252 Int. J. Innovation and Sustainable Development, Vol. 7, No. 3, 2013 Promoting decision making through a Sustainable Remediation Assessment Matrix (SRAM) Aspasia Kalomoiri* and Washington Braida Center for Environmental Systems, Department of Civil Environmental and Ocean Engineering, Stevens Institute of Technology, Castle Point on Hudson, Hoboken, NJ 07030, USA Email: akalomoi@stevens.edu Email: wbraida@stevens.edu *Corresponding author Abstract: This paper describes the steps taken in a decision making process through a Sustainable Remediation Assessment Matrix (SRAM). The development of the SRAM deals with Complex, Large-scale Interconnected, Open, and Socio-technical System (CLIOS). For both large and small contaminated areas, considers potential impacts on neighbouring areas, the contribution to air emissions from the materials of the proposed project and the energy to be consumed. Along this line, the research focused on setting up a model under a systems perspective. A systemigram, from remedial investigation to project closeout, has been developed. For each stage of the remediation project, the process to identify stakeholders has been outlined. Moreover, and as an illustrative example, environmental, social, and economic aspects of remedial operations have been addressed on a specific case using the US Air Force Sustainable Remediation Tool (SRT). Keywords: SRAM; CLIOS; complex large-scale integrated open socio-technical systems; SRTTM; sustainable remediation tool; TCE; trichloroethylene; PCE; perchloroethylene; multicriteria analysis; systemigram; MNA; monitoring natural attenuation; SVE; soil vapour extraction; ISCO; in situ chemical oxidation; ERD; enhanced reductive dechlorination; ERH; electrical resistive heating. Reference to this paper should be made as follows: Kalomoiri, A. and Braida, W. (2013) ‘Promoting decision making through a Sustainable Remediation Assessment Matrix (SRAM)’, Int. J. Innovation and Sustainable Development, Vol. 7, No. 3, pp.252–270. Biographical notes: Aspasia Kalomoiri is a PhD candidate and a Research Assistant at the Center for Environmental Systems (CES) at Stevens Institute of Technology (Hoboken, NJ, USA). Her main research focuses on analysing complex socio-technological systems that involves life cycle assessment, risk and uncertainty analysis in sustainable engineering design and construction, especially in ground water and soil treatment. Washington Braida is an Associate Professor at the Centre for Environmental Systems (CES) at Stevens Institute of Technology (Hoboken, NJ, USA). This paper is a revised and expanded version of a paper entitled ‘A process for developing a Sustainable Remediation Assessment Matrix (SRAM)’ presented at the ‘XI International Conference of Protection and Restoration of the Environment’, Thessaloniki, Greece, 3–6 July 2012. Copyright © 2013 Inderscience Enterprises Ltd. Promoting decision making through a SRAM 1 253 Introduction Sustainable remediation has been defined as the practice of “demonstrating, in terms of environmental, economic and social indicators, that the benefit of undertaking remediation is greater than its impact and that the optimum remediation solution is selected through the use of a balanced decision-making process” (SuRF-UK, 2010). A related definition covering Green Remediation was advanced by US EPA: “The practice of considering all environmental effects of remedy implementation and incorporating options to maximise net environmental benefit of clean-up actions”. As such, the goal of green and sustainable remediation practices is to minimise the footprint of remediation and to avoid the transference of risk between environmental media. This can be achieved by incorporating methods to conserve water, improve water quality, increase energy efficiency, manage and minimise toxics and waste, and reduce emission of air pollutants and greenhouse gases during all phases of remediation including investigation, design, construction, operation, monitoring, and site closure. Moreover, social impacts of any remedial activity must be considered including the safety of workers and residents, noise and vibration impacts, ethical procurement and policy compliances. Last, but not least, under the sustainable remediation umbrella, the economic impacts of remedial activities need to be considered too. As it can be seen, these two definitions are different in scope: while SuRF-UK considers remediation activities as part of the broader sustainable development and considers environmental, social and economic factors, EPA focuses on the selection of the most environmentally friendly technology to achieve