Benefits Assessment of Partial PTC Implementation over non-PTC Territories
The Rail Safety Improvement
Act of 2008 (RSIA) established four key functional requirements for Positive
Train Control (PTC) systems in the United States. As outlined in 49 CFR §
236.1005, these are to reliably and functionally prevent:
Overspeed derailments, including derailments
related to railroad civil engineering speed restrictions, slow orders, and
excessive speeds over switches and through turnouts
Incursions into established work zone limits
without first receiving appropriate authority and verification from the
dispatcher or roadway worker in charge
movement of a train through a main line switch in the improper position
In the United States,
railroads are equipping many miles of track, mostly high-density mainlines,
with full PTC protection. At the same time, railroads are equipping many
locomotives with on-board equipment, particularly Global Positioning System
(GPS) based PTC units, which if provided a track database for orientation,
could provide at least some level of safety protection anywhere the locomotives
go. But at present, this equipment is only activated on line segments where PTC
is fully installed. When asked about options for deploying PTC around the
world, Larry Jordan of Wi-Tronix concisely summarized the situation as follows,
it’s not legislated, I think it is a menu. I think there are items on that menu
that can be implemented economically in a very feasible way. When you say you
must take all four, then you might take none. [Emphasis added. As legislated by
Congress] there’s no menu items that address trespasser strikes or crossing
safety. If you had a menu and you were to add those two items to the menu, then
you might make a different choice than the four menu items that are required by
the PTC regulation. And you could do it in such a way that you could save more
lives. I think having a menu with a couple more choices gives you flexibility
and optimizes the economics.”
The proposed research would
develop an approach for better utilizing the capabilities of PTC onboard
equipment, which is already installed on many locomotives operated by freight
railroads and by Amtrak. Doing this would result in additional safety benefits
at very little additional cost. For example, speed restrictions could be
enforced in most instances even without the need for installing any wayside
equipment. It would clearly be better to have partial functionality enabled on non-equipped
lines than to have no supplemental safety protection at all, as is currently
In addition, by activating the on-board equipment, PTC
features beyond the four core functionalities could also be enabled. For
example, trains could be warned about dangerous conditions at highway/rail
grade crossings ahead, and even stopped if necessary. Until now, PTC-enabled
grade crossings could only be added on fully equipped PTC lines. However, by
marketing its X-ITCS as a stand-alone PTC solution, Alstom has already
demonstrated its ability to decouple PTC functionalities, particularly the
ability of grade crossing protection to operate independently of other PTC
The general principle is that on-board functionality
should be enabled for supporting the highest possible level of PTC functionality,
consistent with whatever wayside equipment has been installed along each rail
line. With “layers of functionality,” railroads could leverage the capabilities
of existing onboard PTC units for quickly and cost-effectively expanding the
scope of PTC coverage to as many additional rail lines as possible. A Partial
PTC implementation may be defined as follows:
_Partial PTC would be any PTC implementation that
satisfies some, but not all of 49 CFR § 236.1005’s requirements. It may however
add additional features that are not required by 49 CFR § 236.1005 including
but not limited to grade crossing functionality.
The focus of the proposed research
is not to impose any significant additional cost burdens on railroads, but to
maximize the leverage of investments that railroads have already made. The
1. Define a Concept of Operations (ConOps) for Partial PTC implementation including several variants of partial coverage. Since existing PTC systems have been integrated based on RSIA requirements, some software changes may be needed for decoupling of system functionalities. On a rail line that is not fully equipped, it is possible that PTC functionality may degrade, but it would not shut off completely. In concept:
For preventing overspeed derailments due to permanent speed restrictions,for those system which rely on GPS for orientation, it should only be necessary to download a GPS track map of the intended route which would include all existing permanent speed restrictions. For those systems which rely on transponders for orientation, it should only be necessary to install some transponders at periodic intervals. Once this is done, on many lines, assuming the that permanent speed restrictions could be downloaded at each start-of-trip, the on-board PTC unit would be fully capable of enforcing these limits without any further data communications.
The next enhancement in functionality would be the enforcement of track-specific permanent speed restrictions. For doing this, at a minimum, data communications is needed with the Central Office. On Traffic Control equipped lines the on-board PTC may obtain the positions of controlled switches from the Central Office to avoid the cost of adding WIU’s at the wayside. This could allow the on-board PTC unit to know the specific route to which the train has been aligned so it can apply the appropriate track-specific speed restriction.
Once data communications with the Central Office has been established, it may be possible to also add protection of temporary speed restrictions (TSR) and of work zone incursions. Enroute communications are needed because temporary speed restrictions or work zones could change while the train is en-route. Anytime a TSR changes this needs to be communicated to the trains. There are a number of different communication protocols for doing this which depend on which PTC system is being adapted.
