Evaluating roadway surface rating technologies.

PDFTech ReportRC- 1621UMTRI-2015-192013-0068Surface course (Pavements)RoughnessAccelerometersSmartphonesSoftwareData collectionFeasibility analysisEvaluationMichiganMichigan. Dept. of TransportationBelzowski, BruceEkstrom, AndrewUniversity of Michigan. Transportation Research InstituteUS Transportation CollectionThe key project objective was to assess and evaluate the feasibility and accuracy ofcustom software used in smartphones to measure road roughness from theaccelerometer data collected from smartphones and compare results with PASER(Pavement Surface and Evaluation Rating System) and IRI (International RoughnessIndex) measurement values collected from the same roadway segments. Thisproject is MDOT\u2019s first large implementation of a customized Android smartphoneto collect road roughness data using a methodology developed from previousresearch work performed by UMTRI. Accelerometer data collection wasperformed via Android-based smartphones using a customized software applicationcalled DataProbe. During the project\u2019s initial phase smartphones were installed ineach of nine Michigan Department of Transportation (MDOT) vehicles driven byMDOT employees. These same vehicles also were used during 2012 and 2013 to collect data on road distress using PASER Ratings for comparison.The DataProbe software application was used to collect data and transmit it to aUniversity of Michigan Transportation Research server, where it was sorted, stored,and analyzed. All MDOT regions are represented in this analysis that comparesroad roughness ratings for nearly 6000 one tenth of a mile road segments. For thesecond phase of the project, road distress (PASER Rating) data was collected in2014 simultaneously with an MDOT vehicle equipped with an IRI device and twoDataProbe smartphones and two UMTRI vehicles equipped with five DataProbesmartphones.The analysis of the 2012 and 2013 data found that there were a number ofsignificant predictors of IRI road roughness including: the phone and the vehicleused to collect the data, the speed of the vehicle collecting the data, the type of roadsurface, date of data collection, and accelerometer variance. By including quadraticterms to adjust for non-linear relationships and interactions among the predictorsstudied in this project, the multiple regression model predicted nearly 45 percent and43 percent of the variance in IRI values, respectively. An analysis of commonlyused IRI categories (3 level/5 level) using ordinal logistic regression found thatDataProbe accurately predicted these categories 68/71 percent of the time (2012data), 77/76 percent of the time (2013 data).Analysis of the data collected in 2014 showed multiple regression models withvariance among accelerometer measurements and speed accounting for 37 percentof the variance, while the ordinal logistic regression accurately predicted the IRI (3level/5 level) categories 86/83 percent of the time. These results are promisingwhen considering the near term application of the DataProbe technology for smallerlocales that drive over their local roads more often, generating web-based roadroughness visuals of each of the roads in their jurisdiction. In the longer term, statewideroad roughness measurement may be performed through the crowd-sourcingmodel available through Connected Vehicle initiatives, where all vehicles will beequipped with devices that support safety applications as well as other applicationssuch as those that measure road roughness