Minutes prepared by: Dr. Alan F. Rumsey

John Lewis (MBTA) welcomed the participants to this seventh meeting of the CBTC Peer Group at the MBTA Operations Control Center.

The objective of the CBTC Peer Group meetings is to provide a forum for transit agencies to share past experiences, and discuss future plans, with respect to implementation of Communications-Based Train Control systems. The peer group meetings also provide a means for transit agencies to explore common operational needs, and operational differences, which may influence CBTC standardization initiatives.

The group has committed to actively support the efforts of the IEEE Rail Transit Vehicle Interface Standards Committee, Working Group 2 in developing a CBTC performance/functional requirements standard, and to promote the use of such a standard in future CBTC procurements.

Minutes of CBTC Peer Group Meeting #6 held at MUNI on July 16, 1998 were reviewed and accepted as written.

The Peer Group confirmed that minutes of the CBTC Peer Group meetings could be posted on the web site maintained by Tom Sullivan (www.tsd.org), provided the minutes had been reviewed prior to posting.

Geoff Hubbs (NYCT) provided an update on the status of the NYCT Canarsie Line CBTC project.

The MTA Board, at its July 29 meeting, approved selection of the following three proposers to participate in Phase I (the demonstration test phase) of the Canarsie Line CBTC project:

• Alcatel
• Alstom Signaling Inc., and
• MATRA Transport/US&S

NYCT has awarded contracts for $1.1 million to each of the above proposers and a Phase I kick-off meeting was held on August 5, 1998.

A number of joint working groups have been established or are planned, including working groups addressing wayside and carborne equipment installations, a Phase I test planning working group, and a working group to finalize interfaces with NYCT’s new R143 cars. (This last working group is awaiting award of the R143 procurement contract later this year).

Phase I equipment installation designs are to be submitted by the proposers by mid-October, with equipment delivery and installation on the NYCT test track and test vehicles commencing in early November. Formal demonstration tests are scheduled to commence in January 1999 and will run for approximately six months.

Phase I is a continuation of the competitive procurement process, and as such detailed technical, commercial and price negotiations will continue with the three Phase I proposers in parallel with the demonstration tests. Selection of the "lead" contractor, and award of the Phase II contract to re-signal the Canarsie Line is still scheduled for October, 1999.

The primary objective of the demonstration test program is to demonstrate/evaluate the proposer’s RF data communications system, train position/speed measurement system, automatic train protection functions, failure management functions, other operational features, and maintenance characteristics. Proposers will also be given an opportunity to demonstrate other capabilities of their proposed system.

Geoff Hubbs tabled the Integrated Test Plan which summarizes the Phase I demonstration test segments.

John LaForce (SEPTA) summarized the status of SEPTA’s plans to install a CBTC system in its Light Rail tunnel.

The proposed system is to be installed on 115 4-axle cars operating on 5 miles of track within the Light Rail tunnel, with 8 stations.

The CBTC system needs to support 60-70 cars per hour in peak periods, with 30+ cars normally operating in the tunnel during the peak. In the event of a delay, however, the number of cars in the tunnel could be much greater than 30 cars and hence the CBTC system needs to be able to handle bunching of cars. In addition, as many as 6 cars can be berthed in the terminal station. The bunching of trains and requirement for multiple berthing are seen as the two primary challenges in the project, and have led SEPTA to select a "leaky feeder" cable solution for its RF train-to-wayside communications link.

The proposed system is a "no frills" installation, providing safe train separation assurance and continuous overspeed protection only, with no ATO and no ATS functionality, except for a simple mimic display at the Control Center. There will be no track circuits, except at the one interlocking on the line. A restricted speed of 20 mph will be enforced in the event of ATC failure.

SEPTA have followed a two-step procurement for the CBTC project, using a Performance Specification developed from the draft IEEE WG2 standard. Adtranz were the low bidder. Funding is currently being finalized prior to contract award.

A 30 month implementation schedule is envisaged, with car equipment installation being on the critical path. No more than 2 cars per week can be taken out of service.

SEPTA’s Broad Street subway ATC program is being shelved, with the long range plan now to install CBTC on the subway at the end of the current car useful life (in approximately 15 years). SEPTA anticipate using standards developed by NYCT and the IEEE as the basis for this future CBTC procurement.

MUNI’s Patty DeVlieg summarized the status of the MUNI advanced train control system (ATCS) project.

