Date of Award


Degree Type


Degree Name

Doctor of Philosophy


Management Science

First Advisor

Layth Alwan

Second Advisor

Xiaohang Yue

Committee Members

Xiang Fang, Samar Mukhopadhyay


Integer programming, Last mile delivery, Omni-channel retailing


E-commerce has grown rapidly in the past decade. In 2015, e-commerce was accounted for 7.2 percent of all retail sales in the U.S., which is massively higher than 0.2 percent in 1998 (U.S. Department of Commerce 2017). Worldwide e-commerce sales also show the same trend and reached $2.356 trillion in 2018 and are expected to grow steadily (Statista 2017). This trend has impacted major areas of operations management including supply chain management and revenue management. Today, e-commerce companies cannot satisfy their customers’ demand using traditional distribution systems. Therefore, retailers with e-commerce channels (e-tailers), cooperate with third party logistic service providers to perform or improve their logistic services and last mile deliveries. In a report on logistics trends, same-day delivery was considered “the next evolutionary step in parcel logistics” (McKinsey 2014). Third party providers are either individual drivers working through the e-tailer’s mobile platform such as Amazon Flex (, or start-ups that provide on-demand urban delivery and aggregate demand via their own mobile platforms, such as In the first essay, we propose three mixed-integer mathematical models and an efficient heuristic algorithm to solve routing and location decisions for these distribution systems, which are different from traditional delivery services. We model this problem as a two echelon location-routing problem with open routes (2E-OLRP) since the third parties have their own fleet of vehicles. Besides a fast and low-cost delivery, retailers are trying to improve the whole shopping experience for customers every day. While in-store sales are relatively low, traditional retailers are increasingly offering products through both e-commerce and brick-and-mortar channels and many online retailers are also opening physical stores. This integration, referred to as omni-channel retailing, can help retailers to have a more responsive demand fulfillment process and provide a satisfactory shopping experience for their customers. In the second essay, we model the consumers’ behavior with discrete choice models and examine that how the probability of purchasing from each of these channels can change by different price and delivery options. We also propose two optimization models that consider pricing, fulfillment and inventory decisions. We show that e-tailers can increase their profit if they make these decisions simultaneously.