Webinar: Rocket science and retailing - mixing art and science
January 14, 2014
What stock level for a shop item is optimal for providing the customer with a good service? To answer this question and obtain an insight into the costs involved in good service, the traditional stock theory was tested in practice. The applied mathematical models are programmed in open source software and usually not available in standard ERP systems. Floris Padt performed a bypass based on the theory and discussed it in her webinar on 16 January.
Image: © Nationale Beeldbank
Why does Hunkemöller want to better respond to customer demand?
"Offering World Class Service is part of Hunkemöller's vision. This vision comprises three components, namely becoming the number one lingerie specialist in Europe, being a high street label and offering its customers a world class service .
World class service means for example that 95% of the time, the desired product must be available in-store; it concerns products that are available all year round, or ‘Never-Out-of-Stock products’ (NOS). In total over 0.5 million supply points."
Open source software
What is done to improve the balance between availability and stock costs?
"The traditional, single level theory of how to achieve the right service level, is already fifty years old, but often not available in the ERP system. Before purchasing additional systems we wanted to acquire an understanding of the challenge based on our sales, the required models and the business case.
To do so we used ‘open source software’: R, in which the desired models are often available or easy to construct. The theory can ultimately be applied in practice, without too much effort, in order to test the business case against extremely limited costs and without any risk of a disinvestment."
What is the result?
"Good question, because it is important to know before you make an investment. The stock cost savings must be weighed against the costs of the chosen solution. Safety stock can be cheaper than predicting with all the factors that influence demand. Take for example the weather, March 2013 was the coldest month in twenty years and that is clearly reflected in our sales figures.
The model indicates that ‘gross sales’ could increase by 4% with the same cost by better allocating the same stock. During a three-month pilot we increased the service level by 3.5% for 350,000 supply points by improving store stock levels. The upstream service level is the next challenge."