A daily work report based approach for schedule risk analysis

Abstract

The Critical Path Method (CPM) has been the dominant scheduling method for the past several decades. Most State Departments of Transportation (DOTs) in the U.S. currently apply CPM to estimate project duration for major and complex projects. The method, however, has limitations, and one of them is its poor ability to analyze schedule risks. While previous studies have identified various factors that cause uncertainties in schedules, the estimated duration from CPM is deterministic. To quantify schedule risks, the Program Evaluation Review Technique which uses three-point estimates for activity duration and Monte Carlo Simulation (MCS) based methods that assign uncertainties to work activities through the probability distributions of activity durations have been limitedly used in the industry. After running the simulation thousands of times, the probability distribution of the total project duration can be developed to assist risk analysis, such as determining a schedule contingency for the project. In current practices, the use of the MCS method is still limited due to various reasons. One of the most challenging issues is the difficulties in estimating the probability distributions of activity durations objectively. This study aims to leverage historical digital daily work report (DWR) data available in the DOTs’ databases to determine the probability distributions of the production rates of work activities then estimate the distributions of activity durations when quantities of the activities are provided. Since the DWRs record the daily accomplished quantity of each work item in the construction phase, the actual production rates of the work items can be calculated to obtain a more accurate and realistic duration estimate for a future project. DWR data collected from a DOT were used to conduct a case study that demonstrates the value of this new approach for schedule risk analysis.