Introduction to the TRACE Project "in detail"

The TRACE project presentation leaflet is available for download here.

Top

Preamble

The European Union (E.U., 25 countries) is home to about 456 million inhabitants and about 270 million motorized vehicles. 2 million personal injury road accidents and 50 000 fatalities occur every year, which is now significantly higher than in the U.S.A. (42 000) which has a noticeably lower population, 290 million inhabitants and a smaller vehicle fleet (230 million motorized vehicles).

In most countries, economic losses due to road accidents represent 1 % or 2 % of GNP. In 1997, the European Transport Safety Council (ETSC) estimated the total cost of transport accidents in Europe at 166 billion Euros. 97 % of these costs were directly related to road transport.

Overall road safety has been improving in industrialized countries over the last 30 years, showing that political willingness and the application of countermeasures produce results. For example, the recent SUNFLOWER report concluded that, from 1980 to 2000, in 3 of the countries with the best road safety record, fatality trends had dramatically decreased, due to:

Top


Background

Despite these fatality reductions, Road Safety remains one of today's main societal concerns. Although car manufacturers in particular have gone to great efforts to improve their vehicles’ passive and active safety over the past 15 years, current road safety research has shown that an asymptote is about to be reached in most countries (even though France recently showed an unprecedented reduction in fatalities through stricter law enforcement policy including the installation of automatic speed cameras) and experts agree that preventive and active safety should now be brought to the fore.

That is why in 2001 the EUropean Council for Automotive Research (EUCAR) launched an initiative to develop a systemic approach to the problem of road safety: Integrated Safety, revisiting the Safety problem with a holistic System Approach. In 2005, 4 IP’s and one STREP (AIDE, PREVENT, EASIS, APROSYS, GST) were under way and have started to produce methodologies and results. However, only a few of these research projects required prior accident analysis as a starting point for further tasks (development of models, simulations, technologies, demonstrators, tests, etc.) aimed at a thorough understanding of real-world problems. This lack of knowledge is sometimes considered as a missing cornerstone.

At the same time, an eSafety Forum was established by the European Commission DG Information Society in 2001 as a joint platform involving all road safety stakeholders. The Forum adopted twenty-eight recommendations for the better use of Information and Communication Technologies (ICT) for improved road safety. Again, although former accident causation and impact assessment research produced a tremendous amount of knowledge, the exact nature of ICT's contribution to road safety could not be determined because consistent EU-wide accident causation analysis was unavailable. Consequently, the first of these recommendations sought to consolidate analyses from existing accident and risk exposure data sources in order to determine the most promising and/or effective counter measures. The second recommendation called for the establishment of a common format for recording accident data to allow the development of an EU-wide information system.

The EU is also funding the large-scale SafetyNet project (The European Road Safety Observatory) which is developing homogeneous accident data collection protocols in several EU countries and constituting an injury and fatal accident databank. But the project has just started and results will not be available in the short term. Moreover this project does not aim to identify relevant methodologies to evaluate the effectiveness and efficiency of technology-based safety systems. To try to overcome these problems in the short term, the Accident Analysis Working Group has examined available European data sources including CARE, MAIDS, GIDAS, EACS, CCIS, OTS, IRTAD, etc. These sources contain general accident statistics and/or in-depth details and range from European to national and private institution level. All sources contain useful data, so one of the first tasks has been to see how these heterogeneous sources could be better used to yield a more consistent European picture that would provide a usable safety diagnosis to enable impact assessment and thereby identify priorities for action. There are however significant access restrictions brought about by intellectual property right issues.

Since there is little likelihood of overcoming these restrictions, there is no prospect of making disaggregated data publicly available. To overcome these problems the working group undertook some qualitative analysis of the data sources, assessing the essential characteristics of the data and the potential for the different sources to be used in conjunction with each other. Four criteria were used:

The analysis confirmed the working groups' hypothesis that although many information sources already exist, their heterogeneity precludes the drawing up of a pan-European picture. Some data sources were never designed for coordinated analysis and therefore have little potential. Others focus mainly on passive safety, biomechanics or traumatology and do not give much insight into the causes of the accidents they contain. Others have considerable potential.

Based on this qualitative analysis of existing sources the working group recommended to the eSafety Forum that existing sources could nevertheless help to give a better understanding of accident causation and to evaluate (at least partially) the effectiveness of some on-board safety functions, if common analysis mechanisms are employed to query the different data sources and to share the results. This of course requires the formulation of a set of appropriate questions to establish the analytical focus point and which can be used in the interrogation. A multi-stakeholder workshop was organised to draw up this questionnaire, where participants shared and agreed on items in a list of questions. Since the list was long and resources to carry out the shared analysis were limited, the list was reordered by priority.