MUNI introduced the SELTRAC ATCS into service in the Market Street subway on Saturday, August 22, with the first peak hour service on Monday, August 24.

MUNI has had three primary objectives for its ATCS system: 1) obtaining optimum throughput using service-proven technology; 2) improving safety by eliminating human error & equipment/system failures; and 3) increasing reliability & availability and lowering maintenance costs.

Prior to the introduction of the new train control system, throughput in the off-peak has been 26 trains per hour with throughput of 23 trains per hour in the peak (the reduction in throughput in the peak being a result of the need to couple vehicles into longer trains at the portal).

The new ATCS was introduced into service in parallel with a number of other operational changes, specifically:

On August 24, approximately 50% of the fleet operating in the Market Street subway were "non-communicating", either because the vehicles were not equipped or, if equipped, had failed to be detected and correctly initialized by the ATCS system at the subway portals (through a combination of equipment problems and human errors).

This high proportion of non-communicating trains created havoc to MUNI’s operations, and a deterioration in train throughput, since a communicating train following a non-communicating train must wait for the occupied axle counter block ahead to clear, before the system will allow the train to proceed.

When there are few non-equipped/non-communicating trains operating in the subway, the train control system appears to be working well, and hence the most significant action taken by MUNI to address the initial operating problems (after the intervention of the Mayor), was to limit the number of non-communicating trains operating in the subway. With this, and other actions taken by MUNI, throughput has steadily improved. However, the associated reduction in vehicle availability, due to having removed the non-retrofit cars from service, as well as removing a number of cars with ATCS-related equipment problems, significantly impacted passenger-carrying capacity. Currently MUNI are operating 1-car and 2-car trains in the subway, whereas before August 22 the typical peak train length was 3-cars.

One of the lessons to be learned based on the MUNI experience is not to underestimate the complications of mixed-mode operations.

Other factors also contributed to the frustrations experienced by MUNI’s passengers. For example, many MUNI outbound passengers in the afternoon peak, had been used to taking a train inbound to the Embarcadero terminal station, and remaining on the train in order to secure a seat for the outbound trip. With the introduction of the MUNI Metro Turnback (with trains now turning back under ATCS control behind the station, in a non-revenue area) this practice was no longer allowed, and all passengers were now required to disembark at Embarcadero. The result was hundreds of angry passengers forced out of their seats, to stand in line on an already crowded platform. In hindsight MUNI failed to appreciate the impact this change in practice would have on commute riders, and inadequate notification to the public was made on this point.

Another lesson learned from MUNI’s experience therefore relates to evaluating in advance the impact on passenger riding patterns, and implementing the appropriate outreach to communicating changes in service patterns.

ATCS equipment failures, surface power outages, a bank robbery in the vicinity (which resulted in the police shutting down the subway for a period), as well as incidents of passengers in the subway, also contributed to poor operating performance during the initial start-up period for the new train control system.

The initial start-up problems also brought to light a number of operational and management issues, as MUNI staff struggled to identify the source of problems and implement effective work-arounds. The importance of operational training, and in particular the simulation and management of abnormal and failure scenarios, is another important lesson to be learned from the MUNI experience.

Bob Miller (BART) summarized the current status of the BART AATC implementation project and provided the meeting participants with a copy of the Harmon test report "AATC Propagation and Communications Testing" for radio tests conducted last year in the New York City Transit subway.

Harmon has the license for the Raytheon spread spectrum radios and is currently developing a low cost version with a target to have production radios available by the end of 1998.

The AATC system for Phase II (a safety certified revenue demonstration between Lake Merritt and Fruitvale) is currently in the final design stage. Lab integration is scheduled for early 1999 with an operating system on the BART Hayward test track by late spring, 1999.

The station controllers in the AATC system architecture are based on commercial off-the-shelf hardware (industrial PC’s), with a "numerical assurance" safety concept. In addition to supporting the AATC wayside-centric control philosophy, the system hardware is also capable of supporting a vehicle-centric control philosophy.

BART/Harmon are working closely with the California Public Utilities Commission with respect to safety verification activities.

As part of the AATC program, BART are also planning to change out existing relay-base interlockings to solid state, and in the long term plan to eliminate track circuits, once an acceptable alternative has been identified for broken rail detection.

Vic Grappone (LIRR) summarized the status of LIRR's CBTC projects.

