Analysis of a Single-Grade Workforce Using Stochastic Methods with Correlated Egress Clusters and Two Factors Defining Breakdown Points

This research investigates the challenge of optimizing recruitment timing in a single-grade workforce system that experiences reductions in personnel due to either employee departures or breaks. Employee exits, occurring in random clusters, are modeled using a compound process, with each cluster’s workforce loss represented by exchangeable, exponentially correlated random variables. Break-related exits are classified by the intensity of attrition and modeled as independent and identically distributed hyper exponential random variables. Threshold of breakdown is split into two components: exits and breaks. The paper proposes two stochastic models: Model 1, where all exits are treated uniformly, and Model 2, which differentiates exits into voluntary and involuntary categories. The study applies the univariate CUM recruitment policy to calculate the mean and variance of recruitment time. Using Matlab, numerical illustrations are presented to illustrate the effectiveness of the proposed models. The research also explores how different nodal parameters influence key system characteristics, yielding insights into the dynamics of recruitment timing. A major contribution of this work lies in the development of two comprehensive stochastic models that account for both voluntary and involuntary exits, providing a more nuanced understanding of workforce attrition. The use of statistical analysis and Matlab simulations strengthens the findings, offering practical applications for optimizing workforce management strategies in dynamic environments. This study provides a foundation for future exploration into more sophisticated recruitment policies, potentially incorporating AI/ML and deep learning techniques. 2020 Mathematics Subject Classification: 60H30, 90B70, 91D35.