Including Labor Flexibility in the Integrated Suppliers' Selection and Aggregate Production Planning Problem Under Uncertainty

Abstract

Uncertainty inherent to suppliers and customers must be effectively managed for planning decisions. The former directly affects production capacity, while the latter influences demand, both impacting production planning. To align production capacity with demand requirements, tactical production planning models serve as data-driven decision-support tools. However, uncertainty in key parameters can lead to impractical decisions and unfulfilled demand. To address these challenges, this paper presents an integrated approach that combines a multi-period tactical production planning model with the supplier selection problem, explicitly accounting for stochastic demand and supplier non-compliance by modeling a network that facilitates labor transfers between production lines. This approach mitigates the impact of uncertain demand and enhances workforce capacity utilization. The proposed framework is applied to a manufacturing company that processes a single raw material into multiple final products through multiple production lines. We extend the proposed deterministic formulation into a two-stage stochastic programming model, which incorporates labor flexibility, enabling workers to shift between production lines when needed. To enhance decision-makers' what-if analysis capabilities, a two-phase matheuristic is also proposed. In the first phase, a surrogate model is solved to reduce the search space of selected suppliers in time periods, while in the second phase, the reduced problem is optimized. The results demonstrate the matheuristic's advantages in terms of solution quality and computational efficiency. These findings indicate that the proposed approach holds significant value for practitioners. Furthermore, our computational experiments show that allowing workers to shift between production lines reduces overall production planning costs and improves supplier acquisition decisions. At lower levels of flexibility, however, the model recommends lower investment in supplier transactions, which results in higher initial installed labor capacity and greater inventory accumulation over time.