I think it's underrated the extent to which new lines like the Eglinton Crosstown, Ontario line, and frequent GO service will make shutdowns more palatable — because they will create fully-connected alternate routes for much of the subway network (which down the line can sub in for these new lines when they need their own major overhauls) — but it's also something we need to be thinking about before we shut down huge sections of the network, and only have hourly GO service and no subway alternatives. 

You can imagine, for example, shutting down the Spadina line section in the median of Allen Road and suggesting people take the Barrie line to the Eglinton Crosstown instead. But that isn't a reasonable thing to do until the Barrie line is running every 15 minutes, which is clearly still a long way off.

There is also the constant concern of maintenance efficiency. I wrote an entire blog post on the topic, but suffice to say, one does have to be careful about always answering maintenance problems with more shutdowns, especially given more shutdowns historically and in the recent past seems to be pretty weakly correlated with subway speed performance. There's always the risk that extra time to do maintenance leads to more time doing the same maintenance tasks instead of getting more done, especially since while subway maintenance is absolutely doable (most subways in the world do not seemed to be plagued by Toronto levels of slow zones); it is also high pressure given the limited time available. 

As has been the case since the TTC has been releasing maps of the slow zones which currently apply on the network, slow zones seem particularly concentrated in areas of the network on ballasted track. This includes the section of Line 1 running down the middle of Allen Road, which has gone from a regular demonstration of public transit’s power as subway trains carrying hundreds fly by cars, to one of its mismanagement, as trains filled with hundreds trundle along passed by cars at modest speed. Another place where slow zones have been near constant is in the tight curves on Line 1 as it turns from north-south to east-west, and back to north-south at Union and at St George. 

In both cases, it's helpful to remind ourselves that a city like London has older lines with similarly tight curves, and an order of magnitude more track on ballast. In the case of tight curves, that might mean we have something to learn about maintenance, and in the case of ballasted track, it might leave us in a tight spot. You see, Toronto’s subway system has several ballasted track sections, but perhaps not enough to justify getting good at maintaining this type of trackway (unlike London — they have well over 100 route kilometres on ballast), perhaps then the solution for this particular problem is replacing what little ballasted track we have with more expensive but lower maintenance slab track. The TTC also clearly should have more in-house maintenance and track condition analysis capacity — and the funding to secure the staff and equipment to do this efficiently should be delivered.