MATHEMATICAL RECONSTRUCTION OF ACCIDENTS-ANALYTICAL AND PHYSICAL RECONSTRUCTION OF TEN SELECTED HIGHWAY ACCIDENTS.

Ten actual highway accidents have been mathematically reconstructed in a process comprised of two parts. First, trajectories of the vehicles were reconstructed from physical evidence observed at the scene, using a computer simulation of a two-vehicle, planar collision. Next, the responses of the right front occupant of the subject vehicle were reconstructed from observed internal vehicle damage and passenger injuries, using the predicted vehicle crash history and a nonplanar mathematical model of the crash victim. The predicted vehicle trajectories are in general agreement with on-sit observations for nine of the ten cases. The rest positions of the vehicles in the remaining case were strongly influenced by a secondary collision. In seven of the ten cases, the predicted occupant responses likewise agree generally with the reported evidence. Discrepancies of occupant responses in the remaining three cases are considered attributable to active responses active responses exerted by the occupant in anticipation of the crash. The effect of occupant response is demonstrated through computer simulation. Predicted values of head injury indicators in current use (head severity index and HIC number) do not correlate significantly with observed injuries as measured by the ten-point injury scale. The predicted values of the head injury indicators do, however, fall largely within the broad envelope of corresponding windshield impact test results. Following this analytical effort, the ten accident cases were physically reconstructed on an impact sled, using a sled pulse corresponding to the predicted vehicle deceleration history and an anthropometric crash test dummy to represent the accident victim. The results of most of the sled tests agree generally with observations from the actual crash and the results from the analytical reconstructions of crash victim responses. Differences are traceable to a number of factors, including uncertainties in the vehicle crash history; uncertainties in occupant position, orientation and muscular response; differences in size between the actual and simulated occupants; and differences in material properties and geometry between the actual and simulated vehicle interiors. Measured values of the head severity index show no distinct correlation with the observed injuries measured on the ten-point scale. However, a correlation is shown to exist between observed injury and predicted vehicle speed change. /Author/

  • Corporate Authors:

    Calspan Corporation

    4455 Genesee Street
    Buffalo, NY  USA  14225
  • Authors:
    • Bartz, J A
    • MCHENRY, R R
    • Segal, D J
  • Publication Date: 1974-6

Media Info

  • Features: Figures; References; Tables;
  • Pagination: 283 p.

Subject/Index Terms

Filing Info

  • Accession Number: 00098936
  • Record Type: Publication
  • Source Agency: National Safety Council Safety Research Info Serv
  • Report/Paper Numbers: ZQ-5341-V-1 Intrm Rpt.
  • Contract Numbers: DOT-HS-053-3-658
  • Files: TRIS
  • Created Date: Oct 18 1975 12:00AM