A Comparison of Genetic Algorithm Operators for the Seat Allocation Problem

One of the major aspects of conducting an examination is ensuring proper seating arrangement. Exam seat allocation is a tedious task and is often done manually. Examinations are often weakened by malpractices occurring during the examination process. It is necessary to reduce malpractice to maintain the integrity of examinations. Automatic seat allocation is a simple way to curtail the need for human assistance and save time. It also eliminates the possibility of making mistakes. In this paper, we compare the performances of various genetic algorithm operators in allocating seats to students for examinations. The seat allocation is done in such a way that the student's neighboring seats are occupied by other students who are writing examinations for subjects that are not similar. This helps in minimizing the probability of malpractice taking place in examination halls. We also suggest a crossover operator that can be applied to similar permutation-based problems.