Level:
Advanced
TCM Section(s):
7.6. Risk Management
10.2. Forecasting
Venue: 2022 AACE International Conference & Expo
Abstract: Infrastructure projects are becoming increasingly complex given governments' demand for building innovative and smart cities. Forecasting a realistic schedule contingency is a major concern for managing this class of projects. Existing scheduling models often run based on subjective data collected from members of project delivery team (PDT) without incorporating historical data of the project. The confidence level of the result heavily relies on the experience of the PDT and their subjective judgment. Accordingly, this paper presents a data-driven quantitative scheduling risk analysis model that employs reported schedule performance data to the project in-hand in combination with a set of schedule risk drivers for forecasting project schedule contingency. It utilizes earned value management (EVM) and risk drivers methods to account for correlation between activities’ duration, which is a drawback of the risk-based EVM methods reported in the literature. A real case example in the transit sector is utilized to compare the performance of the developed model with commercially available scheduling software. The results indicate that the developed model yielded schedule contingency close to that generated by the software, but with considerably less effort and less demand on data preparation and processing.
TCM Section(s):
7.6. Risk Management
10.2. Forecasting
Venue: 2022 AACE International Conference & Expo
Abstract: Infrastructure projects are becoming increasingly complex given governments' demand for building innovative and smart cities. Forecasting a realistic schedule contingency is a major concern for managing this class of projects. Existing scheduling models often run based on subjective data collected from members of project delivery team (PDT) without incorporating historical data of the project. The confidence level of the result heavily relies on the experience of the PDT and their subjective judgment. Accordingly, this paper presents a data-driven quantitative scheduling risk analysis model that employs reported schedule performance data to the project in-hand in combination with a set of schedule risk drivers for forecasting project schedule contingency. It utilizes earned value management (EVM) and risk drivers methods to account for correlation between activities’ duration, which is a drawback of the risk-based EVM methods reported in the literature. A real case example in the transit sector is utilized to compare the performance of the developed model with commercially available scheduling software. The results indicate that the developed model yielded schedule contingency close to that generated by the software, but with considerably less effort and less demand on data preparation and processing.