Multicriteria Decision Analysis / Aiding (MCDA)

What is it?
Multicriteria decision analysis is an approach with many existing tools or methods to help making a decision in situations or problems where there is a large number of criteria and a large number of options to consider or to choose from.

Background/context and objectives
Although the approach comes from the discipline of operations research (widely used in business management), the approach separates from other decision making tools like goal-programming because it deals with problems where not always the search is for the best option in terms of one technical goal. Most problems have a variety of aspects which are not always possible to measure in quantitatively or may be conflicting, which may require some trade-off between options. At the same time, most of the time real world decisions must consider social preferences and uncertainty in the available information.
Multicriteria decision analysis methods look for different ways to find answers to situations where all these aspects are present. Since all decision problem are different, many methods have evolved to respond to the different particularities. The most known group of methods are: additive linear modelling, analytical hierarchical process (AHP), outranking (most know examples are ELECTRE and PROMETHEE methods), Multi-attribute Utility Analyisis (MAUT), Multi-attribute Value Theory (MAVT).

Although there are many methodologies within this set of approaches, the main and most common elements are: the alternatives or options that will be evaluated (compared, ranked, sorted, etc.); the criteria or attributes that will be used to evaluate the alternatives; the stakeholders or decision makers’ preferences on each criterion (sometimes represented as weights and sometimes as votes); and the performance or data of each alternative on each criterion that will be evaluated. What usually varies between methods are: the inclusion or not of decision makers’ preferences, the possibility or not to make trade-offs (or compensation between one bad performance with another good performance on another criterion), the mathematical method for aggregation, and the outcome (ranking, sorting, selection, etc.).
Useful guidelines and theoretical descriptions of the most used methods can be found in Dodgson 2000 and Figueira et al. 2005 (see references below).

Guidance for applying the framework
To use an MCDA method, it is necessary to define the decision problem, the result you are looking for and the nature of the information that will be used. Some methods may be more data-demanding (situation that is not always possible), while some others work better with majority of qualitative information. Based on this and on what type of result you are looking for you can select an appropriate method.
Method selection can also depend on the availability of a user-friendly software, especially if the intention is to use it in a participatory process or in a group setting (most known softwares for participatory processes incude Expert Choice, PROMETHEE, NAIADE).
After selecting a method, it is necessary to define who will be the participants or stakeholders. Participants may help providing criteria, their individual weights/importance to each criterion or even by given a value for the performance of each alternative on each criterion. How involved are the participants will depend on the method selected and on how the decision problem will be faced. Next steps require the collection or estimation of data representing the performance of the alternative/option in each criterion and the integration of the information to obtain an outcome. Each MCDA method has a different aggregation approach and different outcomes: ranking, sorting in pre-defined categories, selection of best options, etc.
The outcome of the MCDA aggregation may not necessarily be the ultimate result of the analysis, as this process may be used to generate discussion, stakeholder engagement or more analysis of alternatives.

Good practice tips
The large number of literature on different methods, applications and comparison between methods may be overwhelming for new users. To downsize the number of options is recommended to initally explore older methods and specific applications on the specific field of your interest, e.g. forest evaluation policies, watershed management, etc. The large number of MCDA methods range from very simply to more complex. Although the more complex may give the possibility to include other interesting aspects of a decision (such as the possibility of thresholds, levels of importance of a comparison, etc.) this may not be too easy to understand nor explain in simple terms in a participatory process.
At the same time, although there are very good methods developed, very few of them have been used in participatory settings, have user-friendly software nor have implementation guidelines for non-MCDA experts. More work must be done so practitioners from different disciplines can be beneficiated from this group of methods.

Success stories
The application of these methods have been found to help organize ideas, find useful solutions, reduce conflict and even identify alternatives not previously considered.
Some interesting examples of MCDA applications may be found in:

Hajkowicz, S., 2007. A comparison of multiple criteria analysis and unaided approaches to environmental decision making. Env. Sci. & Pol. 10:177–184.

Hostmann, M., Bernauer, T., Mosler, H.J., Reichert, P., Truffer, B., 2005. Multi-attribute value theory as a framework for conflict resolution in river rehabilitation. Journal of Multi-Criteria Dec. Anal. 13:91–102.

Wilson RS,Arvai JL. 2006. Policy Analysis Evaluating the Quality of Structured Environmental Management Decisions. Environmental science & technology 40:4831–4837.

To learn more
Useful books and manuals:
Dodgson, J., Spackman, M., Pearman, A., Phillips, L. 2000. Multi-criteria analysis manual. UK Department of Transport, Local Government and Regions (DTLR), London.

Figueira, J., Greco, S., Ehrgott, M., 2005. Multiple Criteria Decision Analysis: State of the Art Survey, 1st Eds, Springer, New York, U.S.

Useful articles to select methods:

Kiker, G.; Bridges, T.; Varghese, A.; Seager, T.P.; Linkov, I. 2005. Aplication of Multicriteria Decision Analysis in environmental decision making. Integrated Environmental Assessment and Management. Volº, Number2, 95-108

Wittmer H, Rauschmayer F, Klauer B. 2006. How to select instruments for the resolution of environmental conflicts? Land Use Policy 23:1–9.


Created on 04 Nov 2016 13:39 by Anahi Ocampo
Updated on 28 Nov 2016 14:03 by Barron Orr

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License