A Stochastic Modeling Approach to Faculty Turnover in Higher Education Institutions Using the New XLindley Distribution

Accepted January 2026

Abstract

Employee turnover is a persistent challenge for higher education institutions (HEIs), with direct implications for organizational performance, instructional quality, and research capacity. This study develops a stochastic manpower framework for modeling faculty turnover based on the new XLindley distribution (NXLD), which offers enhanced flexibility for capturing skewness and heavy-tailed tenure behavior. The proposed renewal-based model is used to estimate the expected length of stay of faculty members across academic ranks under alternative promotion and recruitment policy regimes. To account for institutional constraints such as probation periods and mandatory retirement ages, truncated variants of the NXLD are also incorporated. Using controlled simulation experiments and comparative analysis, the NXLD-based framework demonstrates improved goodness-of-fit relative to classical exponential-based models. The results illustrate how stochastic modeling can support evidence-based workforce planning and the evaluation of retention and recruitment policies in HEIs.