Raman, M. (2009). Knowledge Management is Support of Crisis Response. International Journal of Information Systems for Crisis Response and Management, 1(3), 69–82. Panko, R. P. (2007). Two experiments in reducing overconfidence in spreadsheet development. Journal of Organizational and End User Computing, 19(1), 1–23. Turoff, M. (2002). Past and future emergency response information systems. Communications of the ACM, 45(4), 29–32. doi:10.1145/505248.505265 Bandura, A. (1982). Self-efficacy mechanism in human agency. The American Psychologist, 37(2), 122–147. doi:10.1037/0003-066X.37.2.122 Vanmechelen, K., Stuer, G., & Broeckhove, J. (2006). Pricing substitutable grid resources using commodity market models. Retrieved from http:// www.coms.ua.ac.be/publications/files/KVM_GECON_2006.pdf Caulkins, J. P., Morrison, E. L., & Weidemann, T. (2005). Spreadsheet errors: Are they undermining decision making in your organization? Public Management, 34(1), 22–27. Venkatesh, V., Morris, M., Davis, G., & Davis, F. (2003). User acceptance of information technology: Toward a unified view. Management Information Systems Quarterly, 27(3), 425–478. Davis, F. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. Management Information Systems Quarterly, 13(3), 319–340. doi:10.2307/249008 Whybark, D. C., Melnyk, S. A., Day, J., & Davis, E. (2010). Disaster Relief Supply Chain Management: New Realities, Management Challenges, Emerging Opportunities. Decision Line, 4-7. Dulgeroff, A. (2009). Application of San Diego Gas & Electric Company (U 902 M) for authorization to recover costs related to the 2007 Southern California wildfires recorded in the Catastrophic Event Memorandum Account (CEMA). Retrieved from http:// www.sdge.com/regulatory/documents/a-09-03-011/ testimony-dulgeroff.pdf Wilson, S. (2010). Next generation disaster communications technology now a reality with LifeNet. Retrieved September 25, 2010, from http://www. scs.gatech.edu/news/next-generation-disastercommunications-technology-now-reality-lifenet Geier,D. L.(2009). Investigation on the Commission’s own motion into the operations and practices of Cox Communications and San Diego Gas & Electric Company regarding the utility facilities linked to the Guejito Fire of October 2007. Zuber-Skerrit, O., & Fletcher, M. (2007). The quality of an action research thesis in the social sciences. Quality Assurance in Education, 15(4), 413. Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),36-50,July-September2010 49 Teresa Durbin holds a Bachelor of Applied Arts & Sciences degree in Public Administration with an Emphasis in City Planning, and a Master of Science degree in Homeland Security with a Specialization in Communications and Information Systems. She has over 10 years of management experience in utility accounting and regulatory work, supply chain systems administration, and help desk supervision for a large investor-owned utility. She has experience in responding to the logistics effort supporting gas and electric service restoration after large scale service interruptions resulting from regional emergencies or catastrophes. Ms. Durbin has an understanding of intergovernmental relations and specifically private-to-public interactions and requirements. She has participated in policy analysis and Rule-making filings at the State level related to merger-related filing and compliance requirements. Ms. Durbin coordinates the Business Resumption and Continuity Plan as well as the Pandemic Plan and Response training in a division-wide effort. Her Masters thesis and research analyzed the benefits and potential results of different information systems and decision support methodologies in the logistical effort related to utility service restoration, post-emergency. Murray E. Jennex is an associate professor at San Diego State University, editor in chief of the International Journal of Knowledge Management, co-editor in chief of the International Journal of Information Systems for Crisis Response and Management, and president of the Foundation for Knowledge Management (LLC). Dr. Jennex specializes in knowledge management, system analysis and design, IS security, e-commerce, and organizational effectiveness. Dr. Jennex serves as the Knowledge Systems Track co-chair at the Hawaii International Conference on System Sciences. He is the author of over 120 journal articles, book chapters, and conference proceedings on knowledge management, end user computing, international information systems, organizational memory systems, ecommerce, cyber security, and software outsourcing. Dr. Jennex conducts research for the National Center for Border Security Issues on risk management and technology integration. Dr. Jennex is a former US Navy Nuclear Power Propulsion officer and holds a B.A. in chemistry and physics from William Jewell College, an M.B.A. and an M.S. in software engineering from National University, an M.S. in telecommunications management and a Ph.D. in information systems from the Claremont Graduate University. Dr. Jennex is also a registered professional mechanical engineer in the state of California and a Certified Information Systems Security Professional (CISSP) and a Certified Secure Software Lifecycle Professional (CSSLP). Robert Judge holds undergraduate degrees in Biology and Botany, an MBA, and a PhD in the Management of Information Systems and Technology from Claremont Graduate University. His career spans the semiconductor, aerospace, consumer electronics, and Internet Service industries at mid-management and executive levels. He has held functional responsibilities that include Materiel, Manufacturing, Information Systems, Marketing, Project Management and Customer Support. Throughout the last 20+ years of his career, he has served both San Diego State University and the University of San Diego as an adjunct professor teaching graduate and undergraduate courses in Operations, Supply Chain Management, Manufacturing Planning and Control Systems, Project Management and Information Systems. His current research interests lie in understanding how barriers to knowledge flow arise as small organizations grow, and also in how Knowledge Management Systems in Supply Chains can influence innovation and the flow of non-logistical knowledge. Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 50 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),36-50,July-September2010 Eric Frost directs the SDSU Viz Center and co-directs the Homeland Security Master’s Program, which includes about 130 Homeland Security practioners from many Federal agencies, state and local government, NGOs, industry, and governments. Much of the program is based on actual interaction with real operational training in the US-Mexico border region, as well as many other international Homeland Security groups such as in Mexico, Central Asia, India, Africa, and Indonesia. Frost and his colleagues use many new technologies and protocols that are enhanced during exercises such as Strong Angel III (http://www.strongangel3.net/) for situational awareness for many challenges including H1N1 using tools such as http://www.geoplayer.com/ gateways for Banda Aceh, Katrina, Indonesia, and Haiti disasters (http://hypercube.telascience. org/haiti). Frost and co-workers work with sensor networks, wireless and optical communication, data fusion, visualization, and decision support for first responders and humanitarian groups, especially crossing the civilian-military boundary, especially in unusual coalition areas such as Somalia, Afghanistan, India. Haiti, and Mexico including using Cloud Computing (http://www. inrelief.org) with Navy to impact Humanitarian assistance like Haiti and Mexico earthquakes. Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),51-60,July-September2010 51 Curriculum Design and Development at the Nexus of International Crisis Management and Information Systems Keith Clement, California State University Fresno, USA ABSTRACT This case study discusses the role of education, curriculum development, research, and service in supporting information systems for crisis response management. The study describes the Council for Emergency Management and Homeland Security (CEMHS) organization that designs and develops academic programs and courses in these specialized areas. CEMHS combines all levels of education in California (from K-12 and postsecondary education) into a “state-wide solution” and network of academicians and professionals in emergency and disaster management, crisis response, and homeland security education and training. The organizational purpose is constructing a “vertical track” of academic programs and specialized programs to beneit and enhance information resource and crisis management. The implications and lessons learned from building collaborative partnerships between the crisis and disaster response academic and professional communities in academic program development and research initiatives are also discussed. Keywords: Best Practices, Crisis Management, Crisis Response, Crisis Response Community, Emergency Management, Information Systems INTRODUCTION There has been an increasing expectation for active and vigorous government and NonGovernment Organization (NGO) in responding to natural and man-made disasters. This expectation has translated into greater responsibility, higher visibility, and more media attention in the preparation, response, and management of meteorological, environmental, political violence, and other exigencies. In light of several DOI: 10.4018/jiscrm.2010070104 significant and recent international disasters, including the Thailand tsunami, Haiti and Chile Earthquakes, floods in China and Pakistan, forest fires in Russia, and Hurricane Katrina, have dramatically raised global consciousness and salience for a need to enhance crisis and disaster response and management. Due to the critical need to enhance preparedness, prevention, recovery, and mitigation in dealing with catastrophic or emergency incidents, crisp and efficient action is required to accomplish the “all-hazards” crisis response mission. The tremendous loss of life and human misery coupled Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 52 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),51-60,July-September2010 with huge costs for disaster relief and crisis response has evolved into a critical necessity to develop policies, social institutions, technology, and a culture of preparedness to protect lives, society, critical infrastructure, various sectors of the economy, and property. It is with little surprise that there are many changes to the topographical maps and two-way radio of yesteryear’s disaster and crisis planners, managers, and responders. Today, Geographic Information Systems (GIS), web-based applications, Google Earth, readily available satellite data, communications technologies, voice and data fusion on mobile devices and handsets are all common tools of the trade for emergency and crisis managers and responders. In an increasing technological world and social milieu, the role and value of information has grown exponentially. This is particularly true when designing and managing “real time” information resources, remote sensing assets, and additional knowledge systems in response to the data intensive demands of crisis response in the modern digital age. Developing and deploying information systems are critical tools in the response, preparedness, decision-making, and communication of emergencies. Crisis response personnel must