Two-Stage robust optimization approach for supplier selection, development, and order allocation under uncertainty

Effective supplier management in competitive markets involves more than selecting the most capable suppliers. It requires a comprehensive approach that combines supplier selection with long-term development efforts to ensure competitive advantages. However, traditional models often emphasize short-term efficiency and overlook supplier development as a critical strategy for achieving sustained performance improvement. This study proposes a novel two-stage robust optimization framework that unifies strategic and operational decisions to address this limitation. Different from traditional approaches, the proposed model strategically integrates supplier development programs (SDPs) with supplier selection in the first stage to minimize total costs, followed by order allocation in the second stage under the realization of worst scenarios. To capture the inherent uncertainty associated with supplier development, a budgeted uncertainty set is constructed to characterize potential variations in performance improvement. Methodologically, a multi-cut version of the column-and-constraint generation (C&CG) algorithm is developed to accelerate convergence by generating multiple cutting planes per iteration and linearizing nonconvex dual subproblems. Numerical experiments verify the computational advantages of the proposed algorithm, which achieves superior efficiency in solving large-scale instances compared to the standard C&CG algorithm. Furthermore, a procurement case study is conducted to demonstrate the practical applicability of the proposed model. The findings reveal a nonlinear relationship between uncertainty and supplier development, and highlight significant differences in management strategies for domestic and foreign suppliers.