FORECASTING PASSENGER TRAIN CATASTROPHES

In a previous paper (Ref. 1), data on the annual number of railroad passenger fatalities per passenger mile were examined. It was found that there has been a significant improvement in railroad passenger safety in the period examined, 1890-1977, (the bulk of the improvement occurring between 1890 and 1930). It was suggested that most of the safety for railroad passengers occurred mainly because of a reduction in the number of annual accidents, but that the survivability in any one accident while also having improved, had not kept pace with the safety ensuing as a result of a reduction in the number of passenger train accidents. In this paper we examine the issue of survivability directly by examining historical data on the maximum number of railroad passenger train fatalities in any one accident in a year for the period 1902-1976. It is found, as suggested in the previous paper, that there is only a relatively small improvement in survivability in this period. Over this time frame, any one year is almost as likely as any other to have a large number of passenger fatalities. That is, unlike the dramatic decrease in the number of annual passenger fatalities over the past 3/4 century, the maximum number of people killed in passenger train accidents in a year remained relatively constant in the 3/4 of a century of data.

  • Corporate Authors:

    Transportation Systems Center

    55 Broadway, Kendall Square
    Cambridge, MA  USA  02142
  • Authors:
    • Kahn, D
  • Publication Date: 1978-6-13

Media Info

  • Features: Appendices; Figures; References; Tables;
  • Pagination: 18 p.

Subject/Index Terms

Filing Info

  • Accession Number: 00194859
  • Record Type: Publication
  • Source Agency: Transportation Systems Center
  • Report/Paper Numbers: SS-20-U5-45
  • Files: TRIS
  • Created Date: Jul 11 1979 12:00AM