Consumer Driven Forecasting to Improve Inventory Flow: Brown Shoe Company's Journey
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Brown Shoe Company, Inc. is a leading supplier of footwear with 12,000 employees and annual revenues of $2.4 billion. Based in St. Louis, Missouri, this consumer-focused company operates over 1,400 retail shoe stores in the United States, Canada, and China under the names of Famous Footwear, Naturalizer, Brown Shoe Closet, FX LaSaIIe, Franco Sarto, and Shoes.com (its e-commerce subsidiary). Its brands, which include Naturalizer, Life Stride, Franco Sarto, Via Spiga, Carlos by Carlos Santana, Etienne Aigner, Dr. Scholl's, Nickels Soft, and Buster Brown, are sold at over 2,000 department stores, mass-merchandisers, independent, and specialty stores worldwide. BACKGROUND With 90 million pairs of shoes sourced annually, orders were previously placed in overseas factories. Orders were delivered as we placed them, but there was no systematic process or mechanism to make adjustments to orders based on consumer demand. This was simply because we did not have an effective way of knowing how our shoes were selling at retail. We were limited to ad hoc reporting and emailing order changes, but we were not reacting to consumer demand. There was no systematic way of updating forecasts based on consumer activity. To become more consumer-driven, a forecasting process was needed to synchronize consumer demand with factory operations so that we could service retail customers more effectively. We needed to capture information about our consumer's purchases, which could then be used for our sourcing and inventory management. In July 2004, Brown Shoe began its journey by manually feeding Point of Sales (POS) information obtained from its 1, 100 Famous Footwear stores into a single PC. We wanted to see what kind of forecast we could generate and what we could do with that information. It turned out to be a great success. Forecasts improved, stockouts went down, and sales went up. At that point, we decided to look for a formal system by which we could automate the entire process. The firm chose Logility Voyager Solutions, in part because of its ability to handle highly seasonal, short life cycle merchandise, since 75% of our product offerings are new each season. During the four-month implementation, the biggest challenge was internal - the need for a huge amount of data to be organized and loaded into the new system. We had a lot of rich EDI Point of Sales history from our customer base, but it had to be organized in a way it could be useful to the company. With Logility 's system, we had highly accurate weekly forecasts of consumer sales, which improved our inventory flow to retailers. When a shoe is in high demand by consumers, the chance to sell more of it in-season requires a quick reaction. POS FORECASTING PROCESS Today, Brown Shoe closely collaborates with its Famous Footwear retail store chain on inventory planning by obtaining POS data on weekly sales in order to get retail replenishment as close to true market demand as possible. Retailers provide POS data by size, width and store, clearly pointing to which shoes are selling best. The system rolls forward weekly and recalculates the forecast for each item using the most recent week of POS we receive. The Demand Planning system generates forecasts for all of Famous Footwear's items. In addition, weekly POS feeds are a key driver for Brown Shoe's In-Season Replenishment (ISR) system. The system calculates item and size inventory needs and adjusts the receipt flow based on how the item is performing in stores. The company has been highly successful in forecasting shoes and their sizes. The improvements in forecast accuracy result in fewer quantity adjustments in purchase orders and allow for the factories to stage materials ahead of time. This reduced the lead-times by about 50 days for core Brown Shoe items selling at Famous Footwear because orders went from being monthly to every other week resulting in smaller quantities. For every style and color of footwear, we have typically 12 to 18 sizes. …
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.006 | 0.004 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it