Travel Time Variability As a New Quality Indicator
VAN DER MEDE P and PALM H, Goudappel Coffeng and FLIKKEMA H, Ministry of Transport, The Netherlands
Traffic management alms to optimize the use &the existing road network capacity. Traffic management may lead to a reduction of congestion, thus increasing the level of service. Traffic management may also lead to increased reliability of the road network
Traffic management alms to optimize the use &the existing road network capacity. Traffic management may lead to a reduction of congestion, thus increasing the level of service. Traffic management may also lead to increased reliability of the road network in other ways. For instance, by providing improved (dynamic) traffic information, road users may be able to predict their travel times more accurately, thus experiencing travel as more reliable. The amount of congestion is a common indicator to describe the quality, or level of service of a road network. However, congestion is not the only quality indicator. If traffic management also aims to improve the reliability ofthe network in other ways, a new quality indicator to describe network reliability is needed.
Reliability can mean two things: predictability and invariability. These are not necessarily the same. When travel times vary strongly, this does not necessarily mean they are unpredictable. Although recurrent congestion implicates travel time variability, this variability can to a certain degree be predictable. At present, it is not clear whether separate indicators for predictability and variability are needed as quality indicators. Already a number of studies have addressed the measurement of travel time variability (Black & Towriss (1990), Knight (1974), McLeod, HounseU & Bain (1993), Montgomery & May (1987), Noland & Small (1995), Polus (1979), Senna (1991), Wakabayashi & Iida (1993)). However, no clear agreement exist on common indicators.
To arrive at a common indicator for travel time variability (TTV), which can be used both on a regional and national scale in the Netherlands, a pilot study was carried out by Goudappel Coffeng for the Traffic Research Center &the Dutch Ministry of Transport. In this study different aspects of travel time variability were assessed and considered as quality indicators. The aim of this study was to develop and test a methodology to determine TTV fi'om measurements of travel times from door to door. In the pilot study this methodology was tested for both work trips and road transport by trucks. A usable methodology should not only yield reliable measurements of TTV, but must also be cost-effective, since both national and regional road authorities should be able to carry out and make use of these measurements.
Association for European Transport