a given remedial objective, paying less attention to social and/or economic considerations. This new paradigm has resulted in the need to develop new assessment tools and metrics such as water intensity (the amount of water necessary to remove one unit mass of contaminant), soil intensity (the amount of soil displaced or disturbed to remove a unit mass of contaminant), material intensity (the amount of raw materials extracted, processed, or disposed per unit mass of contaminant treated), energy efficiency (the amount of energy needed to remove a unit mass of contaminant), and carbon intensity (unit mass of CO2 released per unit mass of contaminant). Moreover, since remedial activities are carried out based upon the objective to reduce risk to potential receptors, a general metric that takes into account the minimisation of overall input (environmental, economic and social) per unit of risk reduced, needs also to be developed. This allows the inclusion in the analysis of remedial activities not based on mass removal. The interdependence of the many components of a “remedial system” (i.e., environmental remediation technologies, costs, risks, society, and environment) presents considerable challenges in the development of a useful framework to guide the implementation of sustainability principles and to evaluate their success. The main challenge identified in this paper is to produce assessment tools that are transparent in the input of data and assumptions that help decision makers to make a balanced decision rather than perform the assessment per se. 2 Methods 2.1 General The explosive population growth experienced in the last 100 years resulted in an increased over exploitation and degradation of land and water resources. As a result, 254 A. Kalomoiri and W. Braida sustainable reclamation of land and water resources through environmental remediation is more than ever necessary in order to continuously provide a sustainable quality of life. Green and sustainable remediation focuses on the “internalities” (remedial objectives, system performance, costs and environmental impacts local to the remediation site) as well as the “externalities” (extended environmental, economic, and social impacts) of a remediation project (Braida and Ogundipe, 2010). The effectiveness of any tool in assessing sustainability depends on the defined boundaries of the system. Hence such boundaries need to be inclusive of enough materials and energy flow streams to accurately portray how truly sustainable the approach is, yet not to burden the analysis with extraneous information. The SRAM summarises environmental, economic, and social indicators for sustainability of a given remedial action plan to support the decision making process. The matrix incorporates relevant metrics, units of measure and factors for consideration, which are relevant to the three dimensions of sustainability; environment, economy and society. The SRAM comprises three columns representing the three dimensions of sustainability and a variable number of rows depending upon the number of the possible remediation alternatives for the site. 2.2 A framework for assessing green and sustainable soil and groundwater clean-up: a theoretical framework approach for SRAM As we deal with an engineering system, it is necessary to look at the entire contaminated site from a holistic point of view. The main goal on this framework is to provide a common methodology in order to analyse such a system regardless the degree of complexity. The American Cybernetics Society defines system analysis as: “An approach that applies systems principles to aid a decision-maker with problems of identifying, reconstructing, optimising and managing a system, while taking into account multiple objectives, constrains and resources”. This paper presents the steps towards the development of a matrix to assess sustainable remediation under a system analysis umbrella, through which the decision makers may be able to select the most optimum remediation approach satisfying Sustainable Remediation constrains. At the end of this section, a ‘systemigram’ is presented which illustrates the overall picture of the whole decision making process. A systemigram is a very important tool in the hands of a decision maker, as it provides an overview of the whole process in only one figure which conveys the project’s information to any category of stakeholders. It is expected that this can enable stakeholders to understand the steps and the importance of their contribution in order to finally select the most sustainable remediation method. Stakeholder identification Different stakeholders value different