UE Multi-criteria Decision Aiding and Artificial Intelligence - 5GUC4202

Goal(s)

(This class will be in English)
Please read the required skills before enrolling in this class

The objective is that students will be able to :

A- implement some multi-criteria decision Aiding methods
B- understand how Artificial Intelligence (AI) techniques (specifically Machine Learning) can be used for decision aiding

Content(s)

The class has two parts A and B. The order of classes depends on the availability of teachers. It means that we may start by part B then part A:

Part A: This part consists of studying some mutlti-criteria decision aiding methods, such as AHP, PARETO, etc. Sessions are proposed in lectures and exercices format (CM and TD). A real case study will be introduced for the AHP method. At the end of this part, the student will understand several multi-criteria decision support methods used operationally. The structure of this part is as follows:
• Introducing the key concepts of multi-criteria decision Aiding
• Describe the main steps of the implementation process of multi-criteria decision support
• Formalize a multi-criteria decision support problem
• Implement the AHP, PARETO, ELECTRE, and other methods….

Part B: This part tackles Artificial Intelligence (AI) in the context of multi-criteria decision aiding. We focus specifically on using machine learning as a decision aiding tool.
• In this part, we first, introduce machine learning techniques for solving multi-criteria decision support problems. The techniques studied are: supervised/unsupervised methods and reinforcement methods. We use these methods for solving case studies, but we do not discuss the details of programming development.
• One or several case studies will be conducted and literature review will be studied for the group work
• Some industrial partners will participate in the lecture by sharing industrial experiences

• English is mandatory (understanding, reading and writing)
• General knowledge of manufacturing and supply chain management: scheduling, inventory management
• Know how to carry out a literature review: search and analyze scientific articles
• Basic knowledge of statistics (correlation analysis, linear regression, etc.), Python (for using some packages), Simulation software (such as Arena).