Optimizing job offer packages in a two-sided matching with bounded rationality: a two-stage stochastic approach

Abstract

Personnel selection is a two-sided market where companies compete for qualified candidates by designing job-offer packages. However, there is a gap in understanding how to optimize these packages considering candidate preferences and associated costs, while decision-makers exhibit bounded rationality due to limited information or cognitive constraints. This study addresses this gap by proposing a matching framework that accounts for bounded rationality based on the Quantal Response Equilibrium (QRE), in which both sides are not perfect optimizers and face uncertainty in the other side’s actions. Maximum Likelihood Estimation (MLE) and analysis of real hiring data confirm that decision-makers exhibit bounded rationality and tend to behave more rationally as the selection process progresses. Finally, a two-stage stochastic optimization approach using Particle Swarm Optimization (PSO) to determine the optimal job offer package for the organization, taking into account its human resource policies and candidate competencies, is presented. The evaluation of the results and a sensitivity analysis are conducted under rational and bounded rational modes. This approach offers valuable insights for organizations to optimize their hiring processes and attract top talent.