So far the work on this task has been done by a group of volunteers. It was assumed that the next stages of the task would require important personal and financial investment and could only be carried out when resources were made available. The necessary resources were expected at the end of 2004 and the work to be carried out over the next two years.

Top


Introduction to TRACE

This dual context (the Integrated Safety program and the eSafety initiative) stresses that the development of Intelligent Transport Systems in vehicles or on roads (and especially in the safety field) must be preceded and accompanied by a scientific accident analysis encompassing two main issues:


Top


Scientific Objectives

These two main orientations can be subdivided into several scientific objectives:

  1. The definition of Accident Causation is unclear. Many factors influence a country’s transportation safety level. These factors concern road safety policy, distribution and crashworthiness of the fleet, road network characteristics, human behaviour and attitudes, travel conditions, environment, etc. These issues have been studied for decades and considerable prevention efforts have been inferred from the analysis and comprehension of these factors. Nevertheless, further efforts are needed. These factors have to be studied together in order to provide a comprehensive and understandable definition of accident causation at the end of the project.

  2. The second objective is to provide the scientific community, stakeholders, suppliers, the vehicle industry and the other Integrated Safety program participants with a global overview of the road accident causation issues in Europe, and possibly overseas, based on the analysis of currently available databases which include accident, injury, insurance, medical and exposure data (including driver behavior in normal driving conditions). The aim is to identify, characterise and quantify the nature of risk factors, groups at risk, specific safety-related or risk-related societal issues, specific conflict driving situations and accident situations.

    3. This objective will be achieved at the end of the project with the public dissemination of most of the final reports.

  3. The third objective is to make this overview comprehensive, understandable and operational. Hence all aspects of safety will be taken into account in order to answer the following questions:

    4. When, where, how, why and to whom do accidents happen?
    When, where, how, why and to whom do injuries occur?

    This objective is to be monitored throughout the project. A system of four levels of quality control (Task Leader – WP leader – Project Coordinator – Wise guys committee) is assumed to provide the best and most understandable deliverables possible. In accordance with the objectives, the quality and comprehension aspects of all reports will be systematically checked at these four levels.

  4. The fourth objective is to improve the multidisciplinary methodologies that are considered necessary to achieve this knowledge and especially methodologies for analysing the influence of human factors as well as the statistical methodologies used in risk and evaluation analysis.

    5. This objective is achievable by the constitution of two specific Work Packages devoted to methodologies. The WP5-Human Factors- should provide the project with results at the end of the first year. The Statistical Methods WP is planned until the end of the second year. Both have a two-folded target:
    - provide the Operational WorkPackages with tools and instruments for accident causation analysis and the assessment of the safety benefits of technologies
    - identify and improve the scientific approaches in Human factors analysis and statistical analysis applied to accident causation and evaluation

  5. The fifth objective is to generate summary documents with popularised figures, statistics, results or other deliverables that can be used for the identification, validation and evaluation of expected or observed effectiveness of safety functions based on current or future vehicle-implemented technologies.

  6. The sixth objective is to supply participants of Integrated Projects and STREP’s, under the umbrella of the Integrated Safety Program, with accident causation inputs for the development of relevant technologies.

  7. The seventh objective is to establish links with the other road safety projects (especially SafetyNet, whose own objective is to establish a Road Safety Observatory in Europe through the construction of several accident and exposure databases at a pan-European level), and also research project(s) that will be dealing with the assessment of the potential impact and socioeconomic cost/benefit (up to 2020) of stand-alone and cooperative intelligent systems in Europe, according to the fourth IST Call.

    Joint workshops have already been scheduled with the E-Impact project (Month 1 and Month 11) and are foreseen with the SafetyNet project. Links have already been established with the Integrated Safety Program (Aprosys, Prevent, AIDES) in order to collect their potential need for accident analysis and evaluation. These links are assumed to be fundamental in ensuring co-ordination between accident data collection, accident analysis and studies on the evaluation of the effectiveness of safety technology-related measures.

Top


TRACE Work Packages

In accordance with these objectives, TRACE is divided into 4 series of Workpackages:

Top


Two important issues concerning the use of databases currently available and methodology improvements:

At this stage, two issues deserve to be highlighted:

A review of current knowledge and available methodologies will be systematically performed and reported by the WPs in order to start our own work with ‘state of the art’ knowledge.

Top