Multi-period optimal manpower planning with two types of employees in an employee leasing center

This paper studies the issue of manpower planning in an employee leasing center. The center is engaged on the business of leasing its own employees and/or acting as an agent for employees outside the center, where there are two types of employees, the high-ranked employees have mixed skills and can be substituted for the other type of employees to work. First we propose stochastically optimal control models for the multi-period manpower planning in an employee leasing center. In the model there are stochastic constraints. Then we transform stochastic constraints into deterministic constraints. Next we derive some properties to limit the scope of the feasible states in each period without affecting the optimal solution and develop a dynamic programming (DP) approach. Furthermore, to improve the computational efficiency, we also propose a heuristic approximate (HA) approach to derive a near-optimal solution with lower computational complexity. Finally, numerical results are provided to show that the approach HA is much superior to the approach DP concerning the computational time, and the relative error between the two solutions derived by them is no more than 6% in most of cases.