The full study is available upon request. You can contact us by email at firstname.lastname@example.org to request a copy.
The objective of this study is to determine the factors that increase the severity of the outcomes for derailments involving dangerous goods, identify appropriate mitigating strategies for various train risk profiles and explore the possibility to amend the Rules Respecting Key Trains and Key Routes  accordingly.
The scope of this study involves conducting research on the factors contributing to train derailment outcomes. Factors include those that have already been identified in previous studies and other possible factors that may not be fully explained in existing research. The factors of interest may include, but will not be limited to, the train length, train speed, the track class, the accident cause, and the position of cars in trains.
The aim of the study was to determine how these factors affect the severity level of derailments considering that dangerous goods would be involved. The outcome of each derailment is unique and can be quantified by many factors, such as the number of cars that derailed, the number of dangerous goods involved, or the number of punctures to tank cars.
The literature identified and provided insight into the factors that contribute to derailment severity, including the effects of train speed, train type, derailment cause, and other factors on derailment outcome. The literature also suggested some potential mitigating strategies for these factors. For this study, factors that are generally recorded and tracked in accident reports in Canada and the US were used to categorize the severity of a derailment.
The project was split into three main tasks. For Task 1 NRC completed a literature review to summarize previous studies in North America that have researched the main factors contributing to train derailment outcomes as well as potential mitigating factors for various train risk profiles. The literature review focused on identifying how train speed and other factors affects derailment outcome for these risk profiles. The literature search was limited to North American freight railways only.
For Task 2, in the results of the literature review, NRC identified potential mitigating factors for the train risk profiles. The train risk profiles included dangerous goods (DG) trains and mixed goods trains compared to dangerous goods unit-trains, and how the outcomes of derailments may differ for these different risk profiles. Train risk profile was defined to be 5 levels of train profile, from a train with no dangerous goods (DG) cars, to a non-key train with 19 or fewer DG cars, a key train with 20 or more DG cars, a key train with one poisonous inhalation hazard (PIH) or toxic inhalation hazard (TIH) tank car, and a unit train consisting of all DG cars, such as crude oil tank cars.
For Task 3, NRC reviewed the Rules Respecting Key Trains and Key Routes and discussed how these rules manage risk and account for the effects of train speed, train type (dangerous goods vs manifest), and track conditions.
The study found that the literature provides evidence that there is a complex relationship between speed, train length, accident cause and other factors on the severity of a derailment. The number of derailed cars generally increases with an apparent linear relationship with accident speed, but the scatter in the data, where some high speed derailments derail few cars and low speed derailments derail many cars, suggests that speed alone is not the controlling factor.
The literature shows that derailments caused by broken rails or welds (i.e. unintended rail discontinuities) were seen to have a much higher derailment rate and derailed more cars per accident for a given speed compared to accidents caused by broken wheels, bearing failures or track geometry defects. Accidents caused by rail discontinuities result in a broader distribution of accident severity with a spread of occurrences with a multiple number of cars derailing compared to other causes such as broken wheels. As well, as speed increased, broken rail or weld caused derailments resulted in more severe accidents compared to other accident causes. For example, at 50 mph broken rail caused accidents would derail an average of twice as many cars as other derailment causes.
This relationship between speed and cause come into play with respect to the data seen for different track classes. Higher classes of track have fewer derailments, but the accidents that do occur are more severe because of the higher allowed operating speeds. Offsetting this is the fact that better track fails less frequently, and as such higher classes of track have fewer accident occurrences related to track issues. The overall result is that higher classes of track have lower overall risk in comparison to the lower classes of track, even though an individual accident occurrence on a higher class of track may be more severe (and perhaps more news worthy), but occur infrequently.
