Skip to Content
Author's profile photo Stephan Kolassa

SAP and the Science of Forecasting

Forecasting future customer demand at the Point of Sale (POS) is possibly the most important aspect of successful Retail tactics and execution: whether we want to optimize prices, plan promotions, set up planograms or replenish store shelves, we always need to look into the future and understand what our customers will want tomorrow, next week, next month or next year! Fortunately, SAP has a lot of people working on the art and craft of forecasting – and we stay in touch with the newest developments in the scientific community.


This year, the International Symposium on Forecasting (ISF), the most important annual scientific conference on all aspects of forecasting, was held on June 24-27 in Boston, MA, USA. Over 420 participants from academe, industry and consulting met to discuss newest advances in the theory and practice of forecasting. New this year was a dedicated practitioners’ track, which attracted 100 forecasting practitioners from diverse industries.


SAP and affiliated attendees included Tim Januschowski from the SAP Innovation Center, Martin Lorenz from the Hasso Plattner Institute, and Roland Martin and my humble self from the Center of Excellence Forecasting and Replenishment, SAP (Switzerland) AG, Tägerwilen. Tim and Martin gave a talk on “Demand Forecasting with Partial POS Data using In-Memory Technology”, introducing participants to what SAP HANA can do for them in the field of big data forecasting. Roland presented a talk on “Percentage Errors can Ruin your Day – and Rolling Dice Shows Why”, giving a simple example to illustrate why widely used forecasting error measures may be misleading under certain circumstances. This talk was based on an article Roland and I recently published in the practitioners’ journal Foresight. In addition, Roland gave a very well received pre-conference workshop on “Forecasting to Meet Demand”, discussing what we have learned in our long experience with forecasting large numbers of demand time series in a Retail environment.


On the weekend prior to the conference, I attended the meeting of the Board of Directors of the International Institute of Forecasting, the scientific body that organizes the ISF and coordinates other aspects of forecasting research, having been elected to the Board of Directors in 2011 for a term lasting until 2014. I could not attend the second day of the ISF, switching instead to the SAP Forecasting and Replenishment Forum, which was held in Boston on June 26-27. I presented the unified demand forecast, which will bring together the forecasting expertise formerly divided between the forecasting methodologies used in Forecasting & Replenishment on the one side and Promotion Management for Retail (PMR) and Trade Promotion Optimization (TPO) on the other side. The SAP Forecasting and Replenishment Forum will be discussed here in a separate entry soon – stay tuned!


The 2013 ISF will be held on June 23-26 at KAIST Business School in Seoul, South Korea. We are all looking forward to once again showcasing what SAP has to offer in the field of forecasting – for Retail and the rest of the world!

Assigned Tags

      1 Comment
      You must be Logged on to comment or reply to a post.
      Author's profile photo Mohamed Amer
      Mohamed Amer

      Great to learn more about SAP's leadership role in both the scientific and the practical business usage of Forecasting!