TRISTAM: Traveller Response and Information Service Technology: Analysis and Modelling

Prof. H.J.P. Timmermans (Eindhoven University of Technology), in collaboration with Delft University of Technology and VU University Amsterdam

'TRISTAM' stands for the research programme entitled ‘Traveller Response and Information Service Technology: Analysis and Modelling’. The programme researches how travellers use travel information such as travel time estimates when there are tailbacks. In this connection, the researchers are making full use of advancing ICT technology to avoid undesirable side effects of travel information, such as merely shifting congestion elsewhere.

Read an interview with Harry Timmermans (April 2009)

General aim

One would expect that the provision of information/guidance will reduce the uncertainty of travel, increasing travellers' Comfort and inducing them to change their behaviour such as to stimulate better use of the existing infrastructure and to reduce or even avoid congestion. One would expect that stimulating tele-activities will reduce physical travel and thus contribute to sustainable accessibility.
Existing research findings however suggest that this is not necessarily true. If too many travellers follow the recommendation to take another route to avoid congestion, the problem of congestion may simply shift from one route to another. If dynamic information is provided based on the wrong assumptions about traveller behaviour, it may turn out to be wrong, implying that next time, travellers may simply ignore it. Although teleworking will reduce the number of commuter trips, the time savings may be used to engage in other activities, with adverse effects on mobility rates.
Provision of information thus requires a good understanding of traveller response, considering changing technology, changing economics of energy use, and increased uncertainty about travel and congestion itself. Successful application of such technology to enhance sustainable accessibility also requires the development of operational behaviourally-sound methods and models.
This program will deliver innovative theoretical and empirical findings, methods and models for analyzing, predicting and assessing the impact of ICT on travel on dynamic network performance, accessibility, spatial externalities/land use and economic welfare. These will be applied to the next generation of intelligent ICT tools in case studies developed jointly with a user group.

Expected scientific results

Substantive

  • New insights about
    • willingness to pay for travel information
    • network, accessibility, spatial and economic effects of ICT use
    • dynamics of traveller learning/adaptation under information provision
    • Empirical test applications of theories of traveller response under information provision
  • New theory development about the pros/cons of public versus private information provision

Methodological

  • Innovative models of
    • strategic traveller decision making
    • nonlinear multi-level choice models (individual response to emerging aggregate patterns)
    • integrated measure of social benefits and willingness to pay
    • accessibility measure, integrating physical and virtual accessibility and the role of travel information
    • changes in activity-travel patterns
    • dynamic network effects incorporating traveller response and adaptation

Data

  • new experience with the design and implementation of so called serious games
  • innovative design and implementation of experiments in the tradition of induced value theory
  • new experiences with larger scale panel studies using 1CT and prompted recall technology

Tools

  • new decision support tools for the design of personal ICT-based transport policies and plans.
  • new and/or improved algorithms for dynamic network effects

Multidisciplinary approach

innovation/partial breakthrough in integrating network performance, accessibility, activity-patterns and advanced ICT applications

Expected results relevant for policy makers

  • Guidelines for service providers and authorities regarding
    • information provision strategies
    • information quality requirements
    • requirements for the data collection system
    • information needs and gains for different user-classes
  • Recommendations how to optimize information services to ensure
    • efficient use of the network under normal, but also during non-recurrent circumstances
    • sustainable accessibility
  • New and improved models to assess the effectiveness and impacts of ICT policies and technology

Plan of dissemination

We will conduct the following activities:

(i) targeted at other Master and PhD students

  • Mutual guest lectures and symposia at each participating university
  • Presentations at three PhD schools: Nethur, Trail and Tinbergen

(ii) targeted at international peers

  • Conference papers and presentations

(iii) targeted at policy makers and other potential users

  • Presentations at applied conferences
  • Publications in professional journals
  • Booklet discussing implications
  • Dedicated sessions/tutorials
  • Organisation of user group

Website, listing presentations, activities, results and papers
Show cases will be chosen jointly with user group: industry has expressed interest in real time personalized information services; personalized day planners, and personalized systems inducing less travel.

