Using Data To Optimize The Fulfillment And Location Decisions Of An Online Grocery Retailer

This dissertation empirically examines how an online grocer can improve its fulfillment and location-related operational decisions. In the first chapter, we examine whether fulfillment issues lead to customer defection. Based on our findings, we develop an inventory allocation rule to improve the retailer's performance. We estimate that our inventor allocation rule would increase revenue from affected customer by 2%. In the second chapter, we empirically solve the spatio-temporal location problem faced by our partner retailer. We introduce a novel procedure that combines machine learning and econometric techniques to estimate demand for potential new locations and optimize the pick-up location configuration and schedule. We estimate a revenue increase of at least 42% from the improved location configuration and schedule. In the third chapter, we study how the retailer can optimize the mix of delivery zones and fulfillment models using data-driven analytics. We show that the retailer’s choice of fulfillment offerings is economically important using a regression-discontinuity design framework. In the next step, we build and estimate a structural model based on our empirical evidence the parameters of our structural model.