Developing location-routing-inventory model under uncertainty: A queuing-based approach

This study develops a mathematical model for the designing of a supply chain network. The uncertain nature of demand and lead time is incorporated into the concerned model. This motivates us to deploy the queuing concept to deal with uncertainties and analysing the number of orders, number of shortages and average of on-hand inventory. Then, in accordance with the outputs of the queuing analysis, a mixed integer nonlinear programming model is devised to design the distribution network of a supply chain. The decisions to be made are facility locations, demand allocations along with inventory management decisions. The objective function of the model aims at minimising the total supply chain costs encompassing location, transportation and inventory costs. Notably, we assume that each facility manages its inventory policy based on a S 1, S policy and stock outs result in lost sales. Inasmuch as the developed problem is difficult to solve by means of exact methods, tailored hybrid solution algorithms based on simulated annealing and genetic algorithm are employed to overcome the computational complexity of the developed model. Finally, using the real information of the Telecommunication infrastructure company, we evaluate the proposed model and the management insights are reported.