Monday, April 20, 2015

DOT Takes Additional Actions on Crude Oil Trains – Incident Report ICR

This is part of a detailed look at several actions that the Department of Transportation (DOT) took last Friday to reduce the hazards associated with the transport of crude oil and other flammable liquids by train. Earlier posts in this discussion include:

In this post I will look at the draft information collection request (ICR) revision that the FRA included in the set of Documents released by DOT. This is the draft of a 60-day ICR notice that will be published in the Federal Register; probably later this week.


This ICR notice is for a revision of the FRA Accident/Incident Reporting and Recordkeeping ICR (OMB # 2130-0500). This ICR was last renewed in July of last year without any significant change noted by the FRA. The previous ICR renewal in May of last year saw a decline in the number of responses projected based on the previous three years’ worth of accident reporting data. The hours per response remained the same, so there was also a proportional decrease in the estimated hourly burden.

Changes in Response Requirements

This ICR revision will require additional information be provided in each railroad accident involving any number of rail cars carrying crude oil or residual crude oil. The additional information will be a reporting of:

∙ The number of tank cars on the train that contained crude oil or residual crude oil;
∙ The number of such cars that were damaged in the accident; and
∙ The number of crude oil or residual crude oil cars that leaked during the accident and response.

This new requirement for providing this data (on form Form FRA F 6180.54 [.PDF Download] in Special Study Block 49b) will be in addition to any other data on the crude oil cars that would have normally been included in the incident report. FRA estimates that this will add 3 minutes to the reporting burden for this form for an estimated annual increase of 30 hour.


This reporting will allow FRA to track the number of crude oil rail cars involved in train accidents; their damage and release rates. This is data that is apparently not currently available to the FRA.

The FRA picked a very simple way to collect this information. There was no change made to the form involved and the data provided will pretty painless to submit. Unfortunately, as is the case in most instances where the easy way to accomplish a task is selected, the data provided will provide almost nothing in the way of information that can be used for analytical purposes.

For example lumping crude oil railcars and crude oil residue cars in the data collection will be expected to about double the number of cars involved in accidents and damaged in accidents since most railcars in crude oil service are not cleaned before being returned for refilling. The number of tank cars that leak will almost certainly not double because of the greatly reduced volume of crude oil available to make it to any holes in the tank car. Additionally, due to the lower mass of the tank car because of the missing crude oil, any damage to the car due to its motion or its resistance to change in motion when struck will be significantly reduced. This will make the tank cars seem less susceptible to damage than they really are in a full state.

Additionally, since there is no effort being made to determine what types of tank cars are actually in use and the rate of failure (measured by leaks) for each type of tank car, the FRA will not be able to adequately describe how the continuing change of the makeup of the crude oil tank car fleet will affect the failure rate of the fleet.

The FRA took the easy way out. They made a show of collecting additional data without significantly increasing the burden on the railroads and shippers making the reports. What they should have done was to design a new report specifically for rail accidents involving damaged and leaking rail cars containing crude oil. That form should have been designed to collect a meaningful data set that could inform additional regulatory actions or non-actions as the data dictated.

As it is, the FRA and PHMSA will still not be able to access meaningful accident information to be able to intelligently discuss the relative damage and failure rates and types of failures of the various types of railcars involved in the shipment of crude oil, or on how the variations in crude oil types affect those failure rates.

NOTE: I will submit a copy of this posting as a comment to the ICR notice when it is published.

No comments:

/* Use this with templates/template-twocol.html */