Authors: U.S. General Accounting Office (GAO)

Date of Publication:  April 1992

Sponsoring Agency:  U.S. General Accounting Office (GAO)

Performing Organization:  U.S. General Accounting Office (GAO)

Report No: GAO/RCED-92-133

Abstract:

Despite rail safety improvements over the past 10 years, over 3,000 railroad accidents occurred in 1990 that were reported to the Federal Railroad Administration (FRA). In 1990, human factors caused the largest portion (36 percent) of such accidents, in contrast to the situation in past years, when track defects were the most prevalent cause. This report
addresses the railroad Hours of Service Act, as amended, which requires that employees (1) may work no more than 12 continuous hours without a minimum of 10 consecutive hours off duty and (2) be given at least 8 consecutive hours off duly in every 24 hour period.

We focused our work on answering three questions:

• Are railroads complying with the above requirements of the Hours of Service Act?
• Can safety be improved by amending the act to reduce the maximum number of hours (e.g., from 12 to 10) that an engineer is allowed to work?
• Do work schedule factors other than the maximum number of hours allowed by the act affect safety?

The findings related to characteristics of engineers’ work schedules are based on our analyses of 1990 work schedules for randomly selected engineers from four Class I railroads. These findings apply only to the four railroads we reviewed: Atchison, Topeka & Santa Fe; Kansas City Southern; Southern Pacific; and Burlington Northern. However, these four represent about 36 percent of the freight ton-miles carried by all U.S. railroads in 1990.

Although the act governs all members of a train crew, we limited our review to engineers because the accident data always identified the engineer on duty when an accident occurred but not necessarily other crew members. This limitation should not affect our findings concerning
compliance with the Hours of Service Act because both industry and FRA officials agree that engineers’ schedules reflect the same conditions found in schedules of other train crew members.
 

No. of Pages:  23