aspects of systems performance in different ways, making decision-making difficult. The stakeholder’s group can include a wide range of categories such as the following ones:  Federal and state regulators  Local government  Site owner  Responsible parties 255 Promoting decision making through a SRAM  Local residents affected by the site  Residents of neighbouring areas affected by the site remediation actions  Members of environmental organisations, etc. Those groups can be narrowed down to two major ones, the technical experts and the public policy group. Communicate the system boundaries to stakeholders Stakeholders need to be aware about the system boundaries, including:  Site specifications (site size and location)  Contaminants (hazardous or not)  Remediation time frame / actual risk (how immediate the remediation actions are)  Current situation on site and the nearby areas considering the environment, economy and the society.  Other constrains pertaining to the specific site Technical experts most of the time lack the ability to communicate the scientific data to the public and at this point decision makers roll is to lead the engagement between the major stakeholders group. The interaction between stakeholders provides a chance for everyone to understand the scientific analysis process and to agree on the bounds of uncertainties acceptable for making decisions (Figure 1). Figure 1 Communicate the system boundaries to stakeholders (see online version for colours) Stakeholders must be communicated through desicion-makers group to System boundaries: Site specifications, current Environmental, Social and Economic impact Source: Own illustration Communicate the Sustainable Remediation Goals to stakeholders This is an interactive relationship. Decision makers need to inform a principle list of what the SR goals for the specific contaminated site should be, but on the other hand, stakeholder’s interaction and opinions could result in a most optimum list on SR goals. The interconnection is illustrated in Figure 2. Sustainable remediation goals could include:  Reducing waste production and maximising the recycling and reuse of clean materials  Reducing the consumption of virgin and non-renewable resources  Minimising energy consumption and greenhouse gas emissions  Improving energy efficiency  Coordinating remediation with green building design 256 A. Kalomoiri and W. Braida  Enhancing ecology and biodiversity during site development  Creating new jobs  Increasing the areas value, etc. Figure 2 Constructing a more complete list of SR Goals, considering stakeholders needs (see online version for colours) Must be cleared to Sustainable Remediation Goals Stakeholders Set the Source: Own illustration Categorise and weighting stakeholders importance Categorise and weighting of stakeholders’ importance is necessary. All stakeholders need to be heard; however some stakeholders have a greater weighting factor than others. This is one of the most difficult steps in the whole procedure. Decision makers can prioritise and weigh the stakeholder’s importance by using a PC (Pairwise Comparison) matrix. The use of a Pairwise Comparison matrix for the assignment of weights to stakeholders’ views is shown in Table 1. Table 1 Stakeholders pairwise comparison matrix S1 S2 S1 1 3 S2 1/3 1 S3 S3 S4 Sm w2 1 w3 1 … w4 1 Sm . 1 Σ1 Σ2 Weighting Factor w1 S4 SUM … Σ3 Σ4 wm Σm Where S1, S2… Sm, the stakeholder categories which have been identified at the stakeholders identification process. The cells of the matrix presented in Table 1 are filled with numerical values indicating the relative importance of one stakeholder with respect to others in the same raw as specified in Table 2. The diagonal is filled with “1s” as each stakeholder is compared to itself. After the decision makers complete the matrix, the determination of the weighting factors is finalised by adding up the figures in each column as represented in equation (1): Σi = Σ (α1i+α2i+...αmi), (1) The weighting factor for each stakeholder is computed as the ratio of each element from the first row and the sum for the respective column as presented in equation (1). 