An Integrated Location-Inventory-Routing Problem: Metaheuristics and Environmental Extensions

  • AbdelHalim Hiassat

Student thesis: Master's Thesis


We develop a mathematical model for supply chains with perishable products that integrates location, inventory, and routing decisions into a single model. Products are assumed to have a fixed shelf life, after which items must be discarded. A single manufacturer ships to a number of warehouses which, in turn, ship to retailers. Open warehouses are chosen from a set of candidate locations. Goods are distributed by a homogeneous fleet of vehicles, each having the same capacity. The integrated model decides how many warehouses to open, where to locate them, which customers to allocate to them, how much inventory to keep in each time period, and the number of vehicles and routes needed to distribute goods to retailers. The integrated approach offers a more realistic model of real-world problems in which strategic, tactical, and operational decisions are considered at the same time. To account for environmental considerations, in the second part of this thesis we extend our model to include CO2 emissions considerations. This is done by allowing for the buying and selling of carbon emissions credits under the Kyoto Protocol. We found that adding carbon emissions considerations greatly affected the retailer-warehouse assignments, routing decisions, and the total cost incurred by the system. This new model can be used by firms that are considering a transition toward greener and more sustainable supply chains. Finally, noting that integrated models like the ones developed here are NP-hard problems which in general cannot be solved in polynomial time, in the third part of this thesis we develop a Genetic Algorithm (GA) procedure to efficiently solve our models to optimality or near-optimality. Our GA deals with all three levels of decisions, and solves medium- and large-sized instances efficiently, reaching optimal or near-optimal solutions in reasonable time. A thorough analysis is conducted to optimize the performance of the heuristic.
Date of Award2012
Original languageAmerican English
SupervisorAli Diabat (Supervisor)


  • Location Problems (Programming)
  • Environmental Monitoring

Cite this