In track warrant territory, train-to-train collisions can be prevented by activating PTC enforcement of the track warrant movement limits, even if other PTC functionalities are not enabled. This enforcement requires only communications with the central office, not with wayside devices. In unsignaled and ABS territory, track warrants themselves provide the primary means for preventing train-to-train collisions. In Traffic Control territory signal authorities may be converted to the equivalent track warrants.
2. Assess the Safety opportunity in terms of the number of PTC equipped locomotives that will likely be operating over non-PTC equipped rail lines, the risk exposure indices for certain types of accidents and likely reduction in accident rates associated with expanding PTC coverage.
3. Assess the Economic opportunity by developing a cost benefit assessment, keeping in mind that the cost of utilizing already-installed PTC on-board equipment and utilizing existing cellular data communications systems is likely to be very low.
Logical layers of PTC functionality will depend on
communications capability and the types of wayside equipment alongside each
rail line. These functionalities could be used either individually or in
combination with one another. Several critical issues need to be directly addressed:
Partial PTC solutions are likely to make much greater use of alternative (non PTC-220) communications systems, such as existing public cellular data networks. Railroads in both China and Europe are using GSM-R as the communications standard for their PTC systems. (GSM-R is a telecom standard based on cellular data.) Furthermore, existing onboard PTC units generally already can use cellular data, WiFi, or public band 2.4GHz frequencies as backup to PTC-220. Can the railroad industry support more diversity in its communications approach, particularly for supporting PTC implementation on lower density lines? This cost reduction is needed for making PTC implementation more affordable.
The ability to "Navigate" using GPS-based PTC currently needs supplemental data on switch positions, so trains can know which track they are on. This data has typically been provided by WIU’s at controlled switches. Some hand thrown switches have not had WIU’s, particularly in block signal territory where the statutory requirement for switch monitoring is fulfilled by the signal system. If a hand thrown switch were used in such territory, the PTC system might not be able to discriminate which track the train is on, and would need to be manually told of a track change. Possible solutions may include the industry adoption of more accurate GPS devices that are already being tested by the railroads. For controlled switches, Traffic Control systems already monitor the positions of both switches and track circuits and send this data to the Central Office using ATCS protocols. The Central Office could relay the needed data directly to trains which would avoid the cost of having to install WIU’s in the field.
The cost for adding switch monitoring to100% of industrial side tracks in dark territory is high, and so some have suggested that the decision whether to monitor any given switch should not be statutorily mandated, but be the result of a risk assessment. A train crew’s failure to restore a switch to the normal position, not vandalism was the cause of both the Graniteville, SC and more recent Cayce, SC accidents. As a result, NTSB has proposed a new operating procedure which would require the first train after realignment of a switch to approach the area at restricted speed, until the switch position can be seen, and to provide an independent confirmation that the switch has in fact been restored. Following the NTSB procedure might effectively mitigate most of the safety risks associated with hand-thrown switches while avoiding the costs of having to install PTC switch monitors on infrequently used switches in dark territory.
PTC Interoperability needs to be maintained in any partial PTC solution. This may be largely addressed by developing industry wide standards (ConOps) for defining partial PTC modes of operation. To the extent that some existing territories may already require special equipment (e.g. ATC cab signaling) there have always been limits to interoperability. However, there will always be special territories where it makes sense to deploy special equipment on board trains. Partial PTC neither resolves this problem nor does it make it any worse. However, in terms of ConOps that may be developed for Partial PTC they should be developed with an awareness of interoperability considerations, and encourage the evolution of PTC technologies to move in a direction towards technological convergence.
Regulatory Requirements for PTC are explicit since 49 USC 20157 defines PTC very precisely. While a “partial PTC system” may offer significant improvements in safety, the statutory definition of what it means to be PTC is inflexible. “Partial PTC” is not PTC ! Nonetheless under current law, it would appear that railroads do have the discretion to implement PTC technologies as they wish on line segments where PTC coverage is not already mandated by 49 CFR § 236. The proposed scope of this research is only to define possible Partial PTC modes of operation that could benefit safety; not to address the legal or regulatory matters. It is recommended that the railroads seek qualified legal advice on this issue.
The most important benefit of extending the geographic
reach of PTC is that it is a proven way to reduce rail accidents and save
lives. The benefits of expanding PTC capabilities will be further amplified as
PTC is enhanced to routinely encompass the area of highway-rail grade
As a result, additional safety and operational benefits
could be obtained from expenditures that railroads have already made. The
current situation is wasteful since existing onboard PTC equipment is not used
at all on non-PTC equipped territories. There is an opportunity to gain
additional safety and operating benefits at a very low cost.
