7
A Platform for ReIDMP

7.1 Introduction

This research proposed the Resilient-Informed Decision-Making Process (ReIDMP) platform for developing resilient city systems. The ReIDMP platform consists of two practical approaches: first is a simulation computer game, SimCity (2013)®, deployed under the concept of serious gaming. The second approach is the analysis and assessment that must take place to address risk and vulnerability. Risk and vulnerability are integrated into evaluation guidelines and methodologies provided by the International Atomic Energy Agency (IAEA) and the Science Applications International Corporation (SAIC). This initiative has evolved to enable stakeholders to conduct various risk and vulnerability assessment (VA) based on their own design in a simulated city in a real-world situation to enhance the efficacy of learning experiences relating to the knowledge of risk and vulnerability. The chapter expounds on the approaches undertaken to understand the simulation mechanism of SimCity® and the productive application of its visualization for sophisticated investigation.

7.2 Design and Development of City Simulation Using SimCity (2013)®

To ensure the sustainment and resilience of existing and next-generation critical infrastructure (CI) systems, decision makers and stakeholders must be well-informed about past and current problems (Vasuthanasub 2019). Decision makers and stakeholders also need the knowledge and tools to help them analyze, assess, and design a viable implementation plan for long-term development. The result or output of studies and research must be able to apply to the decision-making process and resource allocation for solving real-world problems. Figure 7.1 depicts the structured methodology (i.e. platform) for analyzing the technical strategies and actions required for CI system resilience. Sections 7.2.1–7.3.4 describe the four phases of the ReIDMP platform/ model, which can serve as a standard practice when developing resilient city systems.

Figure 7.1 A framework of the ReIDMP model.

7.2.1 Phase I: Project Planning and Management

Project management is a systemic process of planned and organized endeavor to meet a specific requirement, achieve a particular objective, or accomplish a specific goal (Meredith et al. 2013). A body of knowledge usually refers to developing critical phases under management science, including initiating, planning, executing, controlling, and closing. Project management is not trivial as it informs activities that are focused on achieving expected results effectively for individuals, organizations, or society. With this in mind, successful project planning and management are another two critical components of success.

Whether for a simple or unique project – like launching a new product or constructing a megastructure – decision makers, stakeholders, project managers, or team members need clear direction, organized operational procedures, consistently streamlined processes, and a user-friendly approach to maximize their full potential. Frequently, many projects have suffered from inadequate planning, inaccurate information, and unclear roles and responsibilities due to a lack of a decent starting framework. The consequences of those poor performances can be severe or unacceptable (Katina 2020; Stumpe and Katina 2019). Therefore, success does not come easily in project management. However, being better prepared for “what will be coming next?” is not always impossible. Project KickStart Pro 5 is a project management-based software that incorporates intuitive planning and managing processes, offering users a smooth sequence of functions to develop a project plan with precise descriptions and a deeper understanding of goals, obstructions, risks, and solutions. Users can define roles and responsibilities with their features and capabilities and assign each team member tasks accordingly. Then, a planner can perform essential project management tasks, such as assignment tracking, project costs, deadlines, or even personalized objectives. However, the whole team needs to have a say in a project’s key milestones and it’s execution plan.

Overall, Project KickStart Pro 5 helps a project manager to create an innovative, efficient, and organized plan from start to finish. The software can be helpful in one way or another, but in particular it will get project teams on board and unmistakably keep all members on the same page without leaving confusion about roles, responsibilities, expectations, and deadlines.

7.2.2 Phase II: Learning by Doing Through Gaming Strategy

While expertise in risk and vulnerability management of complex interdependent systems is not necessary, it is essential to know the basics of risks and vulnerabilities, infrastructure systems, and gaming. This base knowledge can also include energy generation and distribution or public transportation systems; a learning goal of risk and VA materials is to analyze potential risks and their impacts due to adverse events or pollution activities. Moreover, by fully reviewing and understanding those presented activities within a city or region, stakeholders (be it students or otherwise) should be able to determine the societal risks and asset vulnerabilities for individuals living in the given area. Figure 7.2 depicts Phase II elaborating on using computer game technology to enhance stakeholder engagements and practical experiences in quantifying the resiliency of infrastructure systems.