A key issue for LIRR in implementing CBTC system-wide is the need for interoperability between wayside and carborne equipment procured under separate contracts. As a consequence, LIRR is currently reviewing the scope of its proposed CBTC pilot project, with consideration of NYCT's Canarsie Line project which has similar interoperability objectives. The extent to which the LIRR pilot will be safety certified, whether it will be designed to go into service, and whether it will control grade crossings, for example, remains to be finalized.

LIRR's grade crossing CBTC pilot project at New Hyde Park, with a tie in to an Intelligent Transportation System (ITS) to provide preemption of traffic signals, is continuing, however a formal demonstration of the system has been postponed.

Fred Childs (PATH) advised that consultant selection for their ATC/New Vehicle projects was currently on hold (for internal reasons), but it was still hoped to award a contract before the end of the year. The consultant will assist PATH in the procurement of around 12 vehicles to test newer technologies on the line, such as ac propulsion. The consultant will also undertake initial operations and cost/benefit analysis and recommendations of various ATC system technology alternatives.

The JFK airport access project has been awarded to Bombardier. This system will be fully automated, and will utilizing the Alcatel SELTRAC inductive loop-based CBTC technology.

John Vogler (NJT) described the status of a transponder-based intermittent speed control system being implemented by NJT to provide civil and temporary speed enforcement, as well as automatic stop signal enforcement. Known on NJT as Positive Train Stop (PTS), this system supplements the existing automatic wayside signals on commuter rail lines that are not cab signaled. Plans are to install it over the entire commuter rail system, and in addition it will be functionally compatible with the ACSES system on Amtrak's Northeast Corridor. The NJT PTS system integrates geographic information to support speed/distance profile supervision.

The PTS contractor, US&S, has performed proof of concept demonstration tests using one trainset operating between Boonton and Denville. The tests generally confirmed that the required functionality was achievable, although a number of technical issues were identified. Extensive lab simulation tests are now planned prior to the next field tests in November.

John Lewis (MBTA) noted that CBTC was a consideration for the Blue Line, together with the procurement of new vehicles.

For the information of the Peer Group, John Lewis also made reference to a presentation that had been made to the MBTA by Mobile Electrification Systems, Corp (MESC) who together with MagneMotion (a spin-off of MIT) had developed an "anti-collision, absolute positioning, and communications system". This system was described as fail safe, with features that included "full-duplex communications, non-RF and 100% non-contact". The system was based on a vehicle mounted transducer and a wayside cable with "helical windings". No other technical details were available.

Alan Rumsey, chair of the IEEE Rail Transit Vehicle Interface Standards Committee, Working Group #2 (WG2), provided a summary of WG2 status, and highlighted the key changes that had been introduced in the latest draft of the WG2 standard since the last WG2 meeting.

Alan Rumsey advised that is was still the group’s target to be in a position to ballot the CBTC performance and functional requirements standard early in 1999, and as such it was important for the group to reach closure on the various open issues.

In preparation for the IEEE WG2 meeting scheduled for September 30, 1998, the Peer Group representatives discussed various issues related to system operations, failure management, system safety and availability criteria and environmental requirements.

The CBTC Peer group was thanked for its active support to WG2. As the purpose of the proposed standard is to establishes a set of performance and functional requirements for CBTC, user inputs are essential to ensure all operational needs are addressed.

Venkat Pindiprolu (FTA) noted that there was a high level of interest in the FTA to promote CBTC in agencies that are currently not actively considering this technology, and expressed the hope that this would be a natural evolution of the work of the current CBTC Peer Group.

The Peer Group representatives endorsed the value of the meetings and indicated a desire to continue to meet, approximately 3 times a year, to share experiences.

Although the CBTC Peer Group meetings have to date been scheduled to coincide with IEEE RTVISC Working Group #2 meetings, the Peer Group indicated a commitment to continue to meet, even when the current work of WG2 has been completed.

Lou Sanders (APTA) noted that their was significant activity underway in the area of Intelligent Transportation Systems (ITS) and that a Transit Standards Consortium had been formed to act as a clearing house for various standards activities. This group would cut across traditional institutional borders and was supported by both the FTA and the DOT. The first meeting of the group will be held on October 4, at the APTA Annual Meeting in New York City.

The next meeting of the CBTC Peer Group will be held in January or February 1999. Date and location to be advised.

Maureen Shirkus (MBTA) provide the group a very informative tour of MBTA’s new control center.

In concluding this meeting, John Lewis and MBTA were again thanked for their hospitality in hosting the meeting.