A Multi-Objective Optimization Strategy for League Squad Balancing using Improved NSGA-II

The process of selecting players in a franchise-based league like Indian premier league (IPL) is a multi-objective optimization problem. It entails choice trading offs between cost, the performance of those playing, the balance of a team, and the specialization of roles. Towards achieving a realistic and competitive IPL squad using the NSGA-II, this paper introduces two improved versions of NSGA-II, i.e., Budget-Constrained NSGA-II (BNSGA-II) and Interest-Oriented NSGA-II (INSGA-II). In contrast to the current methods, the suggested ones will combine real-life constraints, such as player cost, team composition, and preferences of interested parties. As experimental results indicate, BNSGA-II gives improved convergence and uniformity under budget constrained settings, whereas INSGA-II offers diverse sets of squads according to user-stated interest vectors. The results showcase the practicality of the multi-objective evolutionary algorithms tools to the sports analytics and decision-making by team owners’ franchises.