Figure 7.2 Deployment of the SimCity® application.

The following proposed procedures utilize a combination of serious gaming concepts in conjunction with risk analysis and VA techniques. Three specific documents are referred to here. First is the Manual for the Classification and Prioritization of Risks Due to Major Accidents in Process and Related Industries. This manual aims to introduce methods and procedures to classify hazardous activities in a region or city of interest by categorizing consequences and probabilities of occurrence. An assessment applies to the risks due to major accidents with off-site consequences in fixed installations handling, storing, and processing dangerous substances and transporting hazardous materials by road, rail, pipeline, and inland waterway (International Atomic Energy Agency 1996). Figure 7.3 is based on this document and depicts the procedural steps of rapid risk assessment.

Figure 7.3 Procedural steps of rapid risk assessment.

The second is A Guide to Highway Vulnerability Assessment for Critical Asset Identification and Protection. The adoption of this guidebook presents the standard procedures for assessing the vulnerabilities of physical assets, such as roadways, highways, tunnels, bridges, and inspection and traffic operation facilities, among others. It provides a holistic set of steps to identify and mitigate the consequences of transportation routes from terrorist threats or attacks (Science Applications International Corporation 2002). Figure 7.4 is based on this document and depicts the procedural steps of vulnerability assessment.

Figure 7.4 Procedural steps of VA.

The third is the Guidelines for Integrated Risk Assessment and Management in Large Industrial Areas. The purpose of using this document is to provide practical guidance and a technical reference for the proceeding of integrated health and environmental risk assessment studies and environmental management strategies in large industrial areas. The methodologies and techniques are best suited to geographical regions that accommodate several industrial and related activities of a hazardous and/or polluting nature (International Atomic Energy Agency 1998). Figure 7.5 is based on this document and depicts the procedural steps of VA.

Figure 7.5 Procedural steps of integrated regional risk assessment.

Again, SimCity (2013)® is a simulation game of organized play, which has a set of stated rules and purposes to guide players to the goals, and the goals can be either fanciful or purposeful. If the requirements are specified and supervision is provided to enable focus on learning outcomes, while preserving playfulness, then a serious learning experience is possible. With this intention, a simulation will allow players to play with complex computer models, interact with each other, and experience revolutionary or evolutionary changes. Ideally, this approach must be made before deciding or implementing a policy. And since this is a simulation, there are no effects on the real world. It also intends to benefit stakeholders – they can better understand and manage infrastructure resilience.

Moreover, researchers have seen real benefits, even with university students. For example, an experimental design was prepared and implemented by the authors for one year with students in a graduate course at Old Dominion University. SimCity® was included in the course syllabus as a required tool and other course materials in Spring 2015. During each experimental session, students, under the supervision of the senior faculty researcher, developed a city under the foundational concepts of sustainable development and infrastructure resiliency. Later that year, the original version of the experimental design was revised and upgraded as a result of a funded Faculty Innovator Grant Program, with the goal of creating a gaming laboratory to enable learning (Vasuthanasub 2019).

7.2.3 Phase III: Multi-Criteria Decision Analysis

Theoretically, decision-making can be considered a logical thinking process for selecting suitable alternatives under different criteria or factors. In the process of decision analysis, decision makers and stakeholders must assess all possible positive and negative impacts and should also be able to judge the consequences of that decision. Multi-criteria decision analysis (MCDA) or multi-criteria decision-making (MCDM) is a framework of analytical techniques that helps decision makers choose between multiple options when multiple objectives have to be pursued. It is a systematic and pragmatic process involving various criteria in rational decision-making. MCDA allows the users to determine a selection as valuable and as efficient as possible, to preserve a degree of consistency within the search, or at least to provide the inconsistencies, if they occur, without imposing unnecessary variables and unjustifiable structure in the decision-analysis process (Stewart 1992). With this in mind, the primary purpose of using MCDA is to assist the decision maker in discovering the most preferred solution to a problem. The application has widely supported many complex decision dilemmas, such as in business, government, and medicine, especially in situations with conflicting criteria (Belton and Stewart 2002).

