Well, they could always do what smart cities do and sell a contract to a bus shelter advertising company and let them maintain them, having to live up to SLAs to keep their contract.

Bus stops aren't always bus shelters and the problems with them may have to do with things like access due to sidewalk conditions, access and safety issues related to street configurations & cross-walks, crime / safety issues related to lighting or nearness to hiding places for would-be attackers, just to provide some examples. Any shelter contracting model would need to include requirements to keep up lower-traffic (= fewer eyeballs per time unit) shelters equally with those that advertisers would like better, and still would involve monitoring costs to check on fulfillment.

Oh, and congrats to OPAL!