Here's the link to the FCMB's presentation (PowerPoint file):
http://www.mbta.com/uploadedfiles/About ... %2019.pptx" onclick="window.open(this.href);return false;. Some confusing detail left unsaid in there.
They say that improving dwell times at the primary Park + DTX traffic clogs would improve performance by up to 20% with minor signal tweaks, allowing for +30 cars to be put in use. OK...that's pretty academic. We can see today that Red operates on better headways at the beginnings and ends of peak periods when crowding is lower vs. the height of peak when the platform dwells are taking their toll, inducing bunching, and actively harming achievable headways. The dwells chart on p. 17 shows pretty starkly how the extreme variability of peak crowding harms headways at the station dwells. And the line graph on p. 16 comparing signaling tech shows a performance convergence between current ATO and future CBTC at those dwells, grinding the entire line's performance down to the limiter imposed by those dwells (i.e.
any signal system is going to be FUBAR'ed Park-to-South Station by a negative feedback response because the peak-period crowding is just that overwhelmingly awful). This conclusion also makes sense.
What doesn't make any sense is how they expect to come up with 20%'s worth of dwell improvements when the crowding is physical: too many bodies smooshing down too-narrow stairs and too-narrow platforms at very old stations, and year-to-year ridership increases smooshing more bodies through those same egresses and platforms already at their limit. There's nothing small-scale or on the ops side that'll organize those crowds well enough to net a 20% improvement. Or a 20% improvement that won't immediately decay back to par...then further erosion...as ridership continues to increase and/or skew heavier to these stops. Taping those door hashmarks to the floor, more cameras, more staff assistance during peak, smarter door sensors, better customer service announcements...none of that adds up to the necessary streamlining. The improved braking on the CRRC cars doesn't have any impact either on dwells because the signal blocks are already clogged and running restricted speed before hitting the stations of max congestion. So either this is a prelude to a breathtaking amount of capital $$$ enlarging station egresses, or they're setting this up to disprove itself when the follow-up presentation on dwells can't add together anywhere close 20%'s worth of savings from small tweaks at Park and DTX.
What it looks like is that the unit price they got quoted by CRRC for replacing rather than rebuilding the 01800's was so spectacularly good they had to do due diligence to dig up any performance reasons to sell the FCMB on padding the order by up to 30 more. And if their bag of tricks started with customer service initiatives and resiliency-related track/signal work they were already considering...what's the minimum in extra tweaks they would have to roll all that up into a package pushed as service increase optics? They'll squint harder at the dwells and hold off that presentation to a later meeting because they don't feel confident enough releasing those numbers to the Board today. But in all likelihood the math is exactly as difficult as we fear and they won't come up with anywhere close to 20%'s worth of dwell improvements without major station capital $$$ an order of magnitude higher than the FCMB is willing to spend. If their claim is that dwells are the cause of service-decaying bunching and not vice versa or a little from Column A and a little from Column B, and If there can be no dwell improvement without invasive station mods...then they have a ready exit for considering the fleet padding or the bucket list of major lineside upgrades. Or pivot to the less-convincing (but potentially still viable) "any fleet padding is good over life-of-fleet" argument re: spares without service increases.
Personally, I wonder if +30 supplemental
Orange cars on top of the 152 already ordered would require less in the way of bags-o'-dwell-tricks to put to good use if those 30 extra units on the 01800 replacement order were switched into the Orange tincan size instead. But since the FCMB seems to be going out of their way to pour dirt on any thoughts of replacing the ATO signals with something better that's probably not going to look any more feasible under their assumptions for Orange either.