Comparison of a Rule-Based Heuristic and a Linear Programming Model for Assigning Mentees and Mentors in a Women in Technology Mentoring Programme

Abstract

Women remain significantly underrepresented in the fields of science, technology, engineering and mathematics (STEM), but positive mentoring relationships can help mitigate the challenges they face when studying and working in these areas. To support female university students in STEM, the Auckland University of Technology (AUT) established the Women in Tech mentorship programme in 2019. Initially, the matching of mentees and mentors was achieved manually, but as the programme’s popularity grew, this process became increasingly time consuming. This study addresses the challenges associated with assigning mentees to mentors by automating the matching process based on mentee and mentor attributes. A rule-based heuristic is proposed and compared with a linear programming (LP) approach. Numerical experiments were conducted to evaluate the performance of these algorithms across various scenarios. The rule-based heuristic provides a simple and easily understandable way to allocate mentees and mentors that performs nearly as well as an optimal matching provided by the LP approach. Applying these algorithms to real data from the AUT Women in Tech mentorship programme, it was found that they outperformed manual matching in several performance metrics.