Bee Colony Algorithm for Proctors Assignment

Abstract

Proctor assignment refers to assigning proctors to examinations with the objective of having the appropriate number of proctors assigned to examinations, subject to conditions such as minimizing the load of proctoring and preventing any conflicting assignments. This problem is intractable, and hence, heuristics algorithms are needed to find good solutions. In this paper, we propose a new solution for the proctor assignment problem based on the Bee Colony meta-heuristic algorithm. The Bee Colony algorithm is a recent population-based search algorithm that mimics the natural behavior of swarms of honey bees during the process of collecting food. The algorithm performs a neighborhood search combined with a random search to balance exploration and exploitation. The food source identified by a honey bee is associated with a candidate solution to the proctors assignment problem. The search accomplished by three types of bees over a number of iterations aiming to find the source with the highest nectar value (fitness value of a candidate solution). We apply the Bee Colony algorithm to previously published data. Experimental results show good solutions that maximize the preferences of proctors while preserving the fairness of the workload given to proctors. The results also show that the Bee Colony algorithm outperforms other methods on most subject problems.