-From Ronald L. Batory, Administrator: This letter is the Federal Railroad Administration's (FRA) response to twelve National Transportation Safety Board (NTSB) Safety Recommendations (see list below). Over half of these recommendations are currently classified as "Open - Acceptable Response," and because the FRA has addressed the intent of the recommendations, no further action is necessary. FRA therefore requests that these be classified as "Closed- Acceptable Action."
For the remaining five, FRA has evaluated each recommendation relative to current and potential new regulations, including requirements for conducting cost-benefit analysis of each potential measure to address each recommendation, and has concluded FRA cannot reasonably take further action on them. Thus, FRA respectfully asks the NTSB to classify each of them as "Closed."
Overall, the twelve Safety Recommendations in question are: • R-01-02 • R-12-21 • R-13-22 • R-14-17 • R-01-17 • R-12-22 • R-13-38 • R-14-44 • R-08-06 • R-12-41 • R-14-16 • R-14-48
In the enclosure, FRA discusses the challenges to implement these recommendations, describes what actions the agency has performed, and explains why FRA cannot proceed further, other than to audit compliance as appropriate. Each recommendation is addressed in the enclosure in the following manner:
• NTSB Safety Recommendation Number;
• Text of the Safety Recommendation as issued by the NTSB;
• Status (e.g., "Open-Acceptable Response");
• FRA's position on the Safety Recommendation (see bolded text in shaded boxes);
• A summary of the accident that led the NTSB to issue the recommendation;
• A summary of the NTSB and FRA correspondence regarding each recommendation; and
• FRA's explanation for why we cannot pursue any further action on the recommendation.
To facilitate closure of these recommendations, FRA met with the NTSB on March 1, 2018, to expound on our reasoning and answer questions. This enclosure only includes those recommendations for which we believe we came to an understanding.
If FRA can provide further information or assistance, please contact Mr. Robert C. Lauby, Associate Administrator for Railroad Safety and Chief Safety Officer.
Please note that Federal agencies like FRA are required to follow the direction of Executive Orders 12866 and 13563 for rulemaking, which require quantifying both costs and benefits, reducing costs, harmonizing rules, and promoting flexibility. Executive Order 12866 specifically states that:
Each agency shall assess both the costs and the benefits of the intended regulation and, recognizing that some costs and benefits are difficult to quantify, propose or adopt a regulation only upon a reasoned determination that the benefits of the intended regulation justify its costs.
To meet the requirements of these Executive Orders, for each proposed and final regulation issued, FRA performs a regulatory analysis to:
1. Establish whether Federal regulation is necessary and justified to achieve a social goal; and
2. Clarify how to design regulations in the most efficient, least burdensome, and most cost effective manner.
While issuing regulations to implement many of these NTSB recommendations could improve railroad safety in the specific railroad accident or incident from which each arose, regulatory action to implement these recommendations would result in financial and safety costs that far exceed the societal benefits of improved safety or accident avoidance. Based on the associated cost-benefit analysis, implementing regulations that are required by some of these recommendations would not meet the intent of the Executive Orders listed above, which is inconsistent with the Administration's regulatory policy. Where applicable, FRA has calculated the anticipated costs and benefits of each recommendation and included that information with each detailed response.
Please also note, in the 2016 Federal Railroad Administration Report to Congress on Actions Taken to Implement Unmet Statutory Mandates and Address Open Recommendations by the National Transportation Safety Board and the Department of Transportation's Inspector General Regarding Railroad Safety, FRA informed Congress that the agency would be taking no further action on these twelve recommendations.
Current Status: Open-Unacceptable Response
FRA believes revising the regulations to meet the recommendation is not technologically or economically feasible, and that the loss of event recorder data is a rare event that does not justify such a burden to mitigate such an unusual event. FRA will take no further action on this recommendation, and respectfully requests that the NTSB close it.
The NTSB issued Safety Recommendation R-13-22 in response to an accident on June 24, 2012, near Goodwell, OK, in which two freight trains-one eastbound and one westbound-collided head-on. The collision derailed three locomotives and 24 cars of the eastbound train, and two locomotives and eight cars of the westbound train. The engineer and the conductor of the eastbound train and the engineer of the westbound train were killed. During the collision and derailment, several fuel tanks from the derailed locomotives ruptured, releasing diesel fuel that ignited and burned. The damage was estimated at $14.8 million.
