Modelling Nonconvex Network Production Technologies in Wheat Supply Chains: Centralized and Decentralised Approaches

This study proposes a nonconvex network data envelopment analysis model for evaluating efficiency in multi-stage supply chains under centralized and decentralized decision structures. Existing network DEA models typically rely on convexity assumptions, which limit their ability to capture indivisibilities and scale discontinuities inherent in real world supply chains. To address this gap, we develop a unified framework that accommodates nonconvex production technologies while enabling efficiency assessment at both the stage and the system levels. The model allows identification of coordination inefficiencies and structural performance differences across decision regimes. An empirical application to national wheat supply chains demonstrates significant efficiency variation across stages and decision units. It reveals that high efficiency at individual stages does not guarantee overall system efficiency. The results highlight the critical role of coordination in improving supply chain performance. The proposed framework provides a more realistic efficiency assessment tool and offers practical insights for improving resource allocation and operational coordination in complex supply chains.