Abstract

Earned Value Management (EVM) is a well-known project management technique for monitoring project progress. Over the last 15 years, many promising EVM metrics have been proposed to get, among other improvements, better actual project duration and cost estimates. Papers comparing the performance of all these metrics are, however, scarce and sometimes contradictory. In this paper, a simulation and empirical comparison of 26 deterministic project duration forecasting techniques under the EVM framework is developed. Among them, two new metrics: Earned Schedule min (ESmin) and Earned Schedule max (ESmax) are proposed. ESmin and ESmax offer a new and simpler activity-level calculation approach of the traditional Earned Schedule metric. Top performing (most accurate) metrics: Earned Schedule (ES), Earned Duration (ED) and Effective Earned Schedule (ES(e)) with Performance Factor 1 (PF = 1), are slightly outperformed by the new metrics which also offer some interesting applications for enhanced project control.