Skip to Main Content
 

Global Search Box

 
 
 
 

ETD Abstract Container

Abstract Header

Evaluating the Accuracy of Pavement Deterioration Forecasts: Application to United States Air Force Airfields

Knost, Benjamin R

Abstract Details

2016, Master of Science, Ohio State University, Civil Engineering.
The US Department of Defense is responsible for the design, construction, operation and maintenance of a vast network of pavement infrastructure, including parking, roadways, bridges, and airfields. This study focuses on airfields maintained by the United States Air Force. With budgets shrinking in recent years, it has become increasingly important to maintain airfields at serviceable levels for the lowest possible cost. Air Force civil engineers use a Pavement Management System (PMS) software known as PAVER to support the maintenance decision-making process they undertake on a regular basis. Airfields are represented by a set of contiguous sections expected to exhibit homogeneous condition. The inputs to PAVER are field observations of pavement distress for each section and the outputs are the values of the Pavement Condition Index (PCI) for each section. The PCI values are then used to model and predict deterioration, determine maintenance and repair requirements, and estimate future budgets. It is important for the PCI forecasts to be accurate to ensure that decision makers are making effective infrastructure investment decisions. The objective of this thesis is to investigate the PCI prediction errors. Historical airfield PCI observations and forecasts at six Air Force installations across the United States are used to compare forecasted PCI values with observed PCI values. The errors in these historical forecasts are then used to develop a forecasting error model, which can be used to correct for systematic errors in the forecasts. Factors such as forecast horizon, pavement age, condition, climate, and location are considered in developing and estimating the error model using ordinary least squares. Alternative models specifications are tested. The estimated models are also evaluated in terms of their effectiveness in correcting for systematic forecasting errors. The results reveal that all the previously mentioned factors are statistically significant contributors to forecasting errors. To evaluate the developed relative error models, they are applied to correct the forecast errors of pavement sections in a test set (the records of the test set are not used in estimating the models). The results indicate that forecast error corrections are in general meaningful. The corrections lead to more improvements in the cases of longer forecast horizons, which is particularly encouraging. The implications of the results, directions for future research, and recommendations are discussed.
Rabi Mishalani, Dr. (Advisor)
89 p.

Recommended Citations

Citations

  • Knost, B. R. (2016). Evaluating the Accuracy of Pavement Deterioration Forecasts: Application to United States Air Force Airfields [Master's thesis, Ohio State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=osu1480665140928498

    APA Style (7th edition)

  • Knost, Benjamin. Evaluating the Accuracy of Pavement Deterioration Forecasts: Application to United States Air Force Airfields. 2016. Ohio State University, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=osu1480665140928498.

    MLA Style (8th edition)

  • Knost, Benjamin. "Evaluating the Accuracy of Pavement Deterioration Forecasts: Application to United States Air Force Airfields." Master's thesis, Ohio State University, 2016. http://rave.ohiolink.edu/etdc/view?acc_num=osu1480665140928498

    Chicago Manual of Style (17th edition)