Open Collections

Adaptive inventory control heuristics for non-stationary demand

University of British Columbia

The objective of the current research is to develop an inventory algorithm that determines the ordering periods and the corresponding order quantities for nonstationary demand. The inventory system is a periodic review with lost sales. We propose a heuristic, Wagner Whiting Plus Forecast (WWPF), in which the forecasts are revised and the parameters for inventory control policy parameters are updated periodically. The demand process is non-stationary with a linear trend. The cost function is constituted by a fixed setup cost and a proportional holding cost. In each period, safety stocks are added to the forecast and the dynamic lot sizing is done as per the Wagner-Whitin algorithm. The proposed heuristic is compared with an adaptive (s, S) policy proposed by Axsater (2000). Both WWPF algorithm and Axsater's heuristic determine inventory parameters for demand data with trend, in a reasonable way. WWPF algorithm exhibits a marginal improvement over Axsater's heuristic and can be recommended for inventory control in practical settings. WWPF algorithm can address seasonality, by using seasonal forecast models, such as Holt-Winters. Moreover, WWPF algorithm is independent of the forecasting method and it can be modeled with other forecasting methods, too.

Thesis/Dissertation

application/pdf

2009-10-07

For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use.

http://hdl.handle.net/2429/13738

Business Administration

University of British Columbia

UBCV

DSpace