Evaluation and Decision Models with Multiple Criteria

This chapter is about the decision aiding process. In professional contexts, there are cases of decision problems which require using formal processes and methods. In the first part of the chapter, we identify and describe the essential steps of a decision aiding process. In the second part, we discuss four practical questions that have to be tackled by an analyst in charge of a decision aiding process.

This chapter deals with a crucial step in the decision aiding process: the aggregation of the alternatives’ performances on each criterion in order to faithfully model the overall preference of the decision maker. The approach we follow is that of conjoint measurement, which aims at determining under which conditions a preference can be represented in a particular aggregation model. This approach is first illustrated with the classical additive value function model. Then, we describe two broad families of preference models, which constitute a framework encompassing many aggregation models used in practice. The aggregation rules that fit with the second family of models rely on the aggregation of preference differences. Among this family we find, in particular, models for the outranking relations (concordance relations with vetoes) that are used in several case studies in this book.

This chapter briefly presents a number of techniques that can be used to build recommendations in each of three classical problem statements (choosing, ranking, and sorting) on the basis of a preference model. We start with the simple case of a preference model based on a value function. We then turn to more complex cases.

The chapter concerns the attribution of the EURO Best Poster Award at the 20th EURO Conference, held in Rhodes, July 2004. We present the historical decision making process leading to the selection of the winner, followed by a thorough discussion of the constructed outranking models and of the best choice recommendation.

The present paper seeks to propose a general framework to composite Web services selection. The proposed framework extends the conventional Web services architecture by adding a new component in the registry devoted to multicriteria classification of compositions into different ordered Quality of Service (QoS) classes. This additional component takes as input the specification of the desired service, a set of functional and non-functional evaluation criteria, a set of QoS-ordered classes, and a set of preference parameters, and generates as output a classification of composite Web services into different QoS-ordered classes. In addition to the description of the proposed framework, the paper proposes solutions to construct, evaluate and classify composite Web services. The paper also briefly presents the developed prototype and then illustrates and discusses some computational aspects of the proposed framework using numerical data.

In this paper we present a case study of a real-life multicriteria decision making problem of choosing the site for a university kindergarten in Madrid. The largest private university of Madrid, San Pablo CEU, needs to build a kindergarten for staff children. This study consists of two phases. In the first phase, an approximate model was presented to the decision makers in order to motivate re-activating the process. In the second phase, a more detailed model with new alternatives was introduced. The criteria measurements as well as the preferences contain large uncertainties. Therefore, the problem is solved by using the SMAA-III method that allows to model uncertainties through joint probability distributions.

This chapter reports on a simulated decision aiding process aiming at choosing a cooling system for a new power plant. The study took place in a research contract between MathRO, the Mathematics and Operations Research unit at UMONS, and Laborelec, a technical competence centre of the Belgian power industry. The goal of the contract was to show whether and how multiple criteria decision methods could help taking the environmental and societal aspects of technical choices into account in a relevant way. In this perspective, the case of the choice of a cooling system for the Saint-Ghislain power plant was conceived as a validation exercise.

The purpose of this chapter is to present the main results of a decision aid process which we have applied to help policymakers for the localization of a wind farm in Corsica (Western Mediterranean Island, French region). This process has been applied in a participatory context of multiple stakeholders open to the public. The first section presents a brief introduction about this real case study ordered by the Economic Development Agency of the Territorial Collectivity of Corsica. The second section is devoted to the study context and the decision aid process. The third section describes the simulated projects of wind farms and the family of evaluation criteria. The fourth section deals with the performance table and the recommendations of localization. A discussion about the case study is delivered in the fifth section. Finally, the last section is devoted to the conclusion and proposals for enlargements.

The size and capacity of Data Centers (DCs) is growing at a rapid pace to meet the increased demand of data processing and storage capacity requested by a digital information society. Since DCs are infrastructures that have large energy consumption, there is a need to change their design approach to make them more efficient and more environment friendly. This research was motivated by the planning of a new DC in Portugal. It proposes a multi-criteria framework to assess the sustainability of a DC, which includes a new metric to evaluate the DC efficiency taking into account the environmental conditions of the DC location. ELECTRE TRI was chosen for aggregating different metrics concerning the environmental sustainability of a DC into sustainability categories. The evaluation methodology allows some freedom for each DC to place more weight on the aspects in which it is stronger, an analysis facilitated by the IRIS decision support system.

In this chapter a fuzzy methodology is presented for the valuation of criteria for which the level of available knowledge is limited or inexistent. Due to this lack of knowledge, statistical techniques and data mining are only useable in a limited way. The criteria valuations rely on the elicitation of experts’ knowledge. For coping with criteria valuations in a similar uncertainty context a three-tiered fuzzy inference system has been developed. Details on the fuzzy rules and implications in this fuzzy inference system are provided. This approach has been used in practice for the cost and financial analysis of radioactive-waste-management projects. The case study on a nuclear-fuel repository is a fictive simulation of an actual project assessment. It is presented along with the Fuzzy Inference System software, called FIS Tool, which has been developed for the analysis. It is thought that the approach can be extended to different criteria valuations, particularly in the field of environmental management.

This work is concerned with land use assessment in a region of Burkina Faso, Western Africa. It can help to support the definition of politics promoting a sustainable development of the region. A spatial decision support model is built, based on a coupling of Multi-Criteria Decision Analysis (MCDA) on the one hand and a Geographic Information System (GIS) on the other hand. The Electre Tri method is used to sort the spatial units to ordered categories corresponding to various levels of response to the risk of landscape degradation. The model allows to aggregate physical, economical, socio-cultural and environmental aspects interpreted in terms of their impact on landscape preservation or degradation. Such a categorization leads to the determination of homogeneous zones in the region under study. It can possibly serve as a basis for allocating resources to the most promising sub-regions or the zones needing an urgent intervention.