have access to actionable information in order to make informed, effective, and speedy decisions as dynamic incident events unfold on the ground. Technology and information management are crucial components of crisis and disaster response. Public safety and emergency response personnel must be intimately familiar with current technological tools to assist in effective crisis response, improved preparedness, and enhanced humanitarian disaster management capability. This is due in part because of the critical “need for speed” in efficient crisis response: establishing situational command and control, restoring basic public health care services and business continuity, restoring transportation, critical infrastructure and additional logistical matters, utilizing geovisual analytics, managing information resources, and communicating information reliably (for those in need and for those responding) and any additional steps necessary towards crisis mitigation and recovery. Several “core principles” in Emergency Management include “comprehensive, progressive, risk-driven, integrated, collaborative, coordinated, flexible, and professional” (IAEM, 2007). These are substantial demands that are placed on our crisis responders and managers and we must work to build additional tools to increase our performance and utilization of information in crisis management. Given the totality of life and death decisions and hazard to property and security, a substantial need exists for quality and accessible education and training programs at the nexus of academic and professional interests in areas related to information systems, crisis response, and international disaster management. The need for academic programs and courses to support and prepare the current (and next) generation of information systems and crisis response management professionals has grown significantly as technology advances near daily. Thus, developing and enhancing education programs in this subject area is an important objective for several reasons. Education programs provide an avenue for faculty research, opportunity for students, enhances pedagogy and teaching practices, and further develops our knowledge base. Education and research programs can contribute to greater levels of community and university service for faculty, staff, and students as well as improves community engagement (as well as communities assisted through better trained, educated, and prepared crisis response teams). However, questions exist about emergency management and crisis response education and training programs. What is the current state of these programs? How are quality and accessible programs and courses designed, developed, and implemented? Are there comprehensive and coordinated academic programs that fully prepare specialized crisis response and managers in the utilization of information systems (and other important contributing fields) when coping with disasters? If this is not currently the case, how can academicians and practitioners work together to develop education standards Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),51-60,July-September2010 53 and programs to further support and enhance academic areas like information management, remote sensing systems, and knowledge systems dedicated to enhanced crisis management and response? These are some of the primary questions, issues, and themes addressed over the course of this case study. In terms of organizing this paper, we first discuss the role of education, curriculum development, research, and service in supporting a variety of academic disciplines and fields that contribute to effective crisis response. It is important to discuss the role and value of education, curriculum development, and implications in the field of information systems and crisis response. Next, and the primary purpose of this case study, is the description and discussion of an organization, the Council for Emergency Management and Homeland Security (CEMHS) that designs and develops academic programs and courses in these specialized areas. CEMHS is an interesting case study because it draws together all levels of education in California (from K-12 and Postsecondary Education) into a “state-wide” solution that links a network of academicians and professionals together in the fields of emergency and disaster management, crisis response, and homeland security education and training. Finally, we discuss the implications and lessons learned from collaborative partnerships between the emergency management and homeland security academic and professional communities in academic program development and research initiatives. THE ROLE OF EDUCATION AND CURRICULUM DEVELOPMENT FOR INFORMATION SYSTEMS AND CRISIS RESPONSE MANAGEMENT What are the implications for academic programs implementation in advanced scientific fields such as information systems management as related to crisis response? These questions quickly become complicated when talking about dynamic, technical, and scientific disciplines and specializations like information resource management. In addition, the past few years have seen an increased need and demand for a specialization in crisis management and disaster response education and training programs. “In most societies education is constantly being asked to do more and more things, to higher and higher standards, with greater accountability and finite (if not diminishing) resources.” (Davies, 1999, p. 108). Due to the relatively new entry of crisis response and management related subjects into academics, colleges, and universities, there is much planning, design, development necessary for implementation and adoption at higher education institutions. This is particularly true for more focused scientific disciplines