Applied AI Methods for Retail

This unit introduces core AI methods (supervised, deep, and reinforcement learning) and applies them to store-level retail problems including computer vision and merchandising. Students will plan and run POC bake-offs, perform exploratory analyses using a one-store-at-a-time approach, and prepare operational plans to execute AI-enabled merchandising solutions.

Learning Objectives

• Analyze and compare supervised learning, deep learning, and reinforcement learning approaches for specific retail problems (e.g., demand forecasting, shelf analytics, pricing), including typical data requirements, performance trade-offs, and deployment constraints
• Apply a one-store-at-a-time exploratory workflow to produce data readiness artifacts (feature plans, training/validation splits, and sample visualizations) for a chosen retail use case
• Design and execute a POC bake-off plan that defines baselines, success metrics, evaluation protocols, and roles for AI specialists, and demonstrate selection of the best-performing model using sample or simulated data
• Evaluate the operational, ethical, and business implications of deploying AI solutions (including computer vision and autonomous fulfillment) and produce an implementation roadmap addressing staffing, ROI, and change management