In its January 2, 2014, correspondence to the NTSB, FRA stated why the recommendation is infeasible to implement, including the issue of railroads not being able to acquire the additional communications spectrum necessary to implement the recommendation. The NTSB stated in its February 24, 2014, reply:
We note that railroads may need to acquire additional communications spectrums to continuously download and transmit the content of locomotive-mounted event recorders to capture the real-time data recommended. We also note that railroads are competing for sufficient communications spectrums to implement PTC systems and other radio communications issues, that you see no additional benefits by taking the recommended action, and that you would like us to reconsider and close the recommendation. We believe that there may be more than one means of implementing this recommendation and urge you to explore all potential solutions. This may include strategic mounting locations (as far aft as practical) for event recorder memory modules that are certified as crashworthy, installation of more than one of these modules, data telemetry options to transmit data continuously from the lead locomotive to an alternate location on the train (such as the end-of-train device), or other redundant solutions that can increase the likelihood of data availability after an accident. Although we recognize that implementing Safety Recommendation R-13-22 will involve some expense, we continue to believe that the redundant storage of event data would provide critical data needed to prevent the recurrence of similar accidents, and that the associated benefits of such information outweigh the associated costs. We believe accidents that destroy event recorder data will continue to occur unless action is taken. Such data loss can hamper future investigations and the ability to accurately identify safety issues and potential corrective actions in a timely manner. Accordingly, we request that you reconsider your position. Pending a reply to this request, Safety Recommendation R-13-22 is classified OPEN-UNACCEPTABLE RESPONSE.
FRA's actions to address R-13-22: FRA has considered this recommendation and determined that implementation is currently neither technologically nor economically feasible, and because the loss of event recorder data is rare, does not add significant benefits to justify the cost. From 2005 through February 2018, FRA investigated 561 accidents and only two of those accidents resulted in the loss of event recorder information (0.35 percent of such accidents). One of those two accidents is the very accident near Goodwill, OK accident that precipitated Safety Recommendation R-13-22. Although event recorder data loss may occur under certain extreme circumstances, the current requirements in 49 CPR part 229, appendix D for a certified crashworthy event recorder memory module include that it survive:
• A temperature of 1,400° F for 1 hour, and 500° F for 10 hours;
• An impact shock of 55 g (100 ms pulse), and static crush of25,000 lbs for 5 minutes; and
• Fluid immersion for over 48 hours.
In the unlikely event that these limits are exceeded, the minimal risk of lost data on the lead locomotive is offset by data retention on the memory modules on trailing locomotives in the consist, which are recording most of the same vital data.
In addition, there are significant cost and technological barriers to addressing this recommendation. For those few accidents involving trains with a single locomotive whose event recorder memory module is destroyed by conditions exceeding the test conditions given above, or accidents in which event recorder memory modules on all locomotives in the consist are destroyed, remote recording of the event recorder data would need to be accomplished in real time, because the data in the last few seconds before the accident occurs is typically what is most needed in an accident investigation. As technology matures, the railroad industry is moving beyond event recorders and towards the transmission and storage of real-time information. This real-time transfer of information offers additional safety and business benefits, such as locomotive performance monitoring. However, live monitoring and storage of information still has barriers. Some railroads are trying satellite and cellular technology to exchange some train performance information, but it is expensive to have a continuous wireless connection, and reception of the transmissions can be irregular or experience periods where no coverage is available.
These test systems work on requests or triggering events to send or receive data. As an example, Amtrak currently is capable of reviewing a specific locomotive's event recorder information on request. In addition, if an engineer induces an emergency brake application, it will automatically e-mail exterior monitoring systems that a brake application has occurred, or other triggering events that the railroad has chosen to preset. Long Island Rail Road can request information on its commuter fleet event recorder and monitor in real-time certain parameters, including the speed of the train. BNSF is capturing event recorder data at set intervals with cellular or satellite transmissions. Other methods utilized include CSX Transportation's (CSX) implementation of an Event Recorder Automated Download (ERAD) system, and a similar system UP has to retrieve data as locomotives pass certain wayside readers.
To have a separate and distinct event recorder to be able to provide continuous communication, the ability to conduct such transmissions would likely require railroads to acquire additional communications spectrum availability. Railroads are already competing for sufficient communications spectrum to implement PTC systems, as documented in FRA's August 2012 report to Congress, Positive Train Control: Implementation Status, Issues, and Impacts (pages 15-18), as well as dealing with additional radio communications issues discussed in that report (pages 18-21). 1 Moreover, in territory where PTC has been fully implemented, sixty percent of main track territory operations will have constantly monitored trains where data is stored onboard the locomotive as well as redundantly sent back to the railroad office to store pertinent information. To meet the letter of this recommendation, railroads would also likely need to install additional wayside communication equipment outside of PTC-equipped areas to transmit real-time information to a remote location.
Due to the very limited number of accidents in which remotely recorded data would be needed, the difficulty and cost of implementing this recommendation, PTC redundancy, and the advancement of locomotive monitoring technology increasingly being implemented in locomotives, FRA sees no additional benefits to its overall rail safety program by acting further on this recommendation, and respectfully asks that the NTSB reclassify Safety Recommendation R-13-22 as, "Closed-Reconsidered."