like information resource management. Any absence in quality education and training opportunities reduces our capability to prepare, respond, mitigate and recover from natural and manmade catastrophes. Thus, any potential challenges and rewards for developing quality disaster management and academic programs justify the creation of partnerships among key stakeholders and partners to achieve long-term program sustainability and development. Scholars and practitioners should develop and communicate best practices” to improve the management and utilization of information as a resource to enhance international disaster response capabilities and performance. Technology and information resources and needs indicate that scientific, technical, and professional workforce and skill demand must be readily prepared and skilled to support the many diverse fields that participate in disaster and crisis response. We must find ways to support development (and subsequent response capability) by designing and implementing education and training programs to support global information needs by enhancing knowledge systems management through education and training strategies. It is important that educators contribute their expertise to crisis response and disaster management academic program organization and structure. Educators contribute to the solution through the development, codification, and transmission of important Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 54 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),51-60,July-September2010 expertise, knowledge, skills, and experience relevant to emergency and disaster management. “The demands being made upon teachers and others who provide education call out for educational practices to be based on the best available evidence as well as the professional skills, experience, and competence of teachers” (Davies, 1999, p. 117). To increase supply and capacity of quality education and training in this area, we must understand the current state of these academic programs, discover potential gaps in content and curriculum, adopt quality control and evaluation measures, and support enhanced academic standards. If education program and course supply and demand needs can be met, we can enhance the availability and quality of crisis response and management through standardization of programs, government incentives and funding for program development, and the benefit of employment of students graduating from these reputable academic programs. One solution may be the creation of a template or blueprint to guide the design, development and implementation of academic programs at various levels of education. This would assist in the furtherance of education, training, and research collaboration, global partnerships, and an enhanced network of experts to deploy in the field, data visualization centers, command and control, joint incident command posts, and on incident response teams. It is also critical to seek educational and training solution to further integrate the areas of crisis response and information systems. CEMHS ORGANIZATION The Council for Emergency Management and Homeland Security (CEMHS) was formed in 2008 to enhance and strengthen Emergency Management (EM) and Homeland Security (HS) education and training programs and research partnerships within the State of California. CEMHS is a multi-campus faculty affinity group within the California State University and promotes quality and accessible EM-HS academics, research initiatives, and university/ community service connections. The purpose of CEMHS is to link comprehensive programs, courses, and curriculum through the promulgation of “model programs” as developed by faculty and professional subject matter experts teams and carefully linked with effective pedagogical practice. The organizational purpose is accomplished through the utilization of a curriculum design development framework and blueprint that guides the construction and development of different types of emergency management and homeland security degree and certificate programs (Associate through Doctoral programs). CEMHS promulgates the development of a “state-wide education and training solution” in the form of comprehensive “vertical track” education and training programs to improve disaster and crisis response and, for example, the efficient utilization of information resources and technology. In the program design and curriculum development phase, it is important to ensure that programs contain all requisite skills and knowledge competency areas necessary to master and perform within a given field or discipline. The “templates” and “best practices” framework are intended for replication as a model curriculum by additional educational institutions interested in offering these programs and courses. Thus, there is the development of a clear road-map in academic programs from professional training to advanced terminal degree program in disaster and crisis response and information systems management. CEMHS provides facilitated coordination to help develop academic programs to enhance crisis response and information systems management and greatly benefit the international disaster response community. The Importance of Partnerships and Collaboration The primary means of achieving organization education and research objectives is through collaboration and partnering between state educational institutions and major stakeholders and the development of participation within the Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),51-60,July-September2010 55 emergency management and crisis response community. As an organization, CEMHS fosters a shared vision in designing and promoting EMHS curriculum and programs with the active participation