This research develops a new mathematicalmodel to study a location-routing problem in healthcare network underdisruption. Almost all of the previous research in healthcare network designhave assumed that network components (e.g., routes, production factories,distribution centers, etc.) are always available and can permanently serve thecustomers (e.g., hospitals and pharmacies). This assumption is no more validwhen the network faces with disruptions such as flood, earthquake, tsunami,terrorists attacks and workers strike. In case of any disruption in thehealthcare network, not only tremendous cost is imposed to the stockholders,but also customers’ health may be jeopardized. Considering disruption in thedesign phase of healthcare network will alleviate the impact of these disastersand let the network to resist against disruption. In this thesis, a mixed integer programming(MIP) model is proposed that formulates a reliable location-routing problemwith pick-up and delivery (RLRPD) services in pharmaceutical distributionnetwork. The objective function attempts to minimize the sum of location costof distribution centers, routing cost of vehicles and cost of unfulfilleddemand of customers. Due to uncertainty in customers’ demand (i.e., delivery)and amount of expired medicines (i.e., pickup), we propose a robust RLRPDmodel. Since the model is NP-Hard for large-size instances, three differentmetaheuristics are tailored and results show the outperformance of hybridalgorithms comparing to classic genetic algorithm.
