Published in IET Intelligent Transport Systems Received on 4th June 2008 Revised on 11th September 2008 doi: 10.1049/iet-its:20080034 Special Issue – selected papers from HCD 2008 ISSN 1751-956X Incorporating intelligent speed adaptation systems into microscopic traffic models I. Spyropoulou M.G. Karlaftis National Technical University of Athens, 5 Iroon Polytechniou Street, Zgrafou, 15573 Athens, Greece E-mail: iospyrop@central.ntua.gr Abstract: Intelligent speed adaptation (ISA) systems are incorporated here into microscopic traffic models; Gipps’ car-following model is discussed and the appropriate model parameters that need to be modified and additional ones that may need to be introduced are investigated. Driver behaviour under three different functionalities of ISA, namely informative, warning and intervening, is investigated through a driver simulator experiment. The impact of ISA systems on driver behaviour is a complex matter because it varies both among drivers and under different scenarios. The main parameters that capture the ‘reaction’ to the system are identified and are quantified through model parameters. These are driver speed, acceleration, deceleration, reaction time and effective size of the vehicle, and are estimated following the analysis of the simulator data. The resulting values confirm the necessity of parameter modification. The analysis performed for the incorporation of ISA into the traffic model indicated that a prerequisite of successful implementation is a deep understanding of the model parameters and dynamics. 1 Introduction Intelligent transport systems (ITS) constitute a rapidly evolving field that is anticipated to contribute to a sustainable road traffic network by improving road safety, traffic and environmental conditions, and mobility [1–3]. A prerequisite for their successful implementation is a determination of the anticipated impact that can be achieved through a variety of tools including on-road and simulator studies, laboratory experiments and traffic/driver model-based simulations; to this end, traffic models must be updated to include the operation of ITS. In the recent years, efforts have been made towards incorporating ITS parameters into traffic models. This allows for the possibility of enhancing model utility, for evaluating ITS, and assessing traffic management and road safety strategies. However, current applications mainly involve advanced traveller information systems (ATIS) rather than advanced driver assistance systems (ADAS), and primarily focus on the impact of information provision on driver route choice and road network conditions. Such examples include the traffic simulation programs AIMSUN (Advanced Interactive Microscopic Simulator for Urban and Non-urban networks) [4], CONtinuous TRaffic Assignment Model (CONTRAM) [5], PARAllel MICroscopic traffic Simulator (PARAMICS) [6] and Verkehr In Sta ¨dten UmlegungsModell (VISUM) [7]. Hidas [8] simulated the differences in lane- changing and gap acceptance behaviour at congested road sections when drivers are informed of traffic incidents downstream using the SITRAS (Simulation of Intelligent TRAnsport Systems) traffic simulation model (simulation of ITS) [9]. Intelligent speed adaptation (ISA) has also been incorporated in the DRACULA (Dynamic Route Assignment Combining User Learning and microsimulAtion) microscopic traffic simulation model [10, 11]; this application, however, concentrates on the system operation itself, as depicted through specific model parameters, rather than representing the impact of system operation on driver behaviour. Research has established that ISA affects driver behaviour, particularly parameters related to speed. In particular, a reduction of the following parameters has been identified: mean speed [12, 13], 85th percentile of speed [14, 15], time driven over the speed limit [16, 17] and distance driven over the speed limit [18]. Varhelyi and Makinen [19] even noted an increase in driving speed in certain cases. In this study, aspects of the impact of ISA systems IET Intell. Transp. Syst., 2008, Vol. 2, No. 4, pp. 331–339 331 doi: 10.1049/iet-its:20080034 & The Institution of Engineering and Technology 2008 www.ietdl.org