Date of Graduation
5-2026
Document Type
Thesis
Degree Name
Bachelor of Science in Data Science
Degree Level
Undergraduate
Department
Data Science
Advisor/Mentor
Dr. Karl Schubert
Committee Member
Dr. David Barrett
Second Committee Member
Ms. Lee Shoultz
Abstract
The thesis presents a data-driven approach to identifying and quantifying product substitution effects within Walmart’s fashion apparel retail sector. Utilizing a comprehensive dataset of store-level sales and shelf fixture counts, this research develops a methodology to detect space cut events, defined as a 25% or greater reduction in a fineline’s relative shelf space share, and analyzes the subsequent shifts in consumer purchasing behavior.
To ensure comparability across diverse store formats, the study normalizes raw data into proportional space and sales shares. By correlating reductions in shelf space for specific finelines with sales gains in neighboring categories within the same Walmart week, the model identifies the primary substitute categories for unique apparel finelines.
The analysis compares observed sales changes during space cut events against historical baseline performance to measure substitution lift. Bootstrapping techniques are applied to estimate confidence intervals around the measured lift values, allowing the study to determine whether substitution effects are statistically meaningful.
These findings provide retail practitioners with a predictive framework for optimizing assortment strategies and shelf-space allocation while minimizing total revenue loss during inventory transitions.
Keywords
retail, sales analysis, fineline, substitutability, retail analytics
Citation
Stepanova, D. (2026). A Data-Driven Framework for Modular Category Optimization in Retail: Fineline-Level Insights for Walmart Apparel. Data Science Undergraduate Honors Theses Retrieved from https://scholarworks.uark.edu/dtscuht/42
