Outpatient Scheduling with Genetic Algorithm: The Power of Mutation Operators

Abstract

Background: Outpatient scheduling is a complex and time-consuming task. To address this challenge, numerous studies have developed various optimization methods, including genetic algorithms.

Objectives: This study aims to develop a task-specific genetic algorithm and investigate the effect of different mutation operators on its performance, focusing on minimizing the earliest completion time of scheduled examinations.

Methods: Random and two heuristic mutation operators were designed and compared. The effect of these mutation operators and their parameters were evaluated across four fundamentally distinct scheduling scenarios.

Results: The exponential mutation operator outperformed all others across all scheduling problems. It achieved an optimal schedule in 100% of runs for the simplest task and in 74.5% of runs for the most complex one. In comparison, the random mutation operator achieved 100% and 1%, while the polynomial operator reached 75.66% and only 0.22%, respectively.

Conclusion: The efficiency of the genetic algorithm developed for outpatient scheduling is strongly influenced by the choice of mutation operator. The performance of the algorithm can be greatly enhanced by employing a specialized mutation operator tailored to the objective function.

Keywords: Outpatient scheduling; genetic algorithm; mutation operator.