There are many methods for implementing the MCDM approach. In this research, the decision models of multiple criteria, attributes, and alternatives are structured and presented with the implementation of two distinct MCDA methods through decision support system (DSS) tools.

7.2.3.1 Multi-Attribute Utility Theory

Deciding on complex problems in the real world is challenging; decision makers are often challenged by multiple objectives, various stakeholders, future or long-term consequences, and risk and adverse effects (Nikou 2011). In particular, when the problems are bound with the constraint, namely uncertainty, one of the most effective forms of approach for dealing with this type of problem is using “expected utility.” In MCDA, this technique is formally called multi-attribute utility theory, or MAUT (Li et al. 2009). It was invented to ease the difficulty in solving the problem under uncertainty and aims to answer the question: Which alternative is the best option? The method offers users the steps to assign scores and to compare possible alternatives in which, afterward, all of them can be identified and analyzed. It also allows a group of stakeholders to search and examine the consequences in different ways for evaluating the options (Nikou 2011).

A core concept of MAUT is based on the theoretical foundation that the decision maker’s preferences can be transformed and represented by a function called the utility “U” (Ishizaka and Nemery 2013). This utility function is being used to replace the value associated with each criterion for providing a degree of satisfaction to the decision maker. The MAUT method works best when a decision maker consciously tries to optimize the performance of alternatives under a set of conditions and points of view (Ishizaka and Nemery 2013). Thus, the preferred desirability of a particular alternative depends on how its associated attributes are being considered and judged. Similar to other MCDA methods, MAUT methodology consists of four key steps:

  • Constructing a decision problem by specifying the objectives and identifying the attributes that need to be measured.
  • Setting up the alternatives and exploring the potential consequences caused by each of them in terms of the attributes identified.
  • Determining the preferences of the decision maker and stakeholders and assigning the weight of attributes reflecting their importance to the decision.
  • Synthesizing the results by assessing the impact of a certain criterion of the decision and lastly comparing the alternatives.

In this research, a DSS tool, called Logical Decisions for Windows (LDW), will be utilized to assist the processes of problem modeling, calculation, and analysis. LDW is the MAUT-based decision-support software package that helps a decision maker and stakeholders to evaluate and select the best option under difficult circumstances and restricted constraints. A package combines features between spreadsheet and database programs that allow a user to organize data and information for all possible prospects at the same time.

The following are the basic required steps in completing the decision model and decision analysis with LDW:

  • A decision maker can initialize the process by either manually inputting hard (number) and soft (detail or explanation) data or electronically importing prepared spreadsheets and database files. The software will convert an ordinary data table into a sophisticated hierarchical structure and link all the detailed information and specific preferences to overall goals. It can simply turn a set of incomprehensible numbers into a detailed roadmap that guides a decision maker to achieve the best outcome.
  • Using the MAUT, the value judgments are a crucial stage of the process. LDW makes this phase easier by providing a variety of methods, such as Smarter (easy-to-use), Tradeoff (sophisticated), and AHP (popular), for making the judgments and assigning the weights. The user can select and use one method based on the appropriateness of the decision model.
  • For the final step, the presentation of analysis results is designed to provide insights with interactive displays. A decision maker can rank the best to the worst alternatives, associating any goal or evaluation measures and comparing a particular option against the others to understand the differences.
  • Significantly, a group of stakeholders can also revise the weight of importance to assume the effect of changes in the overall ranking results and conduct a sensitivity analysis to see the effects of uncertainty on the ranking results.

Last but not least, the MAUT method is usually adopted in a group situation. Frequently, a necessary level of detail and specification during a discussion on the determination of attributes and their weights can turn into conflicts or arguments rather than moving toward common ground. This restriction may result in wasting time or even worse. So much so that to implement MAUT through LDW effectively, a decision maker and all stakeholders must be able to agree on a set of attributes and a range of weights to be used in a model.

