$200 Million Caltrain train control project shows signs of project failure

At the February Caltrain board meeting, the staff report on CBOSS positive train control showed strong signs of a failing project.

The signs of a project at high risk presented at board at meeting included:

  • a slipped schedule through the end of 2016 (from an original delivery date in 2015 that was still hoped for in September)
  • the custom software deliverables required for the project are consistently late and problematic
  • the contractor has been under-delivering on a continual basis
  • the federal acceptance requirements to define a completed project are still not complete, and continue to shift
  • staff will return with a proposal to continue funding the project through the coming year
  • the federal deadline for positive train control, originally set for last year, has been extended to 2018 at the earliest so there is not an immediate legal requirement to deliver

(Positive Train Control systems are intended to automatically avoid collisions; a federal mandate was imposed after a deadly crash in 2008 between a Metrolink passenger train and Union Pacific Freight Train; but passenger rail systems had been lagging in implementation and congress granted an exception).

Part of the underlying problem may be organizational transition with the technology vendor. CBOSS is built on top of GE Signalling’s ITCS product.  GE Signalling was purchased last year by Alstom (a French company), one of the biggest players in the worldwide rail signalling market, with a strong market presence with an alternative technology, ERTMS.  Given the organizational change at the vendor, ITCS may no longer be a strategic priority (or there may be other post-acquisition staffing issues). It’s clear based on results that the teams working on the technology aren’t delivering.

So far, Caltrain has spent $230 million on the project.   Forging ahead with a troubled custom technology project, with underperforming developers and contractors, chasing changing federal requirements, may be a recipe for throwing good money after bad. (Attentive observers had been predicting trouble since the early days of the project).

Since teams on a lagging project are often tempted to believe that one more round of work will fix the problems, it may be time for Caltrain to bring in expert reviewers to evaluate the state of the project and assessing whether it is worth continuing in its current form.

The Caltrain-High Speed Rail Compatibility blog suggests that since the High Speed Rail project needs positive train control, and needs to operate on the Peninsula corridor along with other services including Freight and Amtrak, it might make sense to stop the current project and shift to the technology that HSR will need?


Caltrain staff assert that Caltrain needs to run trains much closer together than HSR, so it has different technical requirements. However, Clem Tiller provides this example of an ERTMS system running  trains 110 seconds apart at 200 km/h (125 mph), almost ten years ago.

Given the signs of a project at high risk, it seems time to get a second opinion on what to do next before spending more money on the current approach.