257 Promoting decision making through a SRAM For instance, Wi = α1i/Σi (2) Pairwise comparison matrix measurement scale Table 2 Intensity of Importance Definition 1 Equal Importance 3 Moderate Importance 5 Strong Importance 7 Very Strong Importance 9 Extreme Importance 2, 4, 6, 8 Reciprocals of above For compromises between the above In comparing elements i and j If i is 3 compared to j, then j is 1/3 compared to i System constrains identification The stakeholders weighting importance, leads to the identification of systems constrain. System constrains may include:  Remediation time frame  Future use of the site  Local community involvement  Green energy use etc. System design Once constrains have been identified the design of the remediation system can be performed. Figure 3 System’s design development (see online version for colours) In order to identify the Must to Categorize/weighting stakeholders importance To develop System constrains Decision makers System's design Source: Own illustration Sustainability indicators evaluation – Multi-objective analysis implementation The evaluation of sustainability indicators uses quantitative and qualitative metrics. Quantitative and qualitative information is processed by applying the Puch matrix process, multi-objective analysis such as the Analytical Hierarchy Processes (Saaty, 1990) based on pairwise comparisons or Simple Additive Weighting (SAW), depending upon the accuracy of the data. These approaches have been used to assess sustainability and to optimise several complex engineering systems. 258 A. Kalomoiri and W. Braida Uncertainty analysis In this stage the uncertainty of the variables involved in the decision making process is assessed and acceptable levels are agreed upon in function of the boundaries of the system and perhaps time constrains. Final list of System requirements The information gathered to this point (e.g. sustainable remediation goals, sustainable indicators, preliminary design, and acceptable levels of uncertainty) is used to develop the remedial system requirements and priorities which lead to the selection of the optimal remedial method. At this point, it is recommended to perform a sensitivity analysis in order to identify if the top ranking alternative is sensitive to changes in the weights of the sustainability indicators. Figure 4 Selection of the Optimum Sustainable Remediation method (see online version for colours) Optimum Remediation method In order to sellect the Must apply Sustainability indicators Environmental indicators Social Indicators Economic Indicators LCA (SRT, SiteWise) Followed by Multi-Objective analysis Addresing Final list of Systems requirments To enhance the Uncertainty analysis Source: Own illustration Once the optimum remediation method has been selected, the design of the site remediation is performed which leads to the implementation phase of the remedial project. Figure 5 Site’s remediation Design and Implementation. Two different opposites’ outcomes after monitoring implementation (see online version for colours) Optimum Remediation method To develop That leads to Implementation Site Remediation Design Must apply SR Goals have been achieved: leads to Site closure Source: Own illustration Monitoring SR Goals haven't been achieved: leads to Enhancements / Corrective actions 259 Promoting decision making through a SRAM During the implementation phase, monitoring is essential, to assess the effectiveness of the remedial actions with regard to the sustainable remediation objectives. Monitoring may result to two opposite outcomes: The first and the most desired one, is that the sustainable remediation goals have been achieved. That leads to site close out. After the finalisation of remedial activities, post closure monitoring may be needed for a certain period in order to monitor sustainable remediation goals. Sampling intensity will depend on the site characteristics. Sustainable remediation goals have been achieve (see online version for colours) Figure 6 Site closure Must continue Post close out Monitoring SR Goals Assessment That will provide the Source: Own illustration The second possible outcome is that sustainable remediation goals have not been achieved or that there is a substantial gap between the projected progress of the project and the actual progress. In this case, corrective actions need to be identified and implemented. Before any decision is made, the design team should start over from the systems requirements and priorities identifications stage and fast move forward to the other stages as described before in order to select the optimum corrective actions that will be implemented. Figure 7 Sustainable remediation goals have not been achieved (see online version for colours) Should sustain in the initial In order to sellect the Optimum Remediation method Final list of Systems requirments To develop That leads to Implementation Site Remediation Design Must apply SR Goals have been achieved: leads to Monitoring Enhancements / Corrective actions SR Goals haven't been achieved: leads to Site closure Must continue Post close out Monitoring That will provide the SR Goals Assessment Source: Own illustration 3 Assembling SRAM The SRAM is addressed to the decision makers. It is a technical tool that summarises the information gathered under the system analysis performed and allows the decision maker to select the most sustainable remediation method. 