1. Home
  2. TRID
  3. View Record
https://nap.nationalacademies.org/catalog/27432/critical-issues-in-transportation-for-2024-and-beyond

Wie Motorrad-Unfaelle vom Wetter abhaengen

How Motorcycle Collisions Depend on Weather

Die Studie untersucht den Zusammenhang von Witterung und Risiko von Motorradunfaellen. Die herkoemmliche Analyse der Unfallstatistik ergibt keine ausreichenden Informationen zu diesem Thema, da Wetterdaten in einem Polizeibericht nur zum Unfallort und -zeitpunkt, nicht aber fuer das ganze Jahr erfasst werden und somit die Exposition und nicht das Risiko messen. Fuer die Studie wurde eine Wetter-Datenbank des Instituts fuer Meteorologie und Geodynamik der Universitaet Wien verwendet, die Informationen ueber Niederschlaege an jedem Eckpunkt eines Rasters mit einer Zellengroesse von etwa 16 Quadratkilometern enthaelt. Diese Datenbank wurde mit der oesterreichischen Unfall-Datenbank auf einer georeferenzierten Grundlage fuer die Jahre 2002 bis 2004 verknuepft. Die Ergebnisse zeigen, dass der (neu berechnete) Parameter Regenwahrscheinlichkeit nicht nur in einer engen exponentiellen Korrelation mit Motorradunfaellen steht, sondern dass auch verschiedene Charakteristika von Unfalltypen und Unfallschwere in Zusammenhang mit der Regenwahrscheinlichkeit stehen. Aufgrund dieser Ergebnisse wurde die polizeiliche Unfallstatistik ueberprueft und der Einfluss des Wetters eliminiert. Die Prognose der Zahl von Motorradunfaellen in einem bestimmten Jahr kann demnach mit einer Fehlerquote von weniger als 3 Prozent erfolgen. Zur Gesamtaufnahme siehe ITRD D366371. (KfV/K) sense that motorcycle accidents depend heavily on the actual weather and the weather forecast. This impact may be derived from differences in the intrinsic risk of riding under different weather conditions, differences in the risk-taking behavior of motorbike riders under different weather conditions as well as the impact of weather on exposure. However, such relationships have not been researched thus far. Traditional analysis of accident statistics will never provide sufficient information on this issue, since weather information in a police report is just about the weather at the very time and location of the collision, but not about the remaining 364 days and 23 hours of a year. Hence, traditional analysis of police reports is more about measuring exposure instead of risk. In addition to this, accident statistics on powered two-wheelers are generally considered to be more informative about the weather in a particular year than about developments in motorcycle safety. For the study, a particular weather database prepared by the Institute for Meteorology and Geodynamics at the Vienna University was used. This database provides information about precipitation at each corner point of a grid with a cell size of about 16 kilometers square. This database was linked with the Austrian (police recorded) accident database on a geo-referenced basis, from the year 2002 to 2004, for which the relevant weather information was available. Cross-validation with police-recorded weather data was carried out successfully. As input for further calculation, a parameter called "rain likeability" was calculated for the whole of Austria. Motorcycle collisions and rain likeability were found to have a close exponential correlation. It was also found that there are different characteristics in terms of accident types and crash severity depending on rain likeability. These findings were used to review police-recorded accident statistics and eliminate the impact of weather. It was found that "predicting" the number of motorcycle accidents in a particular year is possible with an error of less than 3 percent based on the weather characteristics of this particular year alone and the relationship between weather and accidents. This will, in the future, move the analysis of motorcycle accidents much closer towards displaying a true picture of developments in motorcycle safety. This methodology can be applied to improve the analysis of any other activity that is dependent on weather conditions, where events take place at different sites. (A)

  • Availability:
  • Authors:
    • WINKELBAUER, M
    • BRANDSTAETTER, C
    • RIEGLER, S
    • Steinacker, R
    • TIEFGRABER, M
  • Publication Date: 2010

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01323662
  • Record Type: Publication
  • Source Agency: Kuratorium für Verkehrssicherheit (KfV)
  • ISBN: 978-3-642-15502-4
  • Files: ITRD
  • Created Date: Dec 22 2010 9:12AM