Title: Assortment Optimization under Customer-driven Substitution

Date: August 18, 2025

Time: 12:30 pm – 2:00 pm (EST)

Meeting Link: Microsoft Teams Meeting

Jisoo Park

Ph.D. Candidate in Industrial Engineering

School of Industrial and Systems Engineering

Georgia Institute of Technology

Advisors: 

Dr. Benoit Montreuil, School of Industrial and Systems Engineering, Georgia Tech

Dr. Walid Klibi, Supply Chain Center of Excellence, KEDGE Business School

Committee:

Prof. Xin Chen, School of Industrial and Systems Engineering, Georgia Tech

Prof. Chelsea “Chip” White III , School of Industrial and Systems Engineering, Georgia Tech

Prof. Andre Calmon, Scheller College of Business, Georgia Tech

Abstract:

Retailers face growing complexity in assortment planning due to blurred boundaries between sales and fulfillment channels, proliferating product lines, and increasing supply chain uncertainty. Customer substitution behavior—the propensity of customers to select alternative products when preferred items are unavailable—serves as a critical lever for improving assortment decisions in capacitated and volatile environments. This dissertation develops optimization- and simulation-based decision support models that explicitly integrate substitution effects into assortment planning at multiple decision levels.

In Chapter 2, we introduce the showroom assortment optimization problem for omnichannel retail. Motivated by the transformation of offline stores into experiential showrooms, the problem is defined as selecting an in-store assortment that maximizes customers’ purchase confidence following a store visit while adhering to capacity constraints. In this context, substitution behavior differs from traditional retail settings, as purchase decisions are shaped by in-store experiences rather than immediate product availability. We formulate the underlying optimization problem as a MIP that captures how customers gain purchase confidence through surrogate products—items that represent or stand in for other products—thereby modeling the customer substitution effect in the showroom environment. Using data from an industry partner, we demonstrate how the approach can design effective showcasing strategies that improve purchase confidence and sales.

Chapter 3 examines the profit-maximizing assortment planning problem under supply and demand uncertainty for a retailer-manufacturer operating in a make-to-stock environment. We develop a stochastic choice-based optimization model that endogenizes customer-driven substitution behavior through rank-based choice model approximations of stockout-driven substitution probabilities. The model extends multi-period assortment planning by integrating assortment decisions with upstream supply availability and production capacity constraints, and downstream fulfillment dynamics. To ensure tractability, we design a rolling-horizon algorithm with lookahead, where each planning window solves a two-stage stochastic program via Benders decomposition. Case study results demonstrate that the approach solves large-scale industry instances efficiently and delivers substantial improvements in profit and service levels, with benefits amplified under volatile supply chain conditions.

In Chapter 4, we develop a hybrid simulation-optimization framework to refine tactical assortments through operational adjustments at finer review intervals. The proposed framework integrates a MILP with a high-fidelity multi-agent system (MAS) that simulates detailed operational dynamics, including stochastic demand, inventory flows across multiple echelons, and demand fulfillment processes. In each review period, the MILP proposes state-contingent adjustments, which the MAS evaluates under realistic operating conditions. A feedback-driven neighborhood search iteratively improves solutions to ensure both operational feasibility and profitability. Computational experiments with industry data show measurable gains in profit and service performance compared to static tactical plans.