Design and Development of a Genetic Algorithm Based on Fuzzy Inference Systems for Personnel Assignment Problem

Abstract

In today’s competitive markets, the role of human resources as a sustainable competitive advantage is undeniable. Reliable hiring decisions for personnel assignation contribute greatly to a firms’ success. The Personnel Assignment Problem (PAP) relies on assigning the right people to the right positions. The solution to the PAP provided in this paper includes the introducing and testing of an algorithm based on a combination of a Fuzzy Inference System (FIS) and a Genetic Algorithm (GA). The evaluation of candidates is based on subjective knowledge and is influenced by uncertainty. A FIS is applied to model experts’ qualitative knowledge and reasoning. Also, a GA is applied for assigning assessed candidates to job vacancies based on their competency and the significance of each position. The proposed algorithm is applied in an Iranian company in the chocolate industry. Thirty-five candidates were evaluated and assigned to three different positions. The results were assessed by ten staff managers and the algorithm results proved to be satisfactory in discovering desirable solutions. Also, two GA selection techniques (tournament selection and proportional roulette wheel selection) were used and compared. Results show that tournament selection has better performance than proportional roulette wheel selection.

Keywords

• fuzzy inference systems
• genetic algorithm
• personnel assignment problem
• disasters management and emergencies
• cost-benefit ratio