An analytics-driven supply chain network flow optimization for carbon emission reduction with minimal operational disruption

Reducing carbon emissions has become increasingly critical in response to growing legislative pressure and heightened consumer awareness. In this context, organizations face significant challenges in achieving meaningful emission reductions within complex logistics networks and a volatile economic environment. This study focuses on the environmental sustainability in forward logistics network design, aiming to identify opportunities to reduce emissions while minimizing deviations from existing operations. A multi-objective mixed-integer programming model is then developed and solved using an exact branch-and-bound algorithm. Machine learning techniques are employed to estimate missing data, enhancing model realism. Experiments conducted on a real-world dataset show that moderate reductions in emissions, costs, and lead times can be achieved by simply rerouting flows without altering processed volumes. Furthermore, substantial improvements are attainable with as little as a 5% reduction in the network’s existing processing volume. The findings provide both theoretical and practical insights for sustainable supply chain design and offer a scalable framework for data-driven decarbonization.