Modeling Behavioral Response to Real Time Traveler Information: An Application of a Continuous Time Integrated Transport Modeling Framework

In the transportation arena, technology today has afforded individuals with freely available up to date information about opportunities and network conditions. The prevalence of information has changed how people plan and execute their activity-travel agendas. In particular, the advent of Real-time Traveler Information Systems (RITS) that provide upto-the-minute system wide information about network conditions has altered how people navigate networks and pursue activities. The implementation of any RTIS solution requires a comprehensive modeling analysis for quantifying the impacts of such services. There are however very limited modeling tools that can accurately capture the impacts of such information services on individual activity-travel agendas without compromising on representation of underlying behaviors. The research proposed in this effort is aimed at exploring travel survey data to characterize dimensions of rescheduling behavior and identify schedule adjustment heuristics. Additionally, the observed behaviors will be implemented in a software prototype of a transportation system modeling tool and applied to the Hartford metropolitan region as a case study. Statement of Financial Interest There are no financial interests and this is purely a research exercise. All research products will be disseminated under open-source licensing agreements. Statement of Innovation The proposed effort will contribute to the state of research and practice in the transportation planning arena. First, the research will add to the literature on travel behavior by attempting to understand activity-travel rescheduling behavior in response to prevailing network conditions. In most implementations of travel demand models, rescheduling behaviors are implemented by imposing consistency constraints (in spatial and temporal representation) and adhering to budget constraints of available resources. For example, rescheduling dynamics are implemented by ensuring that time allocated to activities and trips add up to 1440 minutes in a day. However, such implementations do not consider true rescheduling behaviors and fail to capture the heuristics employed by individuals in making these decisions. In this research effort, travel survey data sets will be explored using data mining principles to understand pre-trip and en-route rescheduling decision heuristics and estimate models of decision processes involved. Second, the research will add to the state of practice on transportation modeling software by enhancing an existing open-source transportation model system to incorporate additional behaviors that accurately mimic rescheduling decision processes exhibited by individuals in real world. The prototype will comprise one of the few implementations of a transportation model system that can comprehensively model the full range of scheduling and rescheduling behaviors exhibited by individuals in response to real-time traveler information. Additionally, the enhanced prototype will be applied to a metropolitan region to demonstrate its applicability for planning and