and critical input of key partners and stakeholders. Particular emphasis is placed on building facilitative and stronger relations in a growing network of academicians and professionals within the field. It is important to foster partnerships and collaboration and develop a strategic plan to implement education and training programs at various levels of government: international, national, state, and local, educational partners, additional public organizations, and the private sector. The discussion of EM-HS curriculum and initiatives should be conducted within the framework of delivering standardized, portable courses and programs that fulfill the needs of key partnering agencies, faculty, students, and respective campuses. It is vital to coordinate streamlined crisis response and emergency management academic programs. This is particularly true for specialized academic disciplines like information systems and management. In order to handle disasters efficiently, crises responders should have achieved specific competencies and mastery to perform their respective tasks and duties. All involved in crisis response should be properly prepared and experienced in handling incidents. All crisis responders and information resource team members should be on the same page to accomplish tasks as necessary. This is accomplished through education, training, and practical experience. To help sate a demand for information and knowledge, key partnerships within the academic and professional communities have formed and coalesced with the purpose of enhancing disaster preparedness, crisis response, through education and training programs. The objectives of CEMHS include building of strategic partnerships among key stakeholders in collaborative partnerships to support the programmatic and curriculum development needs. There is a need to develop education programs to provide a firm foundation for the many specialized areas of crisis response like information management, public health, nursing, civil engineering, victim services, and many other contributing fields and disciplines. This activity is achieved through increased “predictable collaborative action” and through building multi-campus consortiums and university-agency partnerships. Since many of the courses utilized in these programs already exist, it is primarily an issue of getting involved stakeholders together and deciding the critical questions of what these programs look like and making it happen. In this way, the needs of major partners are reflected in the types of academic and training programs developed, curriculum and content designed, and necessary skill and knowledge areas emphasized. Education, research, and service initiatives are supported and represented by a variety of key stakeholders (agencies, universities and colleges, the private sector, public organizations). In terms of educational partners, members are found on campuses from the University of California, California State University, California Community Colleges, County School Boards, and Private Universities. In terms of professional partners, the California Emergency Management Agency (Cal EMA) has provided grant and funding support for CEMHS. As of August 2010, CEMHS has over 525 participants drawn from academic and professional partners. IN ADDITION, presentations have been made to the United States Department of Homeland Security (DHS) and the Federal Emergency Management Agency (FEMA) Emergency Management Institute (EMI). CEMHS has worked closely with the Disaster Resistant California Community Colleges (DRCCC) group and the City of Los Angeles Emergency Management Department’s EM University Consortium (LAEMUC). In addition, the International Association of Emergency Managers (IAEM) has an official liaison with CEMHS workgroups. These partnerships have been mutually beneficial and supportive and results in much feedback from colleagues in the many diverse fields contributing to crisis response management. Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 56 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),51-60,July-September2010 The California Emergency Management and Homeland Security Education and Training Strategic Initiative Current CEMHS activity consists of preparing a strategic vision and framework to design, develop, and implement EM-HS education and training programs for K-12 and Postsecondary Education state-wide. This framework, the California Emergency Management and Homeland Security Education and Training Strategic Initiative, sketches an outline and blueprint to guide and develop a “vertical track” of seamless, standardized, and portable Emergency/ Disaster Management and Homeland Security education and training programs at various levels of education from Training Academy through Associate, Bachelor, Master and Doctoral degrees. The benefit of the vertical track is that students would have the opportunity for academic programs through terminal degrees. The Cal EMA has funded the development and writing of the EM-HS Strategic Initiative for 2009-10. Details of the Strategic Initiative, including mission, vision, and objectives are discussed in the following sections. Strategic Initiative Mission and Vision The primary mission of the Strategic Initiative is to design, develop, and implement a comprehensive, seamless, standardized, and coordinated ‘vertical track’ of emergency/disaster/crisis management education and training programs and curriculum at all stages of educational attainment (K-12 and Post-Secondary Education) to support the critical needs of key stakeholders and partners in this important subject area. The vision of the Strategic Initiative is to facilitate a collaborative partnership between public and private educational