An experimental comparison of exponential and adaptive smoothing forecasting models using actual operating data

Adam Jr., E.E.; Berry, W.L.; Clay Whybark, D.

Computers-Industrial Engineering 2(2): 91-98


ISSN/ISBN: 0360-8352
Accession: 081723920

Download citation:  

Article/Abstract emailed within 1 workday
Payments are secure & encrypted
Powered by Stripe
Powered by PayPal

The forecast accuracy of exponential and adaptive smoothing models is compared using actual operating data. The data, involving the historical demand for twelve different medical supply items, was obtained from a large medical center. The adaptive smoothing model proposed by Trigg and Leach has been extended in this paper to include the continous adjustment of the smoothing constant values for the trend and seasonal factors in the forecasting model. The results indicate that while an adaptive smoothing model can reduce the bias in the forecasts in comparison with those produced by an exponential smoothing model, the adaptive smoothing model does not provide a significant reduction in the variation in the forecast errors (measured in terms of MAD) for most of the medical supply items studied.