A Parallel Genetic Algorithm to Optimize the Massive Recruitment Process

Abstract

HR managers require efficient and effective ways to move forward from traditional recruiting processes and select the right candidates for the right jobs. The kind of staff recruitment that we deal with in this paper is the massive recruitment under several constraints modeled by with the objective of improving the company's performance. It is modeled as a multiple knapsack problem known as an NP-hard problem. Henceforth, solving this problem by a basic GA leads to an approximate solution with large CPU time consumption. For this purpose, we propose a parallel genetic approach to recruitment in order to generate the best quality solution in a reduced CPU time that ensures a better compatibility with what the company is looking for. Operationally, the results obtained in different tests validate the performance of our parallel genetic algorithm for the best optimization of human resources recruitment.