PTC Is FRA pushing too hard? - the railroads continue work on the Positive Train Control program, with emphasis on conforming to the Federal Railway Administration standards, expeciallly in the area of safety

Railway Age, August, 1999 by Tom Sullivan

Contributing Editor Tom Sullivan is a principal at Transportation Systems Designs in Oakland, Calif. Links to additional information on this and related articles can be found on TSD's home page at www.tsd.org.

Though railroads and suppliers have made significant strides toward interoperable Positive Train Control, questions remain as to whether the FRA will allow the industry enough time to "do it right."

The good news is that in under two short years, the major U.S. railroads have made significant progress in working with the industry to develop its next generation train control technology known as Positive Train Control (PTC). The bad news is that it is unclear whether actual PTC deployment will be fast enough to satisfy the Federal Railway Administration--partly because its own timetable may have become infected by a need for political expediency.

Dr. Robert Gallamore, program manager for the North American PTC Joint Program, probably best articulated the railroads' perspective in his paper for the Railway Age/PTG De Leuw, Gather Third International Conference on Communications-Based Train Control in Washington, D.C.: "FRA is considering using its regulatory powers to mandate PTC in high risk corridors. But the railroads argue that without interoperable standards, cost-effective systems are impossible. Without cost-effective systems, wide-scale deployment is unlikely. And without wide-scale deployment, few safety benefits will be realized."

Both major U.S. PTC programs, the North American PTC Joint Program (also known as the IDOT program) and the CSX Transportation/Norfolk Southern PTC program, appear to be converging on common standards and technologies. Despite these positive indicators, however, significant issues surrounding safety regulation of PTC appear still to be open and recent transit experience with CBTC technology suggests it may take longer than expected before PTC technology is ready for prime time.

Tom Prendergast, president of the Long Island Rail Road, is more philosophical about PTC. A long-time proponent of CBTC technology, Prendergast says, "Like a good wine, it takes time." Prendergast is in a position to know. He was head of NYC Transit Subways when the largest system in the world decided to take the time to do it right. A key goal of NYCT's major new CBTC initiative is the same as the railroads: ensure interoperability among multiple vendors.

WHAT, EXACTLY, IS PTC?

Opinions and pilot programs vary, in part because technology is rapidly changing and all architecture elements of PTC have yet to fail in place. For these reasons there is a tendency today to describe PTC not in terms of how it works, but rather, what it does. FRA Administrator Jolene Molitoris, in her testimony before Congress on April 1,1998, stated that from the point of view of safety objectives, a PTC system needs to achieve the following core functions with a high degree of reliability and effectiveness:

* Prevent train-to-train collisions (positive train separation).

* Enforce speed restrictions, including civil engineering restrictions and temporary slow orders.

* Provide protection of roadway workers and their equipment operating under specific authorities.

From an actual design perspective, there is general consensus that PTC is a subset of CBTC technology. Thus, while CBTC may have more features and capabilities PTC uses the same overall architecture of two-way digital radios that communicate information between onboard and wayside computers to locate and control the position of trains.

One key PTC goal is to avoid the use of track circuits. Mike Pracht, vice president-Railroad Transportation Systems for Siemens in the U.S., explains, "With track circuit technology, you must design a signal system for the longest train with the worst-case braking. When you run shorter trains or trains with better braking, infrastructure capacity is wasted."

Is PTC FAILSAFE?

What many may not have noticed in the Administrator's Congressional testimony was what she didn't say. Look again: There is no mention that PTC is expected to be failsafe. Rather, it must "effectively and reliably" perform the safety functions identified.

Improving safety without a "failsafe" design may sound like an oxymoron to many old-school signal engineers. On the other hand, the traditional signal industry may have painted itself into a corner over the years by suggesting it can deliver systems that never fail unsafely.

PTC safety issues are handled by RSAC, the Railway Safety Advisory Committee. The Standards Task Force of the PTC Working Group is exploring whether it is feasible to require that new systems be qualified through a quantitative showing of the system's safety, such as the "mean time between hazardous events."

RSAC is charged with putting its arms around the safety of PTC. But while it is useful to quantify actual system risk with PTC, for many reasons this appears to be difficult. For example, what should and should not be included? Should humans making errors be included? What is the probability of errors in safety-critical software and what really is the mean time between hazards of a vital relay?