Project TitleIntegration of Innovative Sensing Technology and Data Analytics in Transportation Asset Management
UniversityVirginia Tech
Principal Investigator(s)Linbing Wang
PI Contact
Funding Source(s) and Amounts Provided (by each agency or organization)Virginia Tech Federal Amount $150,000
Virginia Tech Match Amount $150,000
Total Project Cost$300,000
Start and End Dates3/1/19—3/1/2023
Brief Description of Research ProjectTransportation asset management is an important tool to maintaining good physical status, quality service and trustworthy safety of infrastructure. Timely repair, rehabilitation, and preventive maintenance are critically important to maintaining such quality status and service, and prolonging service life at reduced costs. Infrastructure (bridges and pavements) degradation is a non--‐‑linear process and critical points where degradation rates change are important in decision making on rehabilitation. Nevertheless, it is challenging to predict the time of occurrence of such critical points. A number of modeling methods or models have been developed and yet the accuracy is very limited. Without reliable performance models, maintenance scheduling and life cycle cost assessments are not reliable and essentially just a mathematics manipulation. The limitations of such models are 1) models mainly based on mixture property testing in laboratories; 2) without considering the complicated loading and environmental conditions; 3) no integration of mix design lab test results and as constructed pavement properties; 4) having only relatively simple methods based on cause effect analysis of single or few causes. Recently a number of sensing techniques such as self-powered sensing, vibration sensor arrays and data analytics methods such as deep learning approaches have emerged and shown some promising features but not yet solidly developed to warrant implementation and realistic applications in asset management.