SKU demand forecasting in the presence of promotions

  • Authors:

  • Özden Gür Ali;Serpil Sayın;Tom van Woensel;Jan Fransoo

  • Affiliations:

  • Koc University, College of Administrative Sciences and Economics, Rumeli Feneri Yolu, Sariyer 34450, Istanbul, Turkey;Koc University, College of Administrative Sciences and Economics, Rumeli Feneri Yolu, Sariyer 34450, Istanbul, Turkey;Technische Universiteit Eindhoven, School of Industrial Engineering, P.O. Box 513, NL-5600 MB Eindhoven, The Netherlands;Technische Universiteit Eindhoven, School of Industrial Engineering, P.O. Box 513, NL-5600 MB Eindhoven, The Netherlands

  • Venue:
  • Expert Systems with Applications: An International Journal
  • Year:
  • 2009

Quantified Score

Hi-index 12.05

Visualization

Abstract

Promotions and shorter life cycles make grocery sales forecasting more difficult, requiring more complicated models. We identify methods of increasing complexity and data preparation cost yielding increasing improvements in forecasting accuracy, by varying the forecasting technique, the input features and model scope on an extensive SKU-store level sales and promotion time series from a European grocery retailer. At the high end of data and technique complexity, we propose using regression trees with explicit features constructed from sales and promotion time series of the focal and related SKU-store combinations. We observe that data pooling almost always improves model performance. The results indicate that simple time series techniques perform very well for periods without promotions. However, for periods with promotions, regression trees with explicit features improve accuracy substantially. More sophisticated input is only beneficial when advanced techniques are used. We believe that our approach and findings shed light into certain questions that arise while building a grocery sales forecasting system.