260 Figure 8 A. Kalomoiri and W. Braida The overall decision-making process illustration (see online version for colours) Source: Own illustration Promoting decision making through a SRAM 261 Figure 9 illustrates the SuRF-US framework to integrate sustainable practices in the design and implementation of remediation projects. Figure 9 Linear phase by phase integration of sustainability (SuRF-US: “Framework for Integrating Sustainability Into Remediation Projects” 2011 US Sustainable Remediation Forum) Source: SuRF-US (2011) At this point, all the process of stakeholder identification, the problem definition, the potential alternatives, and the list of the criteria that need to be considered for the evaluation of the alternative remediation, have been considered. The decision matrix is dependent on the number of the final criteria and the possible alternatives that has already been identified on the first phase. The process to assemble the SRAM is described in the following sections. 3.1 Evaluating remedial alternatives based on qualitative criteria The Pugh method is used to evaluate qualitative criteria. Stakeholder’s involvement is necessary at this point. Each of the chosen alternatives is compared through the qualitative criteria and sub criteria that have been selected (see Table 3). The evaluation is carried out by comparing the selected alternatives against one datum. At a remediation site, the datum is the current condition. The scale use is +1, 0 or –1 depending if the alternative is better, similar or worse than the current condition, respectively. Each of the alternatives will be assessed in the same way, populating in the Pugh matrix. Then, the total of positives and the total of negatives are recorded. The overall score will be the summation of the total positives and the total negative scores. The alternative with the larger positive overall score (comparing with the datum) is the most desired one. Moreover, most of the times, each criterion (or sub-criterion on the specific example) has different importance. At this point, stakeholders can provide a more objective decision as with their contribution, the weights of each criterion (sub-criterion) will be calculated. The addition of the weighting factors may result in the selection of a different alternative. 262 A. Kalomoiri and W. Braida Table 3 Pugh decision matrix Decision Matrix/Qualitative criteria evaluation Criteria: Ethical and equity considerations Weight factor Sub criteria Alternatives Scale Datum A1 ….. +1 ….. Social justice implementation –1, 0, 1 0 Ethical operating involved business –1, 0, 1 0 0 ….. . +1 ….. . . . ….. . . ….. Sub criteria i –1, 0, 1 0 -1 Aν ….. Total Positives Total Negatives Overall Values 3.2 Evaluating remedial alternatives based on quantitative criteria evaluation The values of quantitative criteria are calculated from LCA tools. The list of social, economic and environmental indicators is presented in Table 4 (SuRF-UK, 2010) Moreover, every indicator has six different criteria and each criterion has as many sub criteria as the stakeholders identify. An example is presented in Figure 10 for the environmental indicator. Table 4 Overarching categories of indicators for sustainability assessment of remediation options (as identified by SuRF-UK, 2010) Social Indicators Environmental Indicators Economic Indicators Human health and safety Impacts on air Direct economy costs and benefits Ethical and equity considerations Impacts on soil and ground conditions Indirect economy costs and benefits Impacts on neighbour or regions Impacts on groundwater and surface waters Employment and employment capital Community involvement and satisfaction Impacts on ecology Inducted costs Compliance with policy objectives and strategies Use of natural resources and generation of wastes Life span and project risks Uncertainty and evidence Instructiveness Project flexibility As explained before, each sub criteria may have a different weighting factor that should be taken into account at the time of the assessment as shown in the case study discussed later. The overall score for each alternative is computed by the weighted contribution of each indicator (criteria and sub criteria). Wen1 Wen2 Impact on Air w1 CO2 Emissions Impacts on Soil and Groundwater conditions w1 NOx Emissions w2 w2 Sub criteria i Change in physical, chemical Soil conditions