In case of an accident concerning a nuclear installation, two intervention phases are distinguished: an emergency phase which calls for a rapid and organized response through intervention plans, and a post-accidental phase in which postponed actions are carried out on medium and/or long-term so that the situation comes back to a state judged as acceptable by stakeholders. The PRIME project has developed a decision aiding tool for risk managers involved in an industrial accident involving radioactive substances, through the evaluation of radio-ecological sensitivity of a territory in a post-accidental phase. The proposed decision aiding tool is grounded on the integration of Multiple Criteria Decision Aid (MCDA) and a Geographical Information System (GIS). The proposed methodology relies on the concept of decision map which corresponds to a planar subdivision of the territory in which each subdivision is evaluated on the basis of several criterion maps. This results in a set of disjoint spatial units evaluated on an ordinal scale using the ELECTRE TRI method. Hence, the result is a decision map representing the radio-ecological sensitivity of the territory; such maps prove to be very useful for stakeholders to design relevant post-accidental strategies.

The local public administrations of the Italian cities are responsible for the hazardous material transport within the municipality boundaries. The city of Milan is interested in implementing a system to support the municipal transport department in planning the shipments and control their positions. The paper presents an ongoing project to design and develop a spatial multicriteria decision support system (DSS) based on a Geographical Information System for hazardous material transport in the city of Milan. The DSS considers both the problems of routing and scheduling the shipments in urban and suburban road networks, taking into account the viewpoints of the interested parties (e.g. population, shipping company, vehicle driver, environment agency). We use a risk assessment model that considers the consequences of an accident for each road segment on population, territorial infrastructures, natural elements, critical areas (e.g. areas which may be a target for a terrorist attack). The DSS is in a prototype phase and has been tested on Niguarda, an area of the city of Milan characterized by the presence of an important hospital. The prototype considers the position and time of activities of schools, railways, park and agricultural areas, and hospital buildings that are located in the area. The DSS has been applied to an exemplificative shipment in the area, and the results are presented.

The paper reports a real world decision aiding process concerning rural road maintenance in Madagascar. The issue arises within AGETIPA, the National Agency in charge of conducting Public Works in Madagascar, and can be summarised as a problem of resource allocation to a number of competitive projects. The problem has been modeled using multiple criteria and a classification procedure under two objectives: make the most rational use of the limited available resources and promote participation and commitment of the local actors in the maintenance process. The project is part of an on-going partnership between the LAMSADE and AGETIPA aiming to enhance Decision Support Capacity within AGETIPA.

Hydrogen, a non carbonated energy carrier, is often considered as one possible solution to reduce greenhouse gas emissions from human activity. The use of hydrogen as a possible alternative fuel for automotive applications is envisaged by car manufacturers. However, before a large scale commercialization of hydrogen vehicles, numerous challenges have to be faced, among which the on-board storage of hydrogen. This paper provides a description of the implementation of an MCDA approach for evaluating various competing hydrogen storage technologies for future vehicles. This implementation has been conducted within the STORHY European research project. The MACBETH method has been identified as an appropriate approach for the evaluation and comparison of the technologies from a technical point of view. The evaluation process has been entirely implemented on one hand with several experts from CEA and on the other hand with one of the STORHY car manufacturers. The implementation within the project confirmed that this evaluation method could be used for “application-oriented” multicriteria evaluations. The advantages and drawbacks of the method are finally discussed.

Personal Financial Planning (PFP) is the preparation of target-oriented decisions concerning assets, incomes, and expenses. As people have different preferences for different financial goals, and the goals are flexible, PFP is a Multicriteria Decision Analysis (MCDA) problem that is often addressed by trial calculations under different scenarios. We provide an MCDA model to derive a financial plan that maximizes the value of the expenses for a decision maker with respect to height, time, and type preferences. Specifically, we show how the problem can be solved through a mixed integer programming approach where the weights for the mathematical program are determined with the help of the Analytic Hierarchy Process.

Bank rating refers to the analysis of a bank’s overall viability, performance and risk exposure. Within the recent financial turmoil, bank rating has become extremely important. Typically, bank rating is performed through empirical procedures that combine financial and qualitative data into an overall performance index. This paper presents a case study on the implementation of a multicriteria approach to bank rating. The proposed methodology is based on the PROMETHEE II method. A rich set of evaluation criteria are used in the analysis, selected in accordance with widely accepted bank rating principals. Special emphasis is put on the sensitivity of the results with regard to the relative importance of the evaluation criteria and the parameters of the PROMETHEE method. Analytic and Monte Carlo simulation techniques are used for this purpose.

Up to recently, the processing of a decision problem via several Multiple Criteria Decision Aiding (MCDA) programs required to write the data in various formats. This chapter presents XMCDA, an initiative of the Decision Deck Consortium, which is a standard data model enabling to encode classical concepts from MCDA in XML. Among other things, it eases the analysis of a problem instance by various MCDA techniques compatible with XMCDA without requiring data conversions and it simplifies the sequencing of MCDA algorithms for the resolution of complex decision problems.

In this chapter, we illustrate how the MCDA process can be supported by the use of a decision aiding software called diviz. The diviz workbench allows to build, execute and share complex workflows of MCDA algorithms, and as such, is a convenient tool to help the analyst in the decision aiding process. We start by a presentation of diviz, before switching to the detailed description of a didactic MCDA process, based on a classical example from the MCDA literature. We show how each major step of this process can be backed up by diviz, and how the software can help to build the final recommendation.