institutions to develop “standardized, seamless, and portable” education and training programs with the consultation and guidance of key partners, critical stakeholders, and Subject Matter Experts (SMEs). Strategic Initiative Objectives There are a variety of objectives to foster collaborative relationships and partnerships within the academic and professional community in the disaster and crisis management and help reinforce a culture of preparedness, awareness, and safety on-campus and within our communities. The primary objective of the Strategic Initiative is to provide a comprehensive and linked series of academic programs designed to meet the complexity and rigor of needs and skills to fully support the academic and professional communities for safer, more resilient, and better prepared communities. The first objective is to develop a common understanding of crisis and disaster management education and training programs (i.e. core principles and goals found in these program; types of important knowledge, skills, and courses to include within programs and courses; and how programs are linked together to support the development of an academic culture and student support means) required to provide this important area of education and training, workforce development, and enhance operational capability. The second objective is to seek to determine current capacity (and gaps) within the EM-HS education and training programs and fill these capability gaps. This includes a discussion of the mission and vision of EM-HS education and training programs, foundational goals and types of courses/materials to teach and related values like civic engagement, development of a public service ethos, program/course delivery methods, internships, service learning, and other auxiliary functions to support core campus academic programs throughout K-12 and Postsecondary Education. Third, it is important to design, develop, and implement a linked network and portfolio of EM-HS programs (through various levels of educational attainment). Also to assist in matriculation and transfer opportunities among these linked programs to develop a coordinated “vertical and horizontal track” of EM-HS (and Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),51-60,July-September2010 57 related programs) through terminal degree programs. The ultimate objective is to design a model curriculum and series of program/ course templates replicable across schools, universities, and colleges to further support emergency management and homeland security education and training. This is accomplished through the development of “templates” and “best practices” for a model curriculum that details courses, content, and student learning objectives for programs and courses in this critical subject area. EM-HS program templates should be structured and designed with respect to the academic and curriculum review process of a particular campus (i.e. covering student learning outcomes, course descriptions, and relevant institutional policies). Towards this end, a tremendous amount of courses and materials already exist within the public domain to coordinate into a workable set of organized programs easily structured and modified by respective campuses based on their program specifications and curriculum review process. In other cases where little program and course material currently exists, new programs courses, and specializations to “fill in the gaps” will be designed and built to round out EM-HS education and training programs. We need to frame the discussion of key components of the Strategic Initiative in the context of a “minimum baseline” and guidelines to enhance EM-HS academic programs. At the end of the day, campuses should be able to utilize a template for minimum baseline requirements for whatever degree or certificate program they are interested in; add respective institutional strengths and faculty specializations, geographical location considerations, community interests; any additional requirements (a capstone or internship class) and call the program their own. Some have kindly referred to this as the “CEMHS Menu Approach” to education/training program development. The “menu approach” reference signifies that there are many different factors that educational institutions keep in mind when making the decision to build and offer various programs and courses. Some of these factors —types of program, intended student audience, anticipated demand, available resources, faculty and institutional strengths, etc. are instrumental in the decision to offer (or not) respective academic programs. Programs should be planned and designed with these factors in mind. IMPLICATIONS AND LESSONS LEARNED In order to “assist the development of an educated and well trained scientific and technical workforce to respond to information demands and crisis response,” we link vast resources, programs, and faculty expertise together and provide education and research solutions to designing comprehensive EM-HS core and specialization programs and courses in a multiple campus consortium. Due to the many factors and components involved in academic program development (and regardless of discipline or specialization), there are four linked elements of education, teaching, research and service/ humanitarian outreach implications and lessons learned in the development of information systems and crisis response academic programs. These are examples of the implications and lessons discussed in this case study. • • • • Primary Education Strategies Auxiliary and Education/Training Support Faculty/Student Research Opportunities Service/Community/Humanitarian Outreach Primary Education Strategies Primary education strategies involve matters of core and specialization