Essays on Omnichannel Sale

Thesis (Ph.D.)--University of Washington, 2016-08

In this dissertation, we study three problems related to omnichannel sale. In the first two chapters, we study omnichannel fulfillment where a retailer can use either the online or the offline channel to back up the other one by fulfilling its demand, because of stockout or low inventory availability. In the last chapter, we study an incentive problem where sales agents compete for a common customer base. This problem is loosely related to similar problems in omnichannel sales – when an online order is filled by an offline store, or vice versa, how should sales credit be shared? In the first chapter, we study store fulfillment strategy of an omnichannel retailer that would like to leverage its established offline retail channel infrastructure to capture online sales after stockout. We consider a single newsvendor-type product that is sold in both online and offline channels to non-overlapping markets with independent Poisson demand. The offline store can fulfill online demand at an additional handling and fulfillment cost, k, but not vice versa. We characterize the optimal rationing policy which determines whether online demand should be fulfilled or not given the remaining time and inventory. Due to the challenges of implementing the optimal policy, we further propose two simple and effective heuristics. We also show that integrating the rationing policy into retailer’s higher-level inventory stocking and supply chain design decisions can have a significant impact on the retailer’s inventory level and profitability. As a result, we propose an integrated policy, where the retailer builds separate inventory stocks for each channel but can use the offline inventory to back up online sales, subject to a rationing heuristic. In the second chapter, we study discounted home delivery strategy of an omnichannel retailer that would like to leverage its online channel to help with offline sales when offline store has limited inventory. We consider a single newsvendor-type product that is sold in both online and offline channels to non-overlapping markets with independent Poisson demand. Store has option to offer customers discount d to incentivize them to accept home delivery of item rather than taking an inventory unit from the store. We assume customers are heterogeneous in their discount sensitivity, ranging from price sensitive to leadtime sensitive. We characterize the optimal dynamic discounting policy which determines at any point in season whether the offline store should offer discount or not, and if yes, the optimal discount level. Again, due to the challenges of implementing the optimal policy, we propose two simple and effective heuristics. We conduct an extensive numerical study and find that retailers can considerably benefit from discounted home delivery policy. In the last chapter, we study sales agent’s competition for a common customer base when sales agents are heavily paid by commission. We assume agents can affect customers perceived service quality through their service time. We build a simple model that captures agents time and quality trade off while competing for a common customer base. We assume that the agent’s service time can increase sales probability, but the agents incur cost for their effort. We show that agent’s speed up in the decentralized setting, in comparison to the central solution, which results in lower expected sale for the system. We suggest financial, operational, and informational tools that can align agent’s incentive with that of the service provider’s.