Mathematical programming approach for optimally allocating students' projects to academics in large cohorts

Abstract Many university degree programs (including chemical engineering ones) require final year students and Masters’ students to do an extended research project under the supervision of an academic staff member. However, obtaining a satisfying allocation for both students and supervisors is often a challenging task, especially when the amount of available supervisors is particularly tight and their popularities are highly diverse. In this article we propose a novel method based on a ranked list of supervisors and categories provided by each student, where a category corresponds to a general research area, incorporating this information into the allocation process. A student’s satisfaction may therefore correspond to getting a project either with a highly ranked supervisor and/or in a highly ranked category. With this information, we propose here a systematic approach that relies on a novel mixed-integer linear programming (MILP) model based on a flexible definition of students’ satisfaction. Our MILP overcomes the limitations of manual allocation approaches, which when applied to large cohorts are highly time consuming and may produce suboptimal solutions leading to poor satisfaction levels. This MILP has been applied successfully in the School of Chemical Engineering and Analytical Science of The University of Manchester with increased levels of student satisfaction.