SVR-based method forecasting intermittent demand for service parts inventories

Authors:
Yukun Bao;Wen Wang;Hua Zou
Affiliations:
Department of Management Science and information System, School of Management, Huazhong University of Science and Technology, Wuhan, China;Department of Management Science and information System, School of Management, Huazhong University of Science and Technology, Wuhan, China;Department of Management Science and information System, School of Management, Huazhong University of Science and Technology, Wuhan, China
Venue:
RSFDGrC'05 Proceedings of the 10th international conference on Rough Sets, Fuzzy Sets, Data Mining, and Granular Computing - Volume Part II
Year:
2005

Citing 5
Cited 1

Forecasting error evaluation in material requirements planning (MRP) production-inventory systems

Management Science
The nature of statistical learning theory

The nature of statistical learning theory
A new cluster validity index for the fuzzy c-mean

Pattern Recognition Letters
Neural Networks: A Comprehensive Foundation

Neural Networks: A Comprehensive Foundation
On cluster validity for the fuzzy c-means model

IEEE Transactions on Fuzzy Systems

A differential evolution algorithm for joint replenishment problem using direct grouping and its application

Expert Systems: The Journal of Knowledge Engineering

Quantified Score

Hi-index	0.00

Visualization

Abstract

Intermittent Demand forecasting is one of the most crucial issues of service parts inventory management, which forms the basis for the planning of inventory levels and is probably the biggest challenge in the repair and overhaul industry. Generally, intermittent demand appears at random, with many time periods having no demand. In practice, exponential smoothing is often used when dealing with such kind of demand. Based on exponential smoothing method, more improved methods have been studied such as Croston method. This paper proposes a novel method to forecast the intermittent parts demand based on support vector regression (SVR). Details on data clustering, performance criteria design, kernel function selection are presented and an experimental result is given to show the method's validity.