7.2.3.2 Evidential Reasoning

When dealing with problems under fuzzy weights and utilities, Zhou et al. (2010) suggest that “various types of criteria must be taken into account, which may be quantitative, measured by numerical values with certain units, or qualitative, assessed using subjective judgments with uncertainties.” Consequently, this research employs evidential reasoning (ER) as a second MCDA method to tackle problems of uncertainty and subjectivity. The technique is the latest development and a technical breakthrough in handling hybrid MCDA problems, which can be in either hierarchical or non-hierarchical conditions (Xu and Yang 2001). Essentially, the ER approach is different from other conventional MCDA methods. Its algorithm was developed based on the fundamental principle of decision theory and the combination rule of Dempster–Shafer evidence theory (Xu and Yang 2001; Yang 2001; Zhou et al. 2010). When solving a problem with this method, the weights of importance are not necessary for assembling attributes in the model. Instead of aggregating average scores, it uses a belief structure, formally called the “degrees of belief,” to represent an assessment and to reach a conclusion as a distribution. This belief degree is the level of expectation that can be purposely managed to obtain a decision maker’s preferences.

Recent developments in ER algorithms have resulted in DSS tools, including the Intelligent Decision System, or IDS (Xu et al. 2006; Xu and Yang 2003; Yang and Xu 2002; Xu and Yang 2001; Yang and Xu 2004). IDS is an ER-based decision-support software designed to assist large-scale MCDA problems. The application has a unique capacity to handle thousands of attributes. IDS is also considered a flexible and versatile tool since it can be applied to various types of information, like deterministic numbers, random numbers, and subjective judgments in different formats (Xu et al. 2006; Xu and Yang 2003; Vasuthanasub 2019).

To support a decision-making process with a standard ER approach through IDS, Xu et al. (2008) suggest five required basic steps: (i) model implementation, (ii) data collection, (iii) group information and opinion, (iv) assessment aggregation, and (v) analysis results presentation.

  • Problem modeling: a model implementation in IDS implies identifying the alternative courses of action, or simply choices, criterion weights, and evaluation scales for assessing selections on criteria (Xu et al. 2008). Using the software, the construction of an assessment criterion tree is straightforward. After a tree is constructed, a user needs to define each criterion accordingly. At this point, it is important to note that the IDS software was designed to use the five-point grading scale as a standard estimation for assessment procedures, which means all criteria are assessed as qualitative measurements. The highest grade is five. For this reason, whether it is a quantitative or a qualitative criterion, descriptions are required for both types of criteria. Meanwhile, the number of grades or points must be included as well. In addition to a default setting in IDS, relative attribute weights can be assigned through either pairwise comparison or an interactive chart, where the bars can be instantly dragged up and down to adjust the desirable criteria weight. A function is considered a useful feature since a user can observe the differentiation of individual viewpoints or group decision standpoints on criterion importance early on. At the end of this step, a user may administer an assessment model to individual participants to initiate the next phase.
  • Data collection: practically, the assessment in this step involves a number of activities, such as evidence collection, comparison, judgment resolution, and determining the grade (Xu et al. 2008). With a completed problem structure in model implementation individual participants can assess each option and record their scores and opinions. Each of them needs to verify the grades, then the ER algorithm will automatically generate a degree of belief next to each answer. This degree of belief represents the strength level of describing an alternative to the criterion. Yet, in our research, there was not enough detail or information for the participant group to make accurate judgments; this is the case where there isn’t a clear majority. Participants could only select a grade that was most appropriate based on their knowledge without worrying about the distribution of the belief degrees (Xu et al. 2008). IDS will handle a process by equally dividing 100% of the degrees of belief and automatically allocating them to the checked items.
  • Group decision support: when utilizing IDS, individual participants may independently record and anonymously register their assessments of each alternative to prevent the risk of potential disagreement among them. Consequently, individual lists can be either separately reviewed or privately imported as a single file. After collating all inputs, a user will have two options to select from: comparing assessment data via a function of graphical representations or generating collective assessment information for each alternative (Xu et al. 2008).
  • Assessment aggregation: again, the aggregation process of assessment information from lower-level to higher-level criteria is analyzed through the ER algorithm. With IDS functionality, this compilation is automatic and updated in real-time in the background whenever either initial assessment data is modified or an additional detail is entered for any criterion (Xu et al. 2008). Thus, a user should be able to obtain the original outcomes promptly if nothing is changed or to see the updated reports at any stage of revision, even before the assessment on some criteria is finished.
  • Assessment results presentation: IDS can generate different types of analysis results in graphical formats to support decision communication, such as performance ranking, performance score range, and performance distribution. However, while the ranking and score graphs present general results information, the distribution chart provides more insight. The selected alternatives can be compared to each other through all those diagrams and distinguished from any preferred areas in different levels of the assessment hierarchy (Xu et al. 2008). Moreover, the properties and appearances of those graphs can be configured and then exported to MS Word documents or MS PowerPoint files. The software also provides a search function to help a decision maker identify strengths and weaknesses. Afterward, a model can be used to study the effect of action plans when simulating various improvement scenarios.