The literature has also clearly shown that loaded unit trains (including non-key unit trains) derail more cars and are involved in a larger percentage of broken rail or broken weld accidents compared to unit trains with all empty cars. The literature concluded that the risk levels of loaded unit trains are clearly higher than for empty unit trains. This conclusion can be extended to apply to unit DG trains, as there is no effective difference between a loaded unit train of non-DG products and one carrying dangerous goods.
Marshalling has also been studied as a method to reduce DG transport risk, with the prevailing opinion that the rear quarter to third of a train may be the safest location for placement of DG cars or blocks of cars. However, the increase in train handling in yards needed to achieve this goal for all shipments could result in an increase in yard accidents, as each car or car-block movement requires coupling and uncoupling of cars by yard workers, which increases their risk of injury, and the increased car movement increases the risk of the DG car derailing or becoming damaged during the yard movement. These scenarios point to the possibility that any benefits to marshalling DG cars to the rear portion of a train to potentially decrease accident severity could be offset by these and other unintended negative consequences during marshalling activities.
Other factors, such as seasonal conditions, cannot be controlled, and the mitigating factors available to offset increased risk due to these factors will be speed reductions (as currently practiced by railways in cold weather conditions), increased frequency of maintenance and inspections of track, cars and car related equipment such as brake components.
Improved tank car structure design has been shown to reduce the probability of DG release, reducing the potential severity of an accident. However, although improved designs reduce the probability of DG release, the risk of a tank being punctured and releasing exists in any derailment if the speed is sufficiently high. As well, improved tank designs do not reduce the likelihood of a derailment, or the number of cars derailing.
A review of the Key Train Rules has identified that several areas of the Rules can be improved to account for the repair and maintenance processes of railways in Canada. It was concluded that Paragraphs 5.3 and 5.4 of the Rules concerning joint bars should have a procedure in place for the temporary installation and inspection of joint bars and plug-rails in continuous welded rail (CWR) and that the procedure should include a frequency at which the temporary joint bar and/or plug-rail will be inspected until it is permanently repaired. As well, it is recommended that the inspection frequency should be related to traffic volumes, and the presence of Key Trains in the traffic.
It was identified that the Key Train Rules have limits on train speeds based on route location, wheel bearing faults, track class, and type of goods being transported, but that the rules do not have speed limits for DG unit trains. The rules also do not have recommendations for car placement of DG cars in mixed-goods trains. The Rules also do not have any requirements for more stringent operator experience requirements, or other human factors issues that may have an effect on the occurrence rate or severity of a derailment.
Finally, the factors affecting derailment severity were summarized, and mitigation strategies were suggested. The application of these strategies to the risk profiles identified by TSB in the Gladwick  report was presented as a set of example, or hypothetical, mitigation strategies. The example mitigation profiles used a combination of increased rail flaw and track geometry inspections and repairs, increased car and locomotive inspections and repairs, train speed reductions, and human factors improvements, such as requirements for increased training or work experience when operating Key Trains with large percentage of DG cars present.
These examples represent some, but not all, possible strategies that could be used to mitigate the increased risk associated with DG transportation. The full discussion of all possible mitigation strategies and how they could be formulated and implemented into the operations of the railways in Canada is beyond the scope of this report. As well, strategies that may ultimately be developed and put into practice will most likely require a more global looking cost-benefit analysis that also includes factors such as the economic impact of the strategies on the rail industry and the Canadian economy. This type of analysis was beyond the scope of this project and was not provided here.
However, the literature reviewed does support the risk mitigation profiles suggested, in that the increase in overall risk that occurs as the number of DG cars in a Key Train increases (from one TIH/PIH car in a train up to that of a unit train consisting of 100% DG cars) could be countered with an increasing level of track-related, equipment-related, and human-factors related requirements, each with the aim to reduce the likelihood of an accident occurrence, as well as reduce the severity of an accident should it occur. Although the complete elimination of all derailments from any cause may not be possible, the minimization of the likelihood and outcomes of derailments, without seriously impacting railway operations, should be the ultimate goal of the regulating bodies and railways alike.