How our Demand Forecasting Engine Helped a Retailer Prevent Stock-outs

Problem Statement

The client is a major player in the Fashion Retail Industry having >1000 stores across India. They had an inventory of >25,000 SKUs.

The ability to forecast demand has always been one of the key problem statements for players in the retail industry. Anticipating sales better, has a direct impact on procurement, logistics, store inventory levels and eventually sales as well. 

Approach

Demand in the fashion industry is driven by numerous factors. Some organizations even have specific design philosophies and they prefer to cater to that niche target audience only. Hence, this scenario is unlike a general demand forecasting exercise where a simple time-series model could suffice. We started the diagnostic phase for the project by interviewing store managers, procurement stakeholders and even a few designers to understand the intricacies involved.

Yugen’s team then broke the problem down into 4 parts:

1. Performing in-depth analysis to understand the features driving demand & procurement decisions
2. Developing pipelines to get the required features to the database
3. Developing a customized algorithm to forecast the demand for different product categories
4. Develop a monitoring & measurement system to ensure Quality of predictions and to identify drift

Solution

This system was designed as a post-event system which could be triggered as needed.

Data Engineering

We built ETL Pipelines which would bring in meta-data from various different DataWarehouses into a single DB. Scale of the data in this case wasn’t an issue however, the ability to manage various processes & pipelines was a big consideration while designing it. 

We also foresaw the need to put in a Data Quality Monitoring System which ensures that anomalies in the data are appropriately handled. The system also issues alerts to us and the stakeholders while showing if the alerts are critical or just warnings. 

Optimization Algorithm

The fully customised algorithm coupled with a rule-set based framework was developed which could:

1. Identify SKUs of a style which have depleted and recommend replenishment quantities based on it’s sell-through
2. Identify and maintain service levels for various pivotal & non-pivotal sizes for each store & attribute bucket combinations
3. Maintain Optimum Depth for SKUs across various attributes like Price Band, Sleeve Length, Occasion Type etc. per store 
4. Benchmark styles with it’s peers and identify dead/slow moving inventory in stores for pullback
5. Maintaining acceptable levels of Stock Cover accounting for Buffer Stock, Lead Times & Sell Through
6. Leverage True ROS, pivotal size availability, variant availability etc. for each store-style combination to arrive at recommendations for allocation of new styles to a store
7. Suggest replenishments to regional warehouses based on the sales velocity of SKUs in it’s catchment area.

Inference Pipeline (API)

We used AWS EventBridge to trigger the system deployed on AWS Lambda. The lambda can also be triggered via an API. Different sub-components of the system are modular in nature and can be triggered independently 

Impact

Post release:

1. Improvement in service levels of store inventory was observed (up ~4%-7% in different stores)
2. Lift of ~1%-1.5% in sales was observed due to reduction in number of stock outs driven by higher availability of fast selling sizes & colours
3. Reduction in % dead stock across different stores
4. Savings in terms of manpower & their bandwidth
5. Faster Turnaround and better management due to automated process