Distribution network design with big data: model and analysis

This study addresses the problem of locating distribution centers in a single-echelon, capacitated distribution network. Such network consists of several potential distribution centers and various demand points dispersed in different regional markets. The distribution operations of this network generate massive amounts of data. The problem is how to utilize big data generated to identify the right number of distribution centers to open and the right assignment of customers to opened distribution centers while minimizing the total handling and operation costs of distribution centers, transportation, and penalty. Restrictions on both network capacity and single sourcing strategy are also considered. This study formulates this problem as mixed-integer nonlinear program. The effects of different scenarios on distribution-center locations as demand, the operation costs of distribution centers and outbound transportation, and the number of customers are analyzed through simulation on randomly generated big datasets. Empirical results indicate that the model presented is appropriate and robust. The operational value of big data in the distribution network design is revealed through a case study in which several design alternatives are evaluated.