Publication Date
2020
Document Type
Dissertation
Committee Members
Pratik Parikh, Ph.D. (Advisor); Xinhui Zhang, Ph.D. (Committee Member); Subhashini Ganapathy, Ph.D. (Committee Member); Nan Kong, Ph.D. (Committee Member); Amir Zadeh, Ph.D. (Committee Member)
Degree Name
Doctor of Philosophy (PhD)
Abstract
Although the retail business has been exploring innovative ways to engage shoppers, the COVID-19 pandemic has sped up their effort. Because of its unique benefits, physical stores will continue to remain an integral part of the overall retail business. However, to stay competitive, retailers will be forced to effectively utilize their available space in physical store (and even reduce it if need be), while offering a reasonably large assortment of products on their shelves. For many such retailers, the design of planograms – visual representation of products on shelves – is still driven by prior experience and intuition. Further, existing optimization-based planogram design approaches assume that the shelf length and height are fixed, which often result in unused space on the planogram or suboptimal assignment of SKU facings, both resulting in reduced revenue for the retailer. To address this real-world challenge, we introduce the joint shelf design and shelf space allocation (JSD-SSA) problem to maximize retailer’s revenue. Our proposed mathematical programming model for JSD-SSA determines the optimal shelf design, while determining SKU placement and facings, under SKU family constraints. Because realistic problem sizes pose significant computational challenges in solving this model, we propose a decomposition-based approach. Accordingly, we first partition the planogram area and allocate it to each SKU family via a Particle Swarm Optimization heuristic, then for each partition we determine the shelf design (number of shelves, shelf coordinates, length, and height) and shelf space allocation (SKU placement and facings) using Constraint Programming. In so doing, real-world problem instances (2 families, 100 SKUs total, 192” × 84” planogram size) could be solved within 45 minutes. We also propose a metric to measure variation in SKU shapes within and between SKU families. Our experiments indicate that shorter shelf lengths can increase retailer’s profit by up to 22% depending on the SKU-family shape variation. Higher within-family shape variation can result in higher revenue increases. Further, as the planogram becomes tighter (measured via space tightness), the benefits of shorter shelf lengths increase. Additionally, if SKU and planogram dimensions share a common factor or multiple, then more compact planograms can be designed, in turn reducing unused space and increasing retailer’s profit. We strongly believe that our optimization-based approach will allow retailers to fully utilize the available shelf space, especially during post COVID-19 environment where retailers may opt to reduce their store footprint. Better SKU allocation on highly visible shelf locations will allow better shopper-SKU interaction, in turn reduce expensive trial-and-errors. Our approach will also allow benchmarking of existing and alternative planogram designs depending on the location of the department in the store and corresponding shopper traffic.
Page Count
55
Department or Program
Ph.D. in Engineering
Year Degree Awarded
2020
Copyright
Copyright 2020, some rights reserved. My ETD may be copied and distributed only for non-commercial purposes and may not be modified. All use must give me credit as the original author.
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.
ORCID ID
0000-0001-8536-5882
