Two-echelon distribution network design with collaboration among carriers

Globalization, exponential growth of e-commerce and q-commerce industries, changing market habits and increased need of logistics services result in high competitionamong supply chain pillars. Collaboration is an effective strategy to pursue in thisendeavour. We define a two-echelon distribution network design problem in whichparties can collaborate to complete the last-mile delivery requests in the lower echelon. The objective is to minimize costs which arise from facility opening, transportation and transfer of goods between regional depots. In the upper echelon,goods are transferred from plants to regional depots via direct transportation. Inthe lower echelon, goods are delivered to the customers in a milk-run fashion fromregional depots. We develop three mixed-integer linear programming models whichdiffer in terms of modelling outbound routing decisions. Several valid inequalitiesare proposed to strengthen formulations. To solve a traditional vehicle-based formulation, a cut-generation based method is developed. For the path-based formulation,a heuristic route pool generation procedure which promotes collaboration is developed. Proposed models are tested with different problem sizes to examine solutionqualities and computational times. Moreover, models are tested under different collaborative network settings in which main parameters of the problem such as numberof common customers, demand amounts and number of common depots are variedin order to explore managerial insights such as savings due to collaboration.