Erosion and Soil stability w1 wi Sub criteria i Release of Contaminants Affecting suitability of water for other w2 uses Impacts on Ecology w1 Sub criteria i Consequences on flora Noise and Vibration w2 on ecology . . . wi Wen5 Wen4 Impact on Groundwater and Surface waters . . . . . . wi Wen3 Use of natural resources and generation of wastes w1 Sub criteria i Generating renewable energy by the project Water abstraction w2 and disposal . . . wi Wen6 Intrusiveness w1 Risk on flooding w2 Impacts from alterations of landforms . . . wi Sub criteria i . . . wi Sub criteria i Promoting decision making through a SRAM Environmental Indicator 263 Figure 10 Value hierarchy for evaluating SR Alternatives / Fundamental objective: Environmental Indicator (see online version for colours) Source: Own illustration enν 264 A. Kalomoiri and W. Braida At this point, the process has provided the decision maker with a specific value for each indicator for each alternative remediation method (enν, sν, ecν). The developed scoring system (SRAM), will be a 3 x  matrix, with 3 columns, representing the sustainability dimensions (Social, Economic and Environmental) and one row for each individual alternative. Figure 11 Hierarchy for evaluating SR Alternatives (see online version for colours) Goal: Sustainable Remediation Method Alternative 2 Alternative ? . . . . . . . Alternative 1 Environmental Indicator Environmental Indicator Environmental Indicator Social Indicator Social Indicator Social Indicator Economic Indicator Economic Indicator Economic Indicator Source: Own illustration Environmental Social Economic Alternative 1 Sustainability Dimensions Environmental Social Economic Alternative 2 Sustainability Dimensions . . . . Alternative v Environmental Social Economic Sustainability Dimensions 265 Promoting decision making through a SRAM The SRAM is illustrated as following: SRAM Environmental Social Economic Remediation Alternatives Sustainability Dimensions The final score for each alternative will be given from the following summation:  (en v  s  ec ) (3) where ν is the total number of alternatives being evaluated. The higher the score, the better the alternative is. For a sustainable remediation method, the weighing factor for each sustainability dimension should be equal. SRAM at this point can provide the decision makers with the transparency which allows them to evaluate the different alternatives according to the stakeholders’ priority along the sustainability dimensions. If this is the case, applying multi-objective analysis considering the different weighting factor will provide the decision makers with the optimum alternative choice. Moreover, if at this point the weighting factors on the three sustainability dimensions are not identified, applying sensitivity analysis may help decision makers in finding the optimum remedial alternative. 4 Results/case study This section exemplifies the use of SRT in gathering information for the population of environmental, social and economic matrices. The case study relates to an industrial area site 4.5 acres large located in the state of New Jersey. It has been estimated that 15,000 to 25,000 pounds of TCE and PCE were released from the annex of a former building into the subsurface, more than 25 years ago. This discharge has been the source of many soil and groundwater investigations, interim remedial measures and millions of dollars of expenditure over the years. The remedial goals for this site were to remediate PCE and TCE in shallow and deep soils in the unsaturated zone, below the New Jersey Department of Environmental Protection Soil Clean Up Criteria; to remediate groundwater to achieve the groundwater quality standards or background levels; and to remediate and or address vapour intrusion in on-site and off-site buildings. It can be argued that a sustainability appraisal of the remedial goals for the site (or even the whole environmental policy regarding clean up levels) may be granted, but this issue is outside the scope of this work. As such, an analysis of the data is performed, even though the remedial objective perhaps is not the most sustainable one. Ten model runs were completed evaluating five different technologies. The summary of the outputs are presented in the Table 5. The data were developed by Mr Trevor King from Langan Engineering. 