Advanced decision modeling for real time variable tolling : development and testing of a data collection platform.

PDFTech Report25-1121-0001-478DriversDecision makingAutomatic data collection systemsMathematical modelsReal time informationSmartphonesTravel behaviorVariable tollsUnited StatesMid-America Transportation CenterHanley, PaulSikka, NikhilYu, HaiMa, SuyunUniversity of Iowa. Public Policy CenterUnited States. Dept. of Transportation. Research and Innovative Technology AdministrationNTL-HIGHWAY/ROAD TRANSPORTATION-HIGHWAY/ROAD TRANSPORTATIONNTL-PLANNING AND POLICY-PLANNING AND POLICYUS Transportation CollectionOur current ability to forecast demand on tolled facilities has not kept pace with advances in decision sciences andtechnological innovation. The current forecasting methods suffer from lack of descriptive power of actual behavior becauseof the simplifications used in current economic decision models. These simplifications are in part due to the historicallimitations on data collection. Today, we are seeing advances in the data collection technology that captures naturalisticbehavior and this study is seeking to develop and test such technology. This will be the first phase of a naturalistic drivingstudy on the topic of variable road tolling and decision making.This investigation extends the state of knowledge of decision modeling under risk and ambiguity by developing amobile data collection platform for the capturing of naturalistic choice outcomes, associated environmental states, decisionmakers\ue2\u20ac\u2122 self-articulated perceptions of risk and assessments of ambiguity, and socio-economic attributes. Thissoftware/hardware development is the first step in advancing our ability to forecast future revenue sources intransportation. The data collection platform will be used in an extending study and will allow the study team to obtaininsights for developing, testing and implementing new behavioral models that explicitly describe how we use impreciseinformation\ue2\u20ac\u201din this case, ambiguous information signaled by variable toll lane charges\ue2\u20ac\u201dthat are superior to thoseobtained to date.To support the empirical observations of driver choice of diverting to HOT or non-HOT lanes in real-time, thisstudy documents the development and testing of an in-vehicle perception acquisition device. The device is based on smartphonetechnology and its software is designed using the latest IOS release. Drivers will be able to upload the softwareapplication onto their existing smartphone or tablet. The widespread acceptance and use of smartphones and tablets allowsthe collection and transmission of decision data collected in real-time without installing or altering the drivers\ue2\u20ac\u2122 vehicles inany manner. This both increases the convenience\ue2\u20ac\u201dand perhaps the use of the application\ue2\u20ac\u201dand reduces the cost of datacollection