Feasibility Assessment of a Smartphone-Based Application to Estimate Road Roughness

Transportation agencies spend significant resources to collect pavement roughness data using profiler vans. A potential alternative to collect functionally equivalent data at a significantly lower cost and higher level of temporal resolution is to use existing accelerometers in smartphones as the set of sensors. In this research, a prototype smartphone application was developed to investigate the feasibility of such an approach. Acceleration data were collected using a prototype application running on Android tablets on two routes in Virginia. The analysis results show that the proposed smartphone application can generate consistent data sets from different data collecting runs. In addition, the average of the collected data sets is found to be highly correlated with the International Roughness Index data collected by the Virginia Department of Transportation using profiler vans. Also, a sample size analysis revealed that most pavement sections require fewer than 12 data collecting trips at a 50 Hz sampling rate while fewer than 16 trips are required for a rate of 10 Hz. Finally, a preliminary benefit assessment for Virginia showed that the proposed smartphone application approach allows for collection of comparable roughness data for more roadways, more frequently with significantly less cost.