Causal Analysis of Passenger Train Accidents on Freight Rail Corridors

A number of economic, technical and political factors have limited the development of new, dedicated, very-high-speed rail systems in North America. Consequently, most, near-term development of improved or expanded passenger rail service in the U.S. will involve use of existing railroad infrastructure or rights of way. Comprehensive understanding of train accidents on shared-use corridors is critical for rational allocation of resources to reduce train accident risk. Nevertheless, little research has been undertaken to quantify the risk of a passenger train operating on or next to a freight train in a shared track or shared-use corridor setting. The research described in this paper presents the initial results of a study intended to understand what the most important contributors to the risk of train accidents on shared-used rail corridors are. This work can be used to better understand how to most efficiently and effectively manage the risk on shared-use rail corridors. Although developed in the context of the railroad operation in the U.S., this research can also be applied to other countries by adapting to their safety standards and risk management strategies. Introduction Demand for regional and intercity passenger transport in the United States is increasing, resulting in the need to expand transportation network capacity. For the past half century most of this demand has been met by U.S. highway and air transportation systems; however, these are becoming increasingly congested and adding capacity is more and more constrained. Furthermore, rising fuel costs have added further pressure to both of these modes because of their relatively high energy intensity. Railroads are being viewed as a promising alternative because of their ability to provide, safe, economical, comfortable and reliable passenger transport [1, 2]. A number of economic, technical and political factors have limited the development of new, dedicated, very-high-speed rail systems in North America. Consequently, most, proposed, near-term development of improved or expanded passenger rail service in the U.S. will involve use of existing railroad infrastructure or rights of way [3, 4, 5]. Shared or mixed use corridors refer to different types of passenger and/or freight trains using common infrastructure in some way or another. The U.S. Department of Transportation, Federal Railroad Administration (FRA) defines three types of shared use: shared track, shared right-of-way (ROW), and shared corridor [6] (Figure 1). Figure 1. FRA Definition of Shared-Use Corridor [7] Each type of shared-use system has associated advantages and disadvantages. Among the advantages are potentially lower capital costs, less environmental impact, and potentially easier access to urban cores. Among the potential disadvantages include: safety and risk concerns due to more frequent, higher speed operation of passenger trains in close proximity to freight trains and maintenance of way personnel, reduced line capacity due to more heterogeneous operating characteristics, longer travel time compared to very-high-speed rail, tradeoffs in infrastructure and vehicle designs due to differing characteristics of passenger and freight trains as well as other technical and institutional challenges [1, 7]. A high priority for any rail system is operating safety and there are several concerns associated with operating more frequent, higher-speed passenger trains on shared-use corridors [7]. Among these are, the consequences of a collision between a passenger train and derailed equipment from an adjacent track. Higher passenger train operating speed increases the likely severity of an accident if another train derails and fouls the track on which a passenger train is operating, or alternatively, if a passenger train derails and collides with a freight train on a nearby track [7]. The U.S. approach to this has been to develop robust crash-worthiness standards for passenger equipment that operate in mixed-used circumstances [8]. An alternative approach used in most other countries has been to invest heavily in prevention of such accidents. Both approaches are beneficial, but although the U.S. railroad accident rate has been declining for decades, reaching its lowest level ever in 2012 [9], the latter approach is more difficult in the North American heavy-haul, freight environment because of the 1.5 million railcars owned by hundreds of owners operating over hundreds of thousands of kilometers throughout the continent. Under a system such as this, optimized for highly efficient freight transport, it is difficult to maintain all these railcars in a sufficiently high condition such that components never fail and cause a derailment. Furthermore, the very high axle loads, commonly ranging from 29.8 to 35.7 metric tons put enormous stress on the infrastructure. Although the infrastructure is designed for these heavy loads, components occasionally develop problems that go undetected until they fail and cause an accident. Besides mechanical and infrastructure related causes, a third major cause of accidents are those due to human factors, in which an individual involved in operating or controlling a train's movement, makes an error. Although not necessarily related to heavy-axle-load freight, these accidents also contribute to the risk of shared corridor operations. Another important factor is grade (level) crossing accidents. Although nearly all of the world's dedicated HSR lines have complete grade separation between rail and highway lines, shared uses of existing freight infrastructure makes elimination of all grade crossings infeasible. Analysis of train accident causes is critical for rational allocation of resources to reduce accident occurrence and consequences. However, very little research has been undertaken to quantify the risk of a passenger train operating on or next to a freight line in shared-use corridors. The research described in this paper presents the initial results of a study intended to understand and quantify the most important contributors to the risk of train accidents on shared-used rail corridors. The ultimate objective of this work is to understand and quantify the most effective means of preventing accidents and reducing the risk associated with shared-used corridors. Mainline Passenger Train Accident Analysis Train accident data from the FRA Rail Equipment Accident database were analyzed to examine the effects of different accident causes on the risk of passenger train accidents. The FRA publishes annual train accident statistic summaries [10] but the results are presented at a highly aggregated level [11]. More in-depth insights can be found by analyzing these data in more detail and considering other statistical approaches. Figure 2a shows mainline passenger train accident rate over the 20-year interval from 1993 to 2012 sorted by five types of accidents: grade crossing, derailment, collision, obstruction, and miscellaneous. Annual traffic data were obtained from the U.S. Bureau of Transportation Statistics [12] and the FRA Railroad Safety Statistics Annual Reports [13, 14]. The overall passenger train accident rate has decreased since 1993. Over this period, grade-crossing accidents have been the most frequent type of passenger train accident, followed by obstructions and then derailments.