Some of that data was used in an interview with Republic Parking IT manager Stephen Smith, during which he was challenged to identify parking structures based on graphs of their fill patterns. And he did pretty well, especially considering he was balanced on the end of a teeter totter.

Those graphs are available as a part of the introduction to that interview.

[link]

stopped collecting data about a year ago; he released the source code here

[link]

and it has some amount of data from a year ago still visible, enough to illustrate what could be done.

Two years ago there were independent assessments of detailed time of day structure capacity prepared by multiple individuals. The system currently produces minute by minute accounting of space availability. I don’t see any impediment to the DDA reporting on that to the public, short of the political will to deal with the consequences of that knowledge being widely available for citizens to draw their own conclusions from it.

At the Wednesday committee meeting, some concern was expressed that when people see 35% or 40% or 50% on these graphs, they’ll conclude that the parking structures are vastly underused, because not everyone will reflect on what the denominator is for the calculation. I do think it’s worth taking seriously what the denominator is for that percentage: the total available hours in the structures, Monday-Saturday, 24-hours a day.

So 100% occupancy would translate essentially to round the clock car storage. But there are large swaths of time when I don’t think we’re really expecting more than a handful of cars to be parked in a structure (e.g., 1 a.m.-7 a.m.). So I’m not sure that 35% is low for this metric.

I think it will be interesting to see the eventual breakdown of occupancy during peak periods, say 11 a.m. – 1 p.m., and look at how different structures show different filling patterns during those times. This absolute measure of efficiency/occupancy is, I think, less interesting than the possible use of this metric as a benchmark for measuring the impact of different time/geography-based pricing schedules, which the DDA intends to deploy as part of its transportation demand management strategy. I think it’s reasonable that for each structure a “realistic” overall occupancy