Spical forecast-error measures such as mean squared error, mean absolute deviation, and bias generally are accepted indicators of forecasting performance. However, the eventual cost impact of forecast errors on system performance and the degree to which cost consequences are explahed by typical error measures have not been studied thoroughly. The present paper demonstrates that these typical error measurn often arc not good predictors of cost consequences in material requirements plarming (MRP) settings. MRP systems rely directly on the master production schedule (MPS) to specify gross requirements These MRP environments receive forecast cmrs indirectly when the errors create inaccuracies in the MPS.Our study results suggest that within MRP environments the predictive capabilities of forecast-error measures are contingent on the lot-sizing rule and the productcomponents structure When forecast errors and MRP system costs are coanalyzed, bias emerges as having reasonable predictive ability. In further investigations of bias, loss functions are evaluated to explain the MRP cost consequences of forecast errors. Estimating the loss functions of forecast errors through regression analysis demonstrates the superiority of loss functions as measures over typical forecast error measures in the MPS.Subject Alaas: F m t i n g , Materhl Requlnmcnts Pbnning, prrfonnancc Evaluation, and Simulation.

Recent findings in production research have demonstrated that bias in forecasts may improve system performance in a simulated multiple-product MRP system, and optimal planned bias can be identified for simulated single product MRP settings. This paper provides an analytical approach and theoretical evidence that optimal planned bias exists to minimize the expected loss for the traditional EOQ model, as well as for the Newsboy model, when relaxing the assumption of known demand rate in the EOQ model and the assumption of known boundary of demand distribution in the Newsboy problem. Findings of this paper also provide a rationale for impact of forecast errors on cost behaviour, and facilitate a benchmark for comparing different findings of previous research work.