curriculum design and development. CEMHS is developing a core curriculum and templates corresponding to various EM-HS degree and certificate programs based on “model practices” content and pedagogy and deliverable through various delivery modes (traditional, online, hybridized) and through collaborating campuses. Based on institutional and faculty interests, specializa- Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 58 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),51-60,July-September2010 tions, strategic needs; program coordination is facilitated with program coordination and key partners to develop programs to serve STEM fields (Science, Technology, Engineering, and Mathematics) needs (at the international, regional, and national levels). In addition to core curriculum, a full slate of programs that correspond with special areas of “strategic interest” or addressing specific gaps in “high needs” technology gaps should be designed and developed soon. These specific areas of education and training are “specializations.” Different specializations are found on all campuses as each campus individually develops a respective culture relating to teaching, research, and faculty/ staff expertise. Campuses often collaborate and link together within the context of networks or “multi-campus” consortiums to provide these programs. Each campus in the consortium contributes to the offering of general programs as well as specializations based on important considerations like geographic location, institutional strengths, faculty/staff expertise, additional funding sources/research centers, etc. In other words, programs are blended into various levels of education, and tailored to the needs of students pursuing different educational goals, with the types of programs that institutions chose to develop and implement them. In program areas of exceptionally high academic need, external funding sources would be utilized to support the development of critical topic areas that meet the technical and scientific needs of partners and stakeholders, like in the areas of data visualization, cryptology, communications interoperability, remote sensing and many others. There is a particular need for some types of programs to be designed and develop and guided by the needs and specifications of key stakeholders and their view of current (and future) needs of the EM-HS workforce. In addition to the development of primary education programs, there is an additional need for auxiliary support and the development of faculty and student opportunities in the areas of crisis response and management. Auxiliary Education/Training Support and Faculty/Student Research Opportunities Supporting the implementation of a comprehensive and coordinated education and training programs in a vertical track of academics is an initial first step. However, we must also develop educational infrastructure to assist in student and faculty success during the nascent EM-HS program development process. Without too much detail, one important factor that partially differentiates crisis/disaster management education and training from more “traditional academic pursuits” is in the operational and experiential needs and information systems management connection with international and national government agencies, NGOs, private sector, and others involved in crisis response. It follows intuitively there must be additional linkages between agencies and academics so experiential learning environments should be developed. Linkages and partnerships include internships, field placements, career and job skill development fairs, drills, exercises, tabletops, etc. In addition to providing linkages and connections between academics and professionals in terms of experiential learning and research opportunities, there is also interest in deepening the network between professional and educational partners in the development of EM-HS academic programs at various levels of education. Service/Community/ Humanitarian Outreach It is important to support international disaster response and information systems through engagement of Service Learning and Civic Engagement within our communities and internationally. Commitment to community engagement should reflect in the curricula and programs designed as well as in building experiential learning networks between professionals and academicians. This could be accomplished by developing international information, academic technology, communications, and interoper- Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),51-60,July-September2010 59 ability groups to coordinate disaster response with current technologies, data-analysis, and visualization capabilities to support additional student and faculty education and research opportunities. Finally, support additional collaborative efforts for education, training, research, and community outreach programs and initiatives. Build linkages with additional public and private organizations, and tribal governments to support community outreach in EM-HS fields and promote community and campus preparedness, mental health and victim services, and enhance resiliency. CONCLUSION As relating to the area of information systems and crisis response, it takes many working together as a team to design, develop, and deploy technical solutions to disaster and emergency management. These experts in-training require education and training programs to reach their professional goals. We need geography and geology researchers to understand the dynamics of earthquakes and wildfires as well as utilize remote sensors and enhance GIS technology and delivery. Victim services personnel for crisis response, future educators, and squaring away educational needs for groups making valuable