Production smoothing policies: The case of ARMA demand
Abstract
We investigate the performance of two production smoothing policies. Production smoothing models are designed to minimise the weighted sum of inventory variance (to capture the costs of inventory holding and backlog) and the change in orders from period to period (to capture production adaptation costs). The first policy is the classic production smoothing model proposed by Magee (1958). In this policy, Magee used a convex combination of the classic order-up-to policy and the previous order. The second policy is the so-called “full-state” order-up-to policy that has been adapted to become a production smoothing policy in the same manner as Magee adapted the order-up-to policy. We show that the full-state production smoothing policy outperforms the original policy by Magee.