Strategic Workforce Planning Under Uncertainty

• 成果类型: Article
• 署名作者: Jaillet, Patrick; Loke, Gar Goei; Sim, Melvyn
• 署名单位: Massachusetts Institute of Technology (MIT); Erasmus University Rotterdam - Excl Erasmus MC; Erasmus University Rotterdam; National University of Singapore
• 刊物名称: OPERATIONS RESEARCH
• ISSN/ISSBN: 0030-364X
• DOI: 10.1287/opre.2021.2183
• 发表日期: 2022
• 页码: 1042-1065
• 关键词: programming approach; optimal recruitment; forecasting-model; manpower systems; turnover

摘要

The workforce planning problem of hiring, dismissing, and promoting has been the perennial difficulty of Human Resource (HR) management. To cope with uncertain attrition, we propose a new approach of finding a course of action that safeguards against violating organizational target-meeting constraints, such as productivity, budget, headcount, dismissal threshold, and managerial span of control. As such, this approach leads to a tractable conic optimization model that minimizes a decision criterion that is inspired by the riskiness index of Aumann and Serrano, for which its value can be associated with probabilistic and robustness guarantees in meeting constraints under uncertainty. Additionally, our model departs from the literature by considering employees' time-in-grade, which is known to affect resignations, as a decision variable. In our formulation, decisions and the uncertainty are related. To solve the model, we introduce the technique of pipeline invariance, which yields an exact reformulation that may be tractably solved. Computational performance of the model is studied by running simulations on a real data set of employees performing the same job function in the Singapore Civil Service. Using our model, we are able to numerically illustrate insights into HR, such as the consequences of a lack of organizational renewal. Our model is also likely the first numerical illustration that lends weight to a time-based progression policy common to bureaucracies.