RetailNDA-safe case study

Assortment Optimization for a Global Apparel Retailer

Built Python-based assortment optimization and revenue forecasting models for a global retailer's China operations, paired with executive dashboards.

Client

Levi's (via Lynx Analytics)

Duration

8 months

Stack

PythonBigQueryPower BITime Series

01. The Problem

Client needed to forecast demand across thousands of SKUs and optimize store-level assortment — existing methods relied on spreadsheets and gut-feel.

02. The Approach

Designed a hybrid ML pipeline combining time-series forecasting with assortment optimization heuristics. Built a Power BI layer so regional managers could explore forecasts interactively.

03. The Outcome

Reduced forecasting error significantly versus baseline; decisions moved from monthly cycles to weekly. Adopted by multiple regional teams.

Have a similar challenge?

I bring the same rigor and outcome-focus to every engagement — from enterprise clients to focused startup work.

Start a conversation