Teaching Assignment Based on NASH Equilibrium and Genetic Algorithm

Teaching assignment refers to the process of allocating classes to instructors for a given academic semester. Its outcome is a schedule that must adhere to various constraints, such as ensuring that all classes have assigned instructors, avoiding schedule conflicts among instructors, and adhering to predetermined limits on the number of classes assigned to each instructor. It is a complex task due to the involvement of multiple stakeholders such as instructors, and academic departments, each with their own interests. In many cases, these interests may conflict with each other. In this study, we propose using the NASH equilibrium, a branch of game theory, to analyze conflicts of interest among stakeholders and modeling the teaching assignment as a multi-objective optimization problem. Furthermore, we develop a genetic algorithm to solve this problem. Our algorithm is experimented with a real-world dataset consisting of 153 classes and 25 instructors. The experimental results demonstrate the feasibility and effectiveness of our proposition.