This paper investigates the predictive power of project cost data in earned value management (EVM) as an early indicator of the cost overrun probability in risk management. The predictive power of the cost performance index (CPI) is probabilistically assessed and used (1) to update the cost overrun probability and the estimate at completion (EAC) distribution and (2) to visualize all possible CPI trajectories to project completion. Specifically, this paper presents two decision support tools: a probabilistic EAC (P-EAC) model and a CPI trajectory simulator for visual risk communication. The predictive models were applied to a real project and computational experiments were conducted. The results indicate that a deterministic CPI measurement, for instance, CPI = 0.85 at a 20% completion point, may indicate a wide range of possible cost overrun probabilities from 54 to 100% according to the predictive power of cost data. Improved risk awareness from the proposed analytics can be a vital element for enhanced management visibility and more informed decision-making in project control.
|Original language||English (US)|
|Journal||Journal of Management in Engineering|
|State||Published - Feb 1 2019|
All Science Journal Classification (ASJC) codes
- Industrial relations
- Strategy and Management
- Management Science and Operations Research