This project develops a data-driven solution to optimize supply chain logistics under time and data constraints. Using messy GPS shipment data, we construct a simplified hub-and-corridor network and apply a multi-objective optimisation model (Gurobi MIP) to balance cost, delivery time, and carbon emissions.
The approach combines clustering (KMeans) for network design with prescriptive optimisation, enabling decision-makers to evaluate trade-offs across different operational priorities, such as cost efficiency, on-time delivery, and sustainability.
The model produces actionable outputs, including optimized routing strategies, corridor flow maps, and policy scenarios, allowing managers to make informed logistics decisions.
Overall, the project demonstrates how prescriptive analytics can transform complex, noisy data into clear and practical operational strategies