7.2.4 Phase IV: Object-Oriented Programming

In computer science engineering, object-oriented programming (OOP) is called “structured programming.” It involves a set of procedural programming that administers a logical structure in the program being written to allow it greater efficiency while being easier to understand and modify. OOP is described by Eck (2014) as the approach of structured programming thus:

To solve a complex problem, break the problem into several pieces and work on each piece separately; to solve each piece, handle it as a new problem respectively which itself can be broken down into a smaller piece of problems again; eventually, you will work your way down to problems that can be solved directly, without further decomposition.

However, the OOP paradigm may best be addressed by its corresponding alternate, Object-Oriented Analysis and Design (OOAD). OOAD was derived from adopting the fundamental concept of OOP, and it may be adequately concluded that it is an analytic method that illustrates an information system by identifying things called “objects.” In this case, an object represents a real person, place, event, or system. The end product of object-oriented analysis is an object model, which represents the information system regarding the purpose and object-oriented concepts. It gives researchers an easier way to express or present the important pieces of information and essential features of the application to the stakeholders than any other approach does.

Booch et al. (2007) indicate four main processes for characterizing an object-oriented model’s conceptual framework: abstraction, encapsulation, modularity, and hierarchy.

  • Abstraction/discovery: the first technique refers to a simplified representation of a system. One that captures only those relevant characteristics or essential aspects with regard to the perspective of the researchers.
  • Encapsulation/visualization: this component refers to the hidden details in an object’s internal composition and work. Encapsulation acts as a protector to limit user access to an object’s internal data. Also, it offers a means to reduce system complexity (Booch et al. 2007; Pulfer and Schmid 2006).
  • Modularity/mapping: this process involves partitioning objects into sub-objects called modules. Especially for complex and large-scale system design, thus modularity helps in managing complexity by disintegrating a huge intractable solution into smaller and more manageable ones, which are interconnected and composed of the required large-scale solution.
  • Hierarchy/model and analysis: the last step refers to rank allocation. Since encapsulation manages the hiding of detail or the prioritization of the relevant details to understand the problem at hand, it is necessary to have a hierarchy, as different levels of detail may be required to solve problems (Booch et al. 2007).

OOP can be applied using a tool called TopEase® Designer. TopEase® is a software product that was designed to allow organizations to model the impact of adverse events on every dimension of their business, understand which elements need to be restored within what timeframe, create and maintain organization disaster response plans, and identify any gaps where response plans do not yet exist. In the same way, all its capabilities, including discovery, visualization, mapping, and particularly model and analysis, enable the researchers to design and model a complex or large system structure and then perform a comprehensive system analysis. Figure 7.6 depicts the main window of TopEase®.

Figure 7.6 TopEase®’s main window.

