Towards a sustainable viticultural supply chain under uncertainty: Integration of data envelopment analysis, artificial neural networks, and a multi-objective optimization model

Abstract

The viticultural supply chain plays a critical role in the agricultural sector, yet its optimization remains understudied despite its economic, environmental, and social significance. This study proposes a multi-objective, sustainable viticultural supply chain network design model that simultaneously minimizes costs, mitigates environmental impacts and enhances social benefits. To address the complexity of vineyard selection, a pioneering hybrid strategy, integrating a data envelopment analysis method and an artificial neural network approach is developed, enabling the identification of optimal vineyard locations based on sustainability criteria. Furthermore, a robust optimization approach is devised to handle uncertainties in the supply chain. The augmented ε-constraint method is employed to solve the multi-objective model, balancing trade-offs among conflicting objectives. A real-world case study in Iran validates the model, demonstrating its efficacy in improving network efficiency, minimizing waste, and maintaining product quality. Sensitivity analysis highlights the robust model's superiority over the deterministic approach, particularly in scenarios with limited historical data and high uncertainty. The findings emphasize the effectiveness of the proposed hybrid strategy in fostering a sustainable and robust viticultural supply chain.