Genetic algorithm based on weighted goal programming for doctor rostering problem

In the field of healthcare, undisrupted service is essential for hospitals. Therefore, shift work plays a vital role in the aspect of satisfying the constrictions such as coverage requirements and government regulations. Doctor rostering problem is classified as an NP-hard problem due to its complexity and scale. Involving fairness of assignments, hospital management policies, and government regulations many related factors must be taken into account during a scheduling process. This study aims to generate a rostering system that can satisfy the requirements of the hospital, fairness amongst the doctors, and reckons with the preferences. A genetic algorithm based on weighted goal programming model was proposed to solve the doctor rostering problem. The proposed model was applied to the Internal Diseases Department and the Lateral Branches Department of Kütahya Evliya Çelebi Education and Research Hospital. 15 different scenarios were constructed considering different problem scales and different preference patterns of the doctors for the future. It is approved that the proposed algorithm can be applied to different problem scales and conditions. The parameters of the proposed algorithm parameters were calibrated with an experimental design method. In this study, two main contributions were presented. A model with new constraints was introduced for researchers. In addition, a genetic algorithm based on weighted goal programming was proposed to solve the problem and applied to a real-case study.