2023.2

Fundamentals of Multicriteria Decision Analysis

Code
ENST52089

Syllabus

Multicriteria Decision Analysis (MCDA) methods assist in complex decision-making by evaluating alternatives with multiple conflicting criteria. They involve defining the problem, identifying relevant criteria, assigning weights to criteria, and ranking alternatives based on their performance. MCDA is widely used in various fields to support rational and transparent decision-making. Common MCDA techniques include AHP, ANP, TOPSIS, and ELECTRE, among others, offering valuable tools for addressing multifaceted decision problems. MCDA methods are widely applied in various fields, including business, environmental management, healthcare, and public policy, to support rational and transparent decision-making in complex situations where trade-offs are inherent. These methods provide valuable tools for tackling decision problems with multiple dimensions and can assist in achieving more robust and informed outcomes.

  • Providing students with theoretical and methodological tools to understand the basics of structured decision-making processes;

  • Providing inductive reasoning instead of mathematical foundations, whenever possible, to make it easier for management / business students;

  • Introducing students to the theory of decision-making and their applications in the management / business scenarios;

  • Presenting easy-to-use starter tools to model MCDA decision-making processes to both research projects and industry applications.

Assessment criteria

  • Class participation 20%

  • Project presentation 20%

  • Paper development 60%

Week 01 - General decisions

Welcome! In this class we are going to cover some basics. To start with, I am going to introduce myself and provide you with a syllabus.

We will also do a series of warm-up activities containing several decision-making scenarios, from simple to complex, so we can explore the rationale behind MCDA, but more importantly, its limitations and issues.

SLIDES:

01 - General Decisions

Week 02 - MCDA introduction

In this class, we are going to take a formally explore the basics of MCDA (history, definition(s), basic mathematical concepts).

In addition we are going to use a precursor method to MCDA (even swaps) to analyze a real decision. We will end our meeting with some research papers that use MCDA as a method.

Mandatory reading:

  • Goodwin, P., & Wright, G. (2014). Decision analysis for management judgment. John Wiley & Sons.

  • Belton, V., & Stewart, T. (2012). Multiple criteria decision analysis: an integrated approach. Springer Science & Business Media.

  • Greco, S., Figueira, J., & Ehrgott, M. (Eds.). (2016). Multiple criteria decision analysis: state of the art surveys. 2nd Ed.. New York, Springer.

  • Forman, E. H., & Selly, M. A. (2001). Decision by objectives: how to convince others that you are right. World Scientific.

SLIDES:

02 - MCDA Intro

Class participation / weekly assignments

Throughout this semester, you are going to be exposed to a variety of papers and book chapters dealing with the core concepts of MCDA. You are also going to be exposed to a range of different methods that will allow you to leave this course with a few tools for research and practice.

You are expected to study the papers (and not only skim over the contents) and participate in their discussion, as class participation is a key part of the evaluation process (as well as grading).

Research proposal / Paper development

The proposal and paper development should be developed exclusively for this course and there should be no significant overlaps with any other manuscripts you have developed or are developing for other courses. You may incorporate this proposal into your thesis/dissertation if allowed by your supervisor.

The proposal has to be directly related to MCDA as a tool, but it would benefit the class more if you focused applications close to your research goals. You may bring to your proposal aspects from other fields, but an MCDA applications is mandatory.

As for style guides, choose from ABNT or APA. APA is currently the preferred style guide for management journals and events, including Brazilian ones. You can find a guide on how to use APA here.

Every class we will spend some time (about 30 minutes or more depending on the need) on both paper-writing skills or discussion of your proposals.

The last day of this course is set apart for presentations. You will need to provide your classmates and instructor copies of your manuscript (a pdf file should be enough) at least 5 days prior to your presentation. You should prepare a short presentation about your paper (max. 15 minutes) along with any visual materials you consider necessary for the discussion (a few slides should suffice).

I would like to continue working on all of these manuscripts so you may have them ready for submitting next year to a nice outlet or an event. Please contact me so we can start working on them, ok? Also, contact your supervisor, if you already have one, so we can plan the development and further publishing of said papers, ok?

Week 03 - Analytic Hierarchy Process (AHP)

Today we are going to explore the MCDA method that is probably the most used and cited within the field - AHP (Analytic Hierarchy Process). We are going to demonstrate AHP by the classical approach ("eigenvector method") and a simplified approach ("average method"). We are also going to explore a free AHP software - Superdecisions.

Mandatory reading:

  • Saaty, T. L. (2008). Decision making with the analytic hierarchy process. International journal of services sciences, 1(1), 83-98.



SLIDES:

03 - AHP

Week 04 - Fuzzy AHP

Today we continue with AHP, but with a twist - using fuzzy numbers instead of discrete / continuous values as before. We introduce the notion of fuzzy numbers, membership function, TFNs (triangular fuzzy numbers), fuzzy numbers and operations. We are going to use Chang's (1996) Extent Analysis Method to demonstrate Fuzzy AHP.

We are going to use Online Output's Fuzzy AHP demo as well as my own Fuzzy AHP app (that uses a weighting scheme to consolidate strategic, tactical and operational perceptions). This app is still in development and should be used with caution. Its code is hosted on my GitHub repository and I appreciate feedback, error finding, suggestions and criticisms.

You can learn more about fuzzy numbers from prof. Dr. Marcos Eduardo Valle's (UNICAMP) class notes on fuzzy numbers.

Mandatory reading

  • Chang, D. Y. (1996). Applications of the extent analysis method on fuzzy AHP. European journal of operational research, 95(3), 649-655.

SLIDES:

04 - Fuzzy AHP

Week 05 - TOPSIS (Technique for Order of Preference by Similarity to Ideal Solution) and Fuzzy TOPSIS

Today we will use TOPSIS (Technique for Order of Preference by Similarity to Ideal Solution) to evaluate rankings on real decisions.

In addition we will explore its fuzzy version to account for uncertainty, vagueness or doubt.

SLIDES:

05 - TOPSIS & Fuzzy TOPSIS

Week 06 - ELECTRE (Elimination et Choix Traduisant la Réalité)

In this session we will explore a French method called ELECTRE (that means in French Elimination and Choice Translating Reality). It is a different method and we will compare it to previous ones.

  • Tiwari, R., Agrawal, S., & Kasdekar, D. K. (2020, February). Application of ELECTRE-I, II methods for EDM performance measures in manufacturing decision making. In IOP Conference Series: Materials Science and Engineering (Vol. 748, No. 1, p. 012015). IOP Publishing.

SLIDES:

06 - ELECTRE

Week 07 - PAPRIKA (Potentially All Pairwise Rankings of All Possible Alternatives)

Today we are going to explore a relatively recent method, PAPRIKA (Potentially All Pairwise Rankings of All Possible Alternatives). This model is elegant and simple in its mechanics yet powerful for research and business.

  • Hir Hansen, P., & Ombler, F. (2008). A new method for scoring additive multi‐attribute value models using pairwise rankings of alternatives. Journal of Multi‐Criteria Decision Analysis, 15(3‐4), 87-107.

SLIDES:

07 - PAPRIKA

ENCONTRE-ME EM: