An adaptive large neighbourhood search metaheuristic for hourly learning activity planning in personalised learning

Personalised learning offers an alternative method to one-size-fits-all education in schools, and has seen increasing adoption over the past several years. learning’s focus on learner-driven requires novel scheduling methods. In this paper we introduce hourly, activity planning problem of personalised learning, formulate methods solve it. We present integer linear programming model problem, but does not generate schedules sufficiently quickly for use practice. To overcome this, propose adaptive large neighbourhood search metaheuristic instead. The metaheuristic’s performance is compared against optimal solutions a numerical study 14,400 instances. These instances are representative secondary Netherlands, were developed from expert opinions. Solutions average deviate only 1.6% results. Further, our experiments numerically demonstrate mitigating effects changes structure staffing have challenges satisfying learner instruction demands learning. • An innovative OR educational context introduced. ILP formulated problem. A experimental based Dutch shows complexity ALNS shown good-quality solutions. Insights into presented.