Balanced Student Partitioning to Promote Effective Learning: Applications in an International School

摘要

In this paper, we describe a system that our team developed to solve the problem of heterogeneously partitioning students into classes for the Singapore International School based in Hong Kong. This problem has multiple objectives such as to achieve similar class sizes, similar gender ratios among all classes, each student having at least one old classmate of the same gender, conflict avoidance among students, and similarity of score distribution curves. We proved that this problem is extremely hard and provided an example to show that the number of feasible solutions is astronomical for only medium size cases. We devised and implemented a simulated annealing (SA) algorithm to solve this problem. Our experimental results based on real application data indicate that our SA algorithm is able to improve the quality of the school's partitioning solutions and clearly meets all objectives set out by the client.