Project 1: Traveller Response and Information Service Technology: Analysis and Modelling of Network Effects

Prof. Dr. S. Hoogendoorn, TUD

Many theoretical studies have revealed the importance of providing accurate and consistent traffic information. In addition to the impact on traveller's confidence and compliance, inconsistent information may lead to an overreaction of the system, in particular when a large share of the travellers will respond (Hoogendoorn and Bovy, 1998). Taking into account traveller responses in generating information content is hence very important. Another important question is which traffic data need to be collected in order to enable provision of reliable information, and how these data should be used by the service providers.
Despite the lack of these insights, large investments are planned to establish a Dutch National Data Warehouse, which is aimed at enabling (commercial) service providers to provide traffic information and guidance covering all major urban, rural and motorway roads. Information determined from these data will have specific characteristics in terms of timeliness, accuracy and semantics. It remains to be seen what the effect of providing this information on network traffic operations will be. This project focuses on the system response to information provided via both personalized systems and via collective media, distinguishing descriptive and prescriptive information and time of information receipt. It focuses on the dynamic interplay between information quality, traveller's confidence, and information penetration level.

Project 2: Traveller Response and Information Service Technology: Analysis and Modelling of Accessibility Effects

Dr. Ir. C. Chorus, TUD

A wide range of accessibility-measures has been developed over the last decades, of which the utility-based (logsum) approach is most widely used in practice among transportation planners and spatial economists. In light of recent advances in technology in combination with recent insights into traveller behaviour, a revision of these measures of accessibility is called for. Specifically, the following two important issues are not yet adequately dealt with currently:

  • The role of ICT as possible substitute for physical mobility: it is well known that in some situations, ICT-use substitutes travel. However, current accessibility measures largely ignore the accessibility-implications of this empirical finding (e.g.: an office-location can at the same time be inaccessible by car, but easily accessible by means of ICT).
  • The role of awareness and learning from travel information: it is well known that travellers are not aware of every travel alternative available to them, nor of (changes in) the exact values of the alternatives' attributes. Instead, travellers are often found to gradually learn about relevant aspects of the transportation system over time. These notions are not incorporated in the current generation of accessibility measures, which typically postulate that relevant changes in the transport system lead to immediate changes in accessibility of locations.

Project 3: Traveller Response and Information Service Technology: Analysis and Modelling of Spatial Externalities

Prof. Dr. H.J.P. Timmermans, TUE

New forms of teleworking and the provision of travel information (both pre route and en-route) will likely influence the way individuals and households organize their activities in time and space. Travel information and location-based services allow travellers to make better informed decisions about the choices they have and reduce the uncertainty about various types of attributes influencing their decisions. Likewise, travel guidance systems will imply that travellers experience new and different options. All of this should be viewed in a dynamic context as travellers learn about their spatial and travel environment. The feasibility of activity-travel patterns depends largely on the spatial distributions of land use (and the time space constraints/accessibility involved) and on institutional factors such as office and shopping centre opening hours. Thus, due to increasing congestion, less reliable travel times and the impact of various forms of ICT, individuals and households are likely to change or are advised/stimulated to change one or more characteristics of their activity-travel patterns (destination, timing, transport mode, task allocation etc.). In turn this may impact the market potential of facilities, triggering supply side dynamics. Little is know about the intensity and nature of such changes. Moreover, current activity-based models of activity-travel patterns do not incorporate such dynamics yet (Arentze and Timmermans, 2007).

Project 4: Traveller Response and Information Service Technology: Analysis and Modelling of Economic Effects

Prof. Dr. E. Verhoef, VU

This project aims to provide insight into the economics of accessibility-enhancing ICT applications. We are specifically interested in two types of questions:

  1. Social benefits of and willingness to pay for ICT. These two concepts are of course closely related, but may diverge because of external effects. Specifically, an individual's willingness-to-pay for ICT does not reflect the benefits or costs that other travellers attach to that individual's use of ICT, for example through marginal changes in congestion levels on various links in a network. This project explicitly aims to address both measures, and will also explicitly consider the wedge between them.
  2. Public versus private provision of ICT. An important question is to what extent the public sector should interfere in the provision of accessibility-enhancing ICT. Should it play a leading role, because the main aim is to achieve the public goal of improved accessibility? Or should it play a minor role, because the public sector is inherently less efficient than private operators? Or is it ideal to have a compromise of publicly regulated supply of ICT by private firms? If so, which form should this regulation take, taking into account that various potential market failures co-exist, such as economies of scale and market power in the supply of ICT, congestion and other externalities in transport, and the club-good character of information?

Project 5: Traveller Response and Information Service Technology: Integration and Show Cases

Prof. Dr. H.J.P. Timmermans, TUE

Current research has focused primarily on specific effects of ITC on travel. Moreover, models have largely remained academic in the sense that they have been based on numerical data, or small scale empirical data collection. Field experiments are scarce, certainly in regard to rigorous academic research. The integration of scattered results and prototype models requires a substantial investment in effort, and requires the expertise of a postdoc. It also requires close collaboration with potential users in government and industry.