A retailer with over 200 shops in the Netherlands and 2500 different items in stock
Towards the end of a season stock in the central distribution center runs out for some items, while low performing shops sometimes still have stock left. This leads to lost sales in the high performing shops because they cannot be replenished.
The company has an overall replenishment policy for resupplying a shop. A high performing shop can sell more items than the standard replenishment policy, leading to lost sales in between supply intervals.
We built a model that can recognize high and low performing shops. Based on this ranking the model makes a suggestion as to which shop should return their stock when the distribution center runs out of inventory. This leads to less left over stock at the low performing shops and increased sales because there was more inventory available for the high performing shops.
- Due to redistribution of stock there was 20% less unsold items at the end of the season and an increased overall sales of 8%
- The improved replenishment policy resulted in a 5% increase of overall sales