For example, the proposed approaches could offer a
cost-effective way to start closing PTC coverage gaps that have been allowed by
Main Line Track Exemptions (MLTEs) including for example Amtrak’s Cardinal on the
Buckingham Branch, Downeaster on Pan Am Railways, Music City Star in Nashville,
etc. Although PTC exemptions reflect special the circumstances of the rail
lines for which they were issued, the fact remains that Partial PTC could offer
a low-cost way to get at least some PTC coverage activated quickly in these
territories. In the short term, whatever level of PTC coverage may be feasible
with existing equipment can be activated for these lines. Over time, PTC
functionality can be incrementally improved until the MLTEs are no longer
An important benefit of the proposed research will be to
develop innovative approaches to implementing PTC that will, over time, help
railroads bring PTC costs more in line with the economic benefits that a PTC
system can provide. Continued development is needed for both improving the
functionality and reducing the cost of PTC systems. A more flexible approach to
PTC implementation as proposed here would allow technology development to
become more market-driven and focusing on the creation of operating benefits to
railroads, while also significantly enhancing the safety of existing rail
There is a large body of knowledge of PTC operations,
safety and risk assessment, which can be drawn upon to develop new methods of PTC
operations, identify the risk exposure indices, and assess the potential risk
reduction and associated costs. Exposure indices may differ significantly based
on the type of train traffic, speed, frequency and mix: particularly if
passenger trains are involved.
Most PTC systems in the US today primarily rely on
dedicated data radio communications in the 220 MHz frequency range. However, most
systems also have backup cellular radio capability, which gives them the
ability to operate beyond the range of dedicated 220 MHz coverage. Wi-Fi can
also be used, mostly within terminal areas, to provide a faster means for
initializing route profiles and downloading large on-line databases at the
beginning of each trip.
Thus, even though PTC 220
with peer-to-peer wayside to onboard communications has become the de facto
implementation standard for I-ETMS, railroads do not necessarily need to incur
the high cost of extending PTC 220 data coverage beyond the currently equipped
lines -- since the onboard PTC units are capable of communicating using other
available networks. This suggests that Partial PTC may tend towards more of an
office-centric rather than field-centric architecture. This is because communications between trains
and the Central Office is a core
requirement of all PTC systems. Using the existing Central Office communications
link to relay the status of field devices to trains, avoids the for direct
peer-to-peer communications between trains and wayside devices. However, most Communications
Based Train Control (CBTC) systems have an office-centric architecture, so this
approach to train control is well understood in the rail industry.
1. While use of cellular data service may be an
alternative to installation of a PTC 220 data network, reliable cellular data
service may not be available everywhere. Therefore:
availability of cellular data coverage needs to be overlaid on a map of
candidate rail lines, to identify gaps in cellular coverage.
areas where cellular data coverage is lacking or deficient, consideration needs
to be given to the development of the most cost effective approach that can
support the telecommunications requirements of the system. These may possibly
- Asking cellular providers to extend their coverage to include the areas
- A railroad could implement its own dedicated communications network
- Tolerating the coverage gap, since PTC system are designed to be able to tolerate short communications outages without compromising safety.
2. For assessing the risk improvement from a Safety perspective, a base case is needed for comparative purposes. A methodology for defining this base case needs to be agreed and data collected to define the base condition. Presumably, such a base case would reflect PTC implementation on all lines where PTC is currently required, but no use of PTC on other lines.
3. Develop a Concept of Operations defining different layers and options for extending Partial PTC protection to currently non-equipped lines. For each implementation option:
Assess the cost of the option
- Estmate the risk reduction benefit as applied to a sampling of current lines, which should include all the current non-equipped Amtrak and commuter rail passenger routes, as well as a representative sampling of selected typical non-equipped freight-only lines
- Compare the costs and benefits of each option to develop a cost benefit ratio.
The completion of this research will provide both a solid technical understanding and economic evaluation of the ability to further expand PTC coverage on a cost effective basis, by making better use of the safety equipment that has already been installed in many locomotives but is not currently able to be used.
It is anticipated that this research will define new and
affordable options for leveraging investments that railroads have already made,
for extending safety improvements to additional lines that are not included in
the current PTC mandate. The ability to pursue partial PTC implementations may
also provide shorter term and lower cost solutions for addressing Amtrak’s
safety requirements on its lines for which the FRA had issued MLTE’s.
This research is anticipated to be completed within 12 months, with a final report detailing the ConOps, data collection, risk reduction and cost benefit results. The estimated funding requirement for the requested research is approximately $225,000.
KEYWORDS: Rail, safety, highway grade crossings, risk, positive train control
|Sponsoring Committee:||AR030, Railroad Operating Technologies
|Research Period:||6 - 12 months|
|RNS Developer:||Edwin , CRC AR030|
|Index Terms:||Railroad safety, Positive train control, Implementation, |
Operations and Traffic Management
Safety and Human Factors