contributions to more resilient communities are important as well. Thus, there are a wide variety of knowledge and skill areas to deliver efficiently and effectively. Many of these key curriculum decisions relate to the educational learning objectives promulgated by the specific program under design and development. Workable and feasible “real-life” programs are needed to improve access, availability, and quality education programs in information systems crisis response. One important case study theme is an interest to develop and enhance education, training, and research programs. These programs, courses, and curriculum should be designed with the input and feedback of agencies and professionals (state, federal, local) and private-sector partners regarding necessary skills and knowledge areas required for mastery in these critical subject areas. In addition, these academically rigorous programs and courses must be fitted seamlessly so that students can sequence from training/vocational academy through community college and bachelors and masters degree programs. Various modes of effective program and course delivery are also explored (traditional courses, asynchronous delivery, hybrid/blended models, etc.) There are many necessary details to discuss in the process of designing and developing EM-HS academic programs. There has generally been a low level of coordination and integration of educational resources and programs towards the development of a comprehensive national or state-wide solution for education and training programs. The boundaries of the field, programs, and courses require additional development and specification related to curriculum, content, and program delivery methods. Academicians and professionals should collaborate to develop a “common understanding” of important principles and subject area content to design quality EM-HS education and training programs that meet the needs of both communities. In closing, one purpose of this case study is to provide guidance and information on designing and developing education programs that promote preparedness and response and reflect sound crisis and disaster management principles and “best practices.” This gives us pause to consider themes like how information systems and crisis management should be reflected in post-secondary education programs and course content. It is important to build programs that speak to the critical relationship between information, technology, and how to apply these to crisis response situations that occur across the globe. REFERENCES Davies, P.(199). What is Evidence-Based Education? British Journal of Educational Studies, 47(2), 108–121. doi:10.1111/1467-8527.00106 IAEM. (2007). Principles of Emergency Management. Retrieved August 10, 2010, from http://www. iaem.com/EMPrinciples/documents/PrinciplesofEmergencyManagement.pdf Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 60 InternationalJournalofInformationSystemsforCrisisResponseandManagement,2(3),51-60,July-September2010 Keith Clement is an Assistant Professor in the Department of Criminology at California State University, Fresno and Director of the Graduate Homeland Security Program. Professor Clement teaches undergraduate courses in Comparative (International) Criminal Justice and graduate courses in Psychology of Terrorism and Crisis Response and Essentials of Homeland Security. His research interests include risk perception of border security, transnational crimes, and its effect on security, and interoperability/communications issues. He also serves as Planning Director for the CSU Council for Emergency Management and Homeland Security (CEMHS), a 525+ member organization linking universities and colleges with key stakeholders to build EM-HS education and training programs and curriculum. He is a recipient of a United States Department of Homeland Security (DHS) Early Career Faculty Award under the Scientific Leadership Award for MSI Program. Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. International Journal of Information Systems for Crisis Response and Management An oficial publication of the Conference on Information Systems for Crisis Response and Management Mission The mission of the International Journal of Information Systems for Crisis Response and Management (IJISCRAM)is to provide an outlet for innovative research in the area of information systems for crisis response and management. Research is expected to be rigorous but can utilize any accepted methodology and may be qualitative or quantitative in nature. The journal will provide a comprehensive cross disciplinary forum for advancing the understanding of the organizational, technical, human, and cognitive issues associated with the use of information systems in responding and managing crises of all kinds. Subscription Information IJISCRAM 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/ijiscram. 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. TOC Alerts Please go to www.igi-global.com/IJISCRAM and click on the "TOC Alert" tab on the left menu if you want to be emailed the table of contents of each issue of this journal as it is released. Copyright The International Journal of Information Systems for Crisis Response and Management (ISSN 1937-9390; e-ISSN 1937-9420). Copyright © 2010 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: Murray E. Jennex, San Diego State University, USA Bartel A. Van de Walle, Tilburg University, the Netherlands IJISCRAM@iscram.org 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 IJISCRAM is listed in the following indexes: Cabell's Directory; GetCited; Index of Information Systems Journal; INSPEC; Media Finder; Norwegian Social Science Data Services (NSD); Standard Periodicals Directory; Ulrich's Periodicals Directory