This thesis addresses two issues in the supply chain management.

The first part investigates the impact of consolidation when last mile delivery activities are consolidated by an e-commerce platform. When delivery jobs are coordinated and dispatched by a centralized system like a dominant e-commerce platform player, we propose partition strategy for consolidation. We evaluate the delivery cost of the courier companies under partition and compare it with that before consolidation to measure the impact. We also build criteria to guide a partition and evaluate the efficiency of partition strategy. Moreover, distributionally robust optimization and Wasserstein uncertainty set are used to characterize the random demands, and to evaluate whether the consolidation is robust to perturba...[ Read more ]

This thesis addresses two issues in the supply chain management.

The first part investigates the impact of consolidation when last mile delivery activities are consolidated by an e-commerce platform. When delivery jobs are coordinated and dispatched by a centralized system like a dominant e-commerce platform player, we propose partition strategy for consolidation. We evaluate the delivery cost of the courier companies under partition and compare it with that before consolidation to measure the impact. We also build criteria to guide a partition and evaluate the efficiency of partition strategy. Moreover, distributionally robust optimization and Wasserstein uncertainty set are used to characterize the random demands, and to evaluate whether the consolidation is robust to perturbation in delivery demands. Numerical results using real data from Cainiao demonstrate the feasibility, efficiency and robustness of our partition strategy.

The second part studies the return and resale policies in the presence of strategic consumers. We consider a firm selling to strategic consumers with uncertain valuations in a two-period model. Consumers are heterogeneous, denoted by high type and low type. In each period, the firm offers a return policy regarding selling price and refund. Different from the existing literature, the firm can resell the returned products from the first period, together with the remaining inventory if there is any, in the second period. We analyze the firm’s optimal return and resale policy in both periods as well as its inventory decision. We first show that, when resale is adopted, the firm tends to offer a generous refund in the first period and a stringent refund in the later period. Second, we find that the firm benefits more from resale when the market is less differentiated. Third, two counter-intuitive results are also revealed. One is that a high refund is not always coupled with a high selling price in the optimal return policy. The other finding is that the increase of valuation of high type consumers may hurt the firm’s profit. Our results provide important managerial insights for the managers to determine the return and resale policies.