Publication Date

2019

Document Type

Thesis

Committee Members

Pratik Parikh, Ph.D. (Advisor); Subhashini Ganapathy, Ph.D. (Committee Member); Corinne Mowrey, Ph.D. (Committee Member)

Degree Name

Master of Science in Industrial and Human Factors Engineering (MSIHE)

Abstract

For brick-and-mortar retailers to be successful, it is critical for them to optimally design their rack layout and place products in order to draw attention of shoppers. Literature suggests that racks placed at acute (or obtuse) angles to the main aisle frequented by shoppers can enhance visibility of products compared to racks placed orthogonally (i.e., 90˚). Placing products with high impulse purchase potential in the resulting highly visible locations on the rack can increase shopper impulse purchases. However, placing racks at angles other than 90˚ can increase the required floor space. Additionally, while reducing the height of the racks just below eye-height enhances visibility, it, however, reduces the number of available locations per product and increases restocking costs. To effectively trade off the benefits of visibility (in turn, impulse profit) and limitations of space and restocking costs, we propose the Joint Rack Configuration and Shelf Space Allocation (JRC-SSA) problem. The JRC-SSA jointly determines rack decisions (orientation and height) and product decisions (placement and number of locations) in order to maximize a retailer’s impulse profit (after discounting for space and restocking costs). As JRC-SSA is an extension of the classical SSA that has been shown to be NP-hard, and that the visibility estimation is not in a closed analytical form, standard mathematical programming solvers are not suitable. Consequently, we employed the population-based Particle Swarm Optimization (PSO) framework and designed five subroutines to efficiently find a (near) optimal solution to the JRC-SSA. Using realistic data collected from a major US retailer and that available in the existing literature, we conducted a comprehensive experimental study to derive managerial insights. Results indicate that product decisions were impacted by the angle of the rack; if a high impulse product was placed on the front face near to the endcap in a 90˚ rack, the same product was now placed on the back face in an acute-angled rack. We also noticed that acute-angled racks increased impulse profit over 90˚ racks at low space costs; shorter racks were prominent for low restocking costs. Overall, configurations exist where a retailer can realize up to 8.2% increase in profit through the JRC-SSA compared to a 7 ft height rack placed at 90˚ orientation. We expect that these, and several other insights discussed in our study, will help retailers in quantitatively evaluating their current rack designs and product placements, and optimize them, to increase shopper experience and, in turn, impulse profit.

Page Count

50

Year Degree Awarded

2019


Share

COinS