Title: Evaluation of hierarchical forecasting for substitutable products

Authors: S. Viswanathan, Handik Widiarta, R. Piplani

Addresses: Nanyang Business School, Nanyang Technological University, 639798, Republic of Singapore. ' Nanyang Business School, Nanyang Technological University, 639798, Republic of Singapore. ' School of Mechanical & Aerospace Engineering, Nanyang Technological University, 639798, Republic of Singapore

Abstract: This paper addresses hierarchical forecasting in a production planning environment. Specifically, we examine the relative effectiveness of Top-Down (TD) and Bottom-Up (BU) strategies for forecasting the demand for a substitutable product (which belongs to a family) as well as the demand for the product family under different types of family demand processes. Through a simulation study, it is revealed that the TD strategy consistently outperforms the BU strategy for forecasting product family demand. The relative superiority of the TD strategy further improves by as much as 52% as the product demand variability increases and the degree of substitutability between the products decreases. This phenomenon, however, is not always true for forecasting the demand for the products within the family. In this case, it is found that there are a few situations wherein the BU strategy marginally outperforms the TD strategy, especially when the product demand variability is high and the degree of product substitutability is low.

Keywords: hierarchical forecasting; exponential smoothing; substitutable products; production planning; product families; simulation; demand variability.

DOI: 10.1504/IJSOM.2008.017295

International Journal of Services and Operations Management, 2008 Vol.4 No.3, pp.277 - 295

Published online: 24 Feb 2008 *

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