The grocery store problem

Abstract

In today’s day and age, people care a lot about efficiency. When grocery shopping, people typically do not want to browse each aisle; rather, they want to move quickly throughout the store. However, stores certainly prefer for shoppers to browse aisles, as approximately seventy percent of purchases are not pre-planned (Sigurdsson et al., 2010)i. This leads to a conundrum, from now on referred to as The Grocery Store Problem: how to enable shoppers to quickly acquire the goods they desire while still enticing them to purchase other high-profit items to maximize customer satisfaction as well as profits. If stores know the path that their shoppers will take through the store, they can place items more strategically and increase profits. To accomplish this, stores could ask customers to input their grocery lists to a webpage or mobile app in advance of their shopping trip and, based on the list, provide an optimal path through the store to reach each item on the list. Stores can further increase profits by using the paths to guide customers past high-profit items that the customers are likely to purchase, based on the items they plan to purchase. The unique contribution of this paper is the innovative proposal to use graph algorithms in conjunction with data mining techniques to address the Grocery Store Problem, thus helping customers shop efficiently and increasing store profits.