The original developers of TopEase® suggest that it can be viewed as a combination of comprehensive methodology and integrated tool (Pulfer and Schmid 2006). Moreover, TopEase® and TopEase® XBench offer complete support for sustainable corporate development. The TopEase® approach is probably the most comprehensive framework available for corporate engineering. The traditional approach comes with a box of a wide variety of diverse and (if at all) most loosely coupled and integrated techniques. The framework has an open design and enables other process models to be embedded into the model or the framework after modeling, like Catalyst, Prince, V-Model, Hermes, RUP, etc. (Pulfer and Schmid 2006). The approach is also integrative in that it allows for managing the whole life cycle, from the first development iteration to operations, disposal of the initial system, and the next iteration cycle. The framework is integrated into operations (via the development of systems and control of workflow engines). It provides corporate management with the company’s current data (KPI/risk-/compliance-control) in the form of measurements, hence supporting sustainability concepts.

7.3 Concluding Remarks

This chapter outlines the phases of a platform (ReDMP) necessary to create resilient cities. First, project planning and management emphasize the basics of project and project management. Second is the issue of learning where gaming strategies are necessary, especially if one attempts to address the risk and vulnerability of complex interdependent CI systems without impacting physical, economic security, public health, and safety. Third, there is a need for MCDA and other approaches supporting decision-making, involving alternatives selection under different criteria or factors. Finally, and due to complexity, it is necessary to break the problem into several pieces and work on each part separately while considering the whole; a capability provided by OOP.

However, offering a cautionary tale about the articulated platform/model is essential. A familiar aphorism in statistics is that “all models are wrong the scientist cannot obtain a ‘correct’ one by excessive elaboration” (Box 1976, p. 792). This chapter postulates that this is a fair criticism of engineered models. Moreover, it is suggested that the developed ReIDMP model carries assumptions that potential users must understand. Nevertheless, these assumptions do not make the model “useless.” Rather, the model is useful in the “context” in which it is developed. And those that understand the model can always improve it. Someone who has experience of studying a map will be able to use maps with ease and yet “A map is not the territory it represents, but, if correct, it has a similar structure to the territory, which accounts for its usefulness” (Korzybski 1994, p. 58). Similarly, the ReIDMP model, while wrong, can be useful in the hands of a skilled modeler. In fact, Chapter 8 shows how a skilled modeler can use the proposed model to perform risk and VA in a game-like situation.

