Title: Identification and classification of intermittent demand patterns

Authors: Pallavi Chitturi, Mark Gershon, Jing Chen, John Boyarski

Addresses: Department of Statistics, Temple University, Philadelphia, PA 19122, USA. ' Department of Management Science and Operations Management, Temple University, Philadelphia, PA 19122, USA. ' Department of Statistics, Temple University, Philadelphia, PA 19122, USA. ' C.A.C.I. Inc., Mechanicsburg, PA 17055, USA

Abstract: Intermittent demand is often defined as random demand with a large proportion of zero values. However, the terms |intermittent demand|, |lumpy demand| and |erratic demand| are used interchangeably in the literature. There is also a widely held misconception that low demand items are, by definition, intermittent in nature. In this paper, we show that the term intermittent demand includes at least three distinct sub-groups of demand patterns that we call lumpy, limited and erratic. We define the three distinct demand patterns and provide a step-by-step method for classifying items as members of each demand pattern. We provide examples of each demand pattern using inventory data from the US Navy. Finally, we discuss some preliminary strategies for dealing with each distinct demand pattern.

Keywords: intermittent demand; lumpy demand; erratic demand; limited demand; moving average; forecasting; demand pattern identification; demand pattern classification; random demand; low demand; inventory management; US Navy.

DOI: 10.1504/IJPQM.2010.035118

International Journal of Productivity and Quality Management, 2010 Vol.6 No.3, pp.304 - 317

Published online: 05 Sep 2010 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article