An Assess of Operational Project Performance by Integrating Earned Value Management and Learning Curve Theory

Abstract

This study adopts an integrated performance measurement and prediction model based on a combination of earned value management approach and the learning curve theory under risk condition. The research has two main parts: the first part concerns the project performance measurement and the second part focuses on forecasting performance indicators in terms of time and cost of the project subject to the errors and risks. The contributions of the present study are threefold. First, this study extends to the traditional performance measurement model, which focuses only on forecasting time at completion, by extending the performance measurement domain to analyse both time and cost. Secondly, the learning curve models are explicitly used as a basis to assess the nonlinear effect of learning on the performance. Novel risk performance metrics are proposed and adopted for knowledge-based projects. Thirdly, compared with classic deterministic and static performance measurement models, the proposed performance analysis employs the Kalman-Filter approach to predict the final time and cost performance accurately by taking into account the probabilistic risk factors and the errors in the performance forecasting procedures. The validity of the integrated performance measurement model is justified based on a case study. The computational results demonstrate that the developed performance measurement framework affords more accurate forecasts for the future performance than the traditional deterministic earned value methodology. The integrated performance measurement model developed in this study affords probabilistic prediction bounds and generates less errors than those achieved in classic EVM.