7.4 Exercises

1 Discuss how ReIDMP helps in the development of a resilient city.

2 Discuss how project management influences the ReIDMP platform.

3 What areas of assessment can be developed to enhance the ReIDMP platform?

4 Apply the ReIDMP to the city of your choice.

5 Feedback is not “explicit” in ReIDMP. Discuss why it is “implied” in the ReIDMP platform?

References

  1. Belton, V. and Stewart, T. (2002). Multiple Criteria Decision Analysis: An Integrated Approach. Kluwer Academic Publishers.
  2. Booch, G., Maksimchuk, R.A., Engle, M.W. et al. (2007). Object-Oriented Analysis and Design with Applications (3rd ed.). Addison-Wesley Professional.
  3. Box, G.E.P. (1976). Science and statistics. Journal of the American Statistical Association 71(356): 791–799. https://doi.org/10.2307/2286841.
  4. Eck, D. J. (2014). An Introduction to Programming Using Java (7th ed.). Hobart and William Smith Colleges.
  5. International Atomic Energy Agency. (1996). Manual for the Classification and Prioritization of Risks Due to Major Accidents in Process and Related Industries. http://www-pub.iaea.org/books/IAEABooks/5391/Manual-for-the-Classification-and-Prioritization-of-Risks-due-to-Major-Accidents-in-Process-and-Related-Industries.
  6. International Atomic Energy Agency. (1998). Guidelines for Integrated Risk Assessment and Management in Large Industrial Areas. http://www-pub.iaea.org/books/IAEABooks/5391/Manual-for-the-Classification-and-Prioritization-of-Risks-due-to-Major-Accidents-in-Process-and-Related-Industries.
  7. Ishizaka, A. and Nemery, P. (2013). Multi-Criteria Decision Analysis: Methods and Software (1st ed.). Wiley.
  8. Katina, P.F. (2020). System acquisition pathology: A comprehensive characterisation of system failure modes and effects. International Journal of Critical Infrastructures 16(3): 255–292. https://www.inderscienceonline.com/doi/abs/10.1504/IJCIS.2020.108499.
  9. Korzybski, A. (1994). Science and Sanity: An Introduction to Non-Aristotelian Systems and General Semantics. Wiley.
  10. Li, H., Apostolakis, G.E., Gifun, J. et al. (2009). Ranking the risks from multiple hazards in a small community. Risk Analysis 29(3): 438–456. https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1539-6924.2008.01164.x.
  11. Meredith, J.R., Mantel, S.J., Shafer, S.M. et al. (2013). Project Management in Practice (5 ed.). Wiley.
  12. Nikou, T. (2011). Application of multi-attribute utility theory to heating and cooling system selection for a sustainable building project. Thesis, Clemson University. https://tigerprints.clemson.edu/all_theses/1114.
  13. Pulfer, R. and Schmid, U. (2006). Control Your Business: The Balance Between Principles and Pragmatism. Pulinco.
  14. Science Applications International Corporation. (2002). Guide to Highway Vulnerability Assessment for Critical Asset Identification and Protection. Transportation Policy and Analysis Center.
  15. Stewart, T. (1992). A critical survey on the status of multiple criteria decision making theory and practice. Omega 20(5): 569–586. https://doi.org/10.1016/0305-0483(92)90003-P.
  16. Stumpe, F. and Katina, P.F. (2019). Multi-objective multi-customer project network: Visualising interdependencies and influences. International Journal of System of Systems Engineering 9(2): 139. https://www.inderscienceonline.com/doi/abs/10.1504/IJSSE.2019.100338.
  17. Vasuthanasub, J. (2019). The resilient city: A platform for informed decision-making process. Dissertation, Old Dominion University. https://digitalcommons.odu.edu/emse_etds/151.
  18. Xu, D.-L., McCarthy, G., and Yang, J.-B. (2006). Intelligent decision system and its application in business innovation self assessment. Decision Support Systems 42(2): 664–673. https://doi.org/10.1016/j.dss.2005.03.004.
  19. Xu, D.-L. and Yang, J.-B. (2003). Intelligent decision system for self-assessment. Journal of Multi-Criteria Decision Analysis 12(1): 43–60. https://doi.org/10.1002/mcda.343.
  20. Xu, D.L., Yang, J.B., Carle, B. et al. (2008). Application of an intelligent decision system to nuclear waste repository option analysis. International Journal of Nuclear Governance, Economy and Ecology 2(2): 146. https://www.inderscience.com/info/inarticle.php?artid=18333.
  21. Xu, L. and Yang, J.-B. (2001). Introduction to multi-criteria decision making and the evidential reasoning approach (Working Paper No. 106). University of Manchester Institute of Science and Technology.
  22. Yang, J. and Xu, L. (2004). Intelligent decision system for supplier assessment. DSS2004: The 2004 IFIP Conference on Decision Support Systems. Monash University. https://www.research.manchester.ac.uk/portal/en/publications/intelligent-decision-system-for-supplier-assessment(999748c6-ceb6-403c-8a9b-8ed874ee3724)/export.html.
  23. Yang, J.-B. (2001). Rule and utility based evidential reasoning approach for multiattribute decision analysis under uncertainties. European Journal of Operational Research 131(1): 31–61. https://doi.org/10.1016/S0377-2217(99)00441-5.
  24. Yang, J.-B. and Xu, D.-L. (2002). On the evidential reasoning algorithm for multiple attribute decision analysis under uncertainty. IEEE Transactions on Systems, Man, and Cybernetics – Part A: Systems and Humans 32(3): 289–304. https://doi.org/10.1109/TSMCA.2002.802746.
  25. Zhou, M., Liu, X.B., and Yang, J.B. (2010). Evidential reasoning based nonlinear programming model for MCDA under fuzzy weights and utilities. International Journal of Intelligent Systems 25(1): 31–58. https://onlinelibrary.wiley.com/doi/10.1002/int.20387.
..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset