Learning Curve and Rate Adjustment Models: Comparative Prediction Accuracy Under Varying Conditions

Abstract

Learning curve models have gained widespread acceptance as a technique for analyzing and forecasting the cost of items produced from a repetitive process. Considerable research has investigated augmenting the traditional learning curve model with the addition of a production rate variable, creating a rate adjustment model. This study compares the predictive accuracy of the learning curve and rate adjustment models. A simulation methodology is used to vary conditions along seven dimensions. Forecast errors are analyzed and compared under the various simulated conditions, using ANOVA. As in all simulation studies, findings must be interpreted in light of the assumptions underlying the simulation. Overall results indicate that neither model dominates; each is more accurate under some conditions. Conditions under which each model tends to result in lower forecast errors are identified and discussed. This work was sponsored by the Cost Estimating and Analysis Division of the Naval Sea Systems Command and the Naval Postgraduate School.