Olga Pak

Date of Award

Spring 2020

Document Type

Open Access Dissertation


Moore School of Business

First Advisor

Mark Ferguson


In this dissertation titled “Three Essays on Optimization and Decision-Making Solutions in Retail Operations,” we explore various techniques aimed at optimizing the operational efficiency in a grocery retail store. Specifically, the first essay examines a store manager’s decision of which stock-keeping units (SKUs) from a given category to assign to a promotional display space. We develop a decision support tool that consists of an estimation model and an optimization model. Using a grocery store sales transaction dataset, we introduce a methodology to measure the incremental lift in sales of placing a particular SKU on promotional display space. Our optimization model includes the incremental lifts (from the estimation method) combined with the estimated base-sales rates and profit margins of each SKU so that the profitmaximizing SKU can be chosen for a promotional display space for each week of the year.

The second essay offers a novel methodological solution on the appropriate identification and analysis of submarkets in product categories. Our research contributes to the literature in the following ways. While a vast amount of literature in both marketing and operations management investigate retail decision tree structures, limited information exists on developing algorithms that allow to generate, analyze, and test data-driven decision trees. Understanding how decision trees may drive consumer preferences is critical to a retailer’s choice of product category assortment. We provide a methodology on empirically constructing and evaluating the best fitting decision tree structures using easily accessible and readily available scanner data.

The third essay studies the mechanisms retailers can use to facilitate sales of reduced packaged products, which have a number of advantages that are attractive to retailers, manufacturers, and consumers. Large product packaging creates logistical and operational challenges for retailers who carry such products since these products require more space to be stored and displayed, and more manpower to handle it. In contrast, products in smaller packaging have fewer such problems, and, thus, positively contribute to the retailer’s operational efficiency. We discuss and empirically test two levers that retailers may utilize to influence the sales of reduced packaged products. Using sales data for liquid detergents, we show that retailers with market power are able to announce their preferences for reduced packaged detergents, which results in an industry-wide shift toward reduced packaged detergents. We also show that retailers, with varying degrees of market power, may select higher ratios of reduced packaged detergents and achieve convex levels of sales of reduced packaged detergents.