Finding the sweet spot– UNB researcher finds optimal locations for retail firms
Author: Faculty of Management
Posted on May 30, 2023
It is no secret that location is a major contributor to success for retailers. Dr. H. A. Eiselt, a quantitative methods professor with UNB’s faculty of management, is collaborating on a series of studies that applies gaming theory and customer behaviour to help retailers find the best locations. The gaming theory is based on the work of von Stackelberg, an economist who lived during the first half of the 20th century and whose work contributed to cost theory and sequential games in duopolies.
Big brands make good neighbours, it turns out. Eiselt and his colleagues recently published two studies in the European Journal of Operational Research that use the Stackelberg leader – follower game to determine optimal locations for retailers. “All models are put in the context of a so-called von Stackelberg game that feature a leader [an established facility] and a follower in a scenario, in which each player has some variables at its disposal. In our work, we are mainly interested in the location of firms,” says Eiselt. “The leader locates its firms first, then the follower firm will locate its firms. The game is asymmetric: while the follower only needs to observe the decisions (here, locations) of the leader, the leader will have to guess what the reaction of the follower will be, given each of its own decisions.” In determining optimal locations for individual firms, one question is whether rational behavior by customers and firms contribute to the agglomeration of facilities.
The first study, entitled “The leader multipurpose shopping location problem,” investigates the locations chosen by leaders based on where customers engage in multipurpose shopping, visiting multiple stores to purchase multiple nonessential products. Eiselt explains, “This model includes leader decisions based on foresight. Depending on specifics regarding customer behavior, agglomeration may occur, a feature that can be observed in practice in many types of businesses, for example, the fast-food industry.”
The second study, "Store location with multipurpose shopping trips and a new random utility customers" choice model, develops a new decision rule for customers that more closely resembles actual customer decision making than previous choice rules. According to their study, firms that observe the importance that customers put on the observable attributes of different products, have a clear decision rule that tells them to either co-locate with competitors or to choose individual locations away from the competition.
Based on observations of consumer behaviour and their predictions of how retail follower’s are likely to react, Eiselt and his team’s research can help leaders in the retail industry determine the best locations for their operations.
Eiselt joined the faculty of management in 1986 and is a member of the quantitative methods area. He teaches courses in quantitative analysis and location science and his research interests include competitive location models, and multi-criteria decision analysis. His other recent publications have looked at locations for landfills, hubs, and hospitals, and has applied mathematical measures to counterterrorism.
PHOTO: Dr. H. A. Eiselt, a quantitative methods professor with UNB’s faculty of management, is collaborating on a series of studies that applies gaming theory and customer behaviour to help retailers find the best locations.
Learn more about Dr. H. A. Eiselt and about UNB’s faculty of management.
Media contact: Liz Lemon-Mitchell