A multi-objective, multi-echelon optimization model for perishable fruit supply chain network design using exact methods and NSGA-II: A multi-fruit case study from northwestern Bangladesh
Year: 2026
Type: article
Abstract: Post-harvest losses in perishable fruit supply chains reduce farmers’ income, food availability, and environmental performance. However, few network-design models jointly consider product deterioration, logistics risks, and greenhouse gas emissions. This study develops a three-objective, multi-echelon optimization model for perishable fruit supply chain network design. The objectives minimize total cost, time-based service and disruption exposure, and greenhouse gas emissions from transportation and refrigeration. The model is formulated as a mixed-integer linear program and solved using the augmented ε-constraint and weighted-sum methods, with NSGA-II used for comparison. The model is applied to a stylized, mango-anchored supply chain involving mango, dragon fruit, and orange in the Sapahar–Naogaon corridor of north-western Bangladesh. The network includes four gardens, three local hubs, two cross-docking centers, and four stores. The exact method identifies the Pareto front within seconds. The cost-optimal configuration uses one hub and one cross-dock, while opening all five facilities reduces emissions by 6.7% at a 67% higher cost. NSGA-II identifies the relevant objective-space region but produces solutions with an average cost 29% higher than the exact optimum under equal computational limits. Sensitivity analysis shows that the cost-optimal network remains stable unless capacity is sharply reduced or disruption probability increases substantially. A monthly formulation incorporating harvest calendars reveals peak-month capacity shortages at a single hub, increasing minimum cost by 5.4% and requiring a second hub. These findings highlight the importance of seasonal capacity planning and the trade-offs among cost, service, and emissions. Given the stylized case, the results indicate potential trends rather than direct forecasts.
Source: Sustainable Futures
Language: English
Subfield: Strategy and Management
Domain: Social Sciences
Open Access status: gold
Estimated APC paid: $2,340
OpenAlex ID: W7235935793
