Duration estimators and productivity metrics for highway construction

Estimating durations of projects is an ongoing challenge for owners and contractors in the transportation construction industry. This challenge can be related to difficulties in measuring productivity. The current dissertation studied the influence that construction (bid) quantities and project characteristics have on the duration of transportation construction projects, and also developed a new metric to understand the historical trends of productivity. The author collected historical data of 1,500 transportation construction projects executed between 2004 and 2017. These data were used to conduct a quantitative analysis of how construction quantities and project characteristics can be used to predict project durations. Additionally, the data collected were used to develop a new metric to assess the overall productivity trends at a project-type level. Historically, there are three well know methods for duration estimation in transportation construction: Bar (Gantt) Charts, Critical Path Method, and Estimated Cost Method. The latter being a method that uses a relationship between the cost and the duration of a project. This dissertation used Multiple Linear Regression and Artificial Neural Networks to analyze the relationships that project quantities and characteristics have with the duration of projects. Findings from this dissertation suggest that the cost of a project is not important to estimate the duration of a project when analyzed alongside project quantities and characteristics contrary to what is suggested by the Estimated Cost Method. Additional findings from this dissertation suggest that the productivity of the overall transportation construction industry did not show statistically significant signs of change over the period analyzed when measured with the metric developed during this dissertation. Nonetheless, the productivity of construction of Bridge projects did show an increasing trend over the period described. The current dissertation has practical implications for the estimation of the duration of transportation construction projects, as well as theoretical contributions to the ongoing debate about the trend of productivity in the transportation construction industry.