266 A. Kalomoiri and W. Braida Table 5 Summary of Sustainability Analysis Results, derived from SRT (Sustainable Remediation Tool, USA Air Force, 2008) analysis Alternatives A1 Technologies (soil & GW) A2 Excavation & MNA A3 A4 A5 SVE (& ISCO SVE & ERD ERH In situ In situ In situ Sustainability Metrics Non-Hazardous Hazardous CO2 Emissions (tons) 103.8 103.8 470 290 710 Pounds CO2/TCE (lb/lb) 285 285 21,028 8,428 110 Total Energy (KWh) 404,000 404,000 784,000 762,000 3,100,000 Technology Cost ($) 600,000 1,770,000 2,390,000 1,150,000 350,000 Safety/ Accident Risk (lost hrs) 4.05 4.05 13.2 4.2 0.0003 Pounds TCE (lbs) 13.013 13.013 13.013 13.013 13.013 Remediation Time (yrs) <0.5-5 <0.5-5 <5 <10 <1 Table 6 Criteria been selected Criteria Sustainability Metrics Alternatives SR Indicator A1 A2 A3 A4 A5 103.8 103.8 470 290 710 C1 CO2 Emissions (tons) Environmental C2 Technology Cost ($) Economic Safety/Accident Risk C3 (lost hours) Social 4.05 4.05 13.2 4.2 0.0003 Remediation Time C4 (years) Social 0.5 0.5 3 5 1 600,000 1,770,000 2,390,000 1,150,000 350,000 Using the data presented in Table 5, the evaluation of the five different alternatives with respect to the chosen criteria was performed. The remediation time for Excavation & MNA was estimated in less than 0.5–5 years (0.5 years of excavation and 4.5 years of MNA), for SVE & ISCO and SVE & ERD remediation times were less than 5 and 10 years, respectively. Among the metrics calculated by the SRT, the ones listed in Table 8 were chosen for the sustainability analysis as they closely match SURF-UK indicators (Table 4). In order to find which is the most efficient remediation technology considering environmental, economic and social indicators, we need to calculate the overall values of the selected Alternatives. To do so, the first step was to determine the weights for each output value from LCA implementation for each alternative. Assuming that each criterion varies linearly and as all of them have negative impact, we use the Simple Additive Weighting (SAW) (Zarghmi and Szidarovskzky) method to determine the values. The data was normalised into the interval [0, 1] by using the linear transformation described in equation (4) (Zarghami and Szidarovszky, 2011): aij  aij  M i mi  M i (4) 267 Promoting decision making through a SRAM Where mi and Mi, the computed minimum and maximum values of each criterion, for i = 1,2,3,4 and j = 1,2,3,4,5 (Table 5), respectively. Results of the transformation are presented in Table 7. Normalised overall values for alternatives Table 7 Criteria Weights Alternatives A1 A2 A3 A4 A5 C1 0.33 1 1 0.396 0.693 0 C2 0.33 0.877 0.338 0 0.608 1 C3 0.23 0.693 0.693 0 0.682 1 C4 0.1 1 1 0.444 0 0.888 Si: Overall Scores for Alternatives 0.879 0.689 0.175 0.586 0.649 Si Normalised Overall Scores for Alternatives 0.294 0.232 0.057 0.198 0.220 Following, the overall scores for alternatives was determined using equation (5) (Zarghamiand and Szidarovszky, 2011) where wi, are the weight factors for each criterion. In this regard, a generic approach was used and the same weighting factors, 0.33, were selected for the three sustainability indicators. For C3 and C4 criteria, we use 0.23 and 0.1 respectively, considering that safety has a greater weight than the technology’s time implementation. Finally, the normalised overall scores for alternatives were computed, using the equation (6). Si  Si  n  wi aij (5) Si  Si (6) i 1 The results in Table 7 suggest that the best choice is the first alternative, as it gives the smallest weighted average satisfaction value. The alternatives’ ranking that was obtained from best to the most undesired is as follows: A1>A2>A5>A4>A3 A sustainable remediation decision tool should be able to guide the user in gathering the appropriate information and present the associated trade-offs in a manner that allows the decision maker to reach a conclusion that balances those trade-offs. Furthermore, there are cases which have specific constrains that also need to be taken into account. Those constrains might lead the decision maker into an optimum solution which is not the most sustainable one. The visual presentation of the Table 7 is displayed in Figure 12. The performance sensitivity graph shows the relative importance of each of the criteria as bars, and the relative preference for each alternative with respect to each criterion. The overall alternative preferences (the normalised overall score) are presented in the foremost right column. On Figure 12, one can see that alternative A3 performance is worse than 268 A. Kalomoiri and W. Braida alternatives A1 and A2 for the four criteria analysed and worse than alternatives A4 and A5 for three of the four criteria. Thus, this alternative cannot affect the total top ranking even if the criteria weights change. The same can be assumed for alternative A4. Figure 12 Performance Sensitivity graph under sustainability umbrella (see online version for colours) Source: Own illustration, based on Expert Choice software A sensitivity analysis was performed using Expert Choice software (http://expertchoise. com), in order to determine the impact of weighting on the ranking of the top alternatives. Figure 13 shows that the top ranking is not sensitive to a change in the criteria weight. In order to place the alternative A5 to the top rank, C2 criterion weighting factor should be more than 80% (or 0.8 depending on the scale). That automatically reduces the weighting factors for the rest of the criteria to negligible values (see Figure 13). Thus, A5 could be an optimum solution for a specific contaminated site dealing with harsh budget constraints, but does not satisfy sustainability criteria. Figure 13 Performance Sensitivity graph manipulating C2 criterion for changing the top ranking of alternatives A1 and A5 (see online version for colours) Source: Own illustration, based on Expert Choice software Promoting decision making through a SRAM 5 269 Conclusions Our planet has undergone significant deterioration during the last decades. Today, more than ever, it is necessary that all human activities be carried out under the framework of sustainability. Moreover, the demand of decision tools in order to assess sustainability practices in general and sustainable environmental remediation approaches in particular, has increased. An important conclusion from the real case study is that the final decision in the selection of a remedial approach may change depending upon criteria weights. Thus, criteria weighting is the most important stage of developing an assessment tool and stakeholder involvement in this direction is important and very critical in providing an objective, balanced, and socially acceptable decision that at the same time eliminates uncertainties. Acknowledgements We gratefully acknowledge the QUESTOR Centre for funding the research project “A Process for Developing a Sustainable Remediation Assessment Matrix (SRAM)”. Moreover, we would like to like to acknowledge the contributions of Professor Jonathan Smith from Shell Global Solutions (UK) for his very helpful comments and Mr. Trevor King from Langan Engineering, who provided the data, used in the evaluation the different remediation methods by SRT. References Braida, W. and Ogundipe, A. (2010) ‘A process for developing a Sustainable Remediation Assessment Matrix (SRAM)’, Paper presented at the X Protection and Restoration of the Environment International Conference, 5–9 July 2010, Corfu, Greece. Expert Choice Software 11.5. Available online at: http://expertchoice.com/ (accessed on 1 November 2012). Gernot, D., Moritz, O. and Werner, F. (2013) ‘Introducing a goal-oriented sustainability assessment method to support decision making in contaminated site management’, Journal of Environmental Science & Policy, Vol. 25, pp.207–217. Saaty, T. (1990) ‘How to make a decision: the analytic hierarchy process’, European Journal of Operating Research, Vol. 48, No. 1, pp.9–26. Sussman, J., Dodder, R., McConnel, J., Mostashari, A. and Sgouridis S. (2007) The CLIOS Process, A User’s Guide, Massachusetts Institute of Technology. Scheuermann, K. (2009) ‘Green remediation – estimating the environmental footprint at a corrective action clean-up, pilot study at Romic East Palo Alto’, Paper presented at the 19th Annual NARPM Training Conference, 2–5 June 2009, Atlanta, Georgia, USA. Available online at: http://www.epa.gov/region9/waste/romic-eastpaloalto/pdf/Romic-GreenRemediation-Pilot-Study-NARPM.pdf (accessed on June 2011). SRT (Sustainable Remediation Tool, USA Air Force, 2008). Available online at: http://www. afcee.af.mil/resources/technologytransfer/programsandinitiatives/sustainableremediation/srt/in dex.asp (accessed on 3 September 2011). 270 A. Kalomoiri and W. Braida Sustainable Remediation Forum UK (SuRF-UK) A Framework for Assessing the Sustainability of Soil and Groundwater Remediation. Available online at: http://www.claire.co.uk/index. php?option=com_phocadownload&view=file&id=61:initiatives&Itemid=78 (accessed on March 2010). Holland, K., Lewis, R., Tripton, K., Karnis, C., Petrovskis, E., Bull, L., Taege, D. and Hook, C. (2011) ‘Sustainable Remediation Forum US (SuRF-US) (2011) Framework for Integrating Sustainability Into Remediation Projects’, Willey Online Library, Vol. 21, No. 3, pp.7–38. Available online at: http://www.sustainableremediation.org/library/guidance-tools-and-otherresources/ Zarghmi, M. and Szidarovskzky, F. (2011) ‘Simple Additive Weighting (SAW)’, Multi-criteria Analysis – Applications to Water and Environmental Management, Springer-Verlag, Berlin, Heidelberg, ISBN: 978-3-642-17936-5, pp.26–29.