Project Cost Overrun Risk Prediction Using Hidden Markov Chain Analysis

Construction project cost overrun is a common problem in the construction industry. The of projects thought to have increased by approximately 33% on average. Several types research been conducted and these generally rely historical data. However, whilst each has its own characteristics trend, real-time data are more reliable forecast trend. This paper proposes hidden Markov chain (HMM) model predict risk based project-owned performance corrective actions if adopted. events occurrence this was assumed follow Poisson arrival pattern. Real-time HMM with particle filter used run simulation. One SRC building Taiwan for validation comparison. posterior probabilities from were highly consistent ratios real projects. proposed prediction could provide an early alert overruns manager. Based survey significantly influential factors, we propose effective management plans alleviate frequency overrun.