A New Fuzzy Earned Value Management Method for Incorporating Project Data Uncertainty

Abstract

In projects, regular assessment of the progress and cost works well to detect the deviations from the plans and take corrective actions on time. To monitor the project performance and predict the completion cost and duration of the project, earned value management (EVM) has been commonly used by project managers. The implementation of EVM relies on collecting precise data on cost realizations and activity completion rates. However, in practice the completion rates cannot be determined exactly in many cases, and they are often estimated by the project team. Traditional EVM does not consider the inherent uncertainty caused by subjective assessments in information gathering and assumes that the input data are certain. In this research, we develop a fuzzy-based EVM method that incorporates the uncertainty caused by inexact measurement of percentage completeness and actual cost of project activities. The method makes use of linguistic terms to estimate the inexact values and applies fuzzy mathematics to quantify the inexact measurements. It also considers the errors caused by the judgments of project team. The usability of the method is demonstrated with computational analysis of the performance measurement of a real project.

Keywords: uncertainty, fuzzy numbers, earned value management, project performance evaluation