A New Bayesian-Based Model to Demand Forecast and Inventory Reduction

Abstract

<span>Natural calamities (e.g., hurricane, excessive ice-fall) may often impede the inventory replenishment during the<br /><span>peak sale season. Due to the extreme situations, sales may not occur and demand may not be recorded. This study<br /><span>focuses on forecasting of intermittent seasonal demand by taking random demand with a proportion of zero values<br /><span>in the peak sale season. Demand pattern for a regular time is identified using the seasonal ARIMA (<span><em>S-ARIMA</em><span>)<br /><span>model. The study proposes a Bayesian procedure to the ARIMA (<span><em>BS-ARIMA</em><span>) model to forecast the peak season<br /><span>demand which uses a dummy variable to account for the past years intermittent demand. To capture uncertainty in<br /><span>the <span><em>B-ARIMA </em><span>model, the non-informative prior distributions are assumed for each parameter. Bayesian updating is<br /><span>performed by Markov Chain Monte Carlo simulation through the Gibbs sampler algorithm. A dynamic<br /><span>programming algorithm under periodic review inventory policy is applied to derive the inventory costs. The model<br /><span>is tested using partial demand of seasonal apparel product in the US during 1996-05, collected from the US Census<br /><span>Bureau. Results showed that, for intermittent seasonal demand forecast, the <span><em>BS-ARIMA </em><span>model performs better and<br /><span>minimizes inventory costs than do <span><em>S-ARIMA </em><span>and modified Holt-Winters exponential smoothing method.</span></span></span></span></span></span></span></span></span></span></span><br /><br class="Apple-interchange-newline" /></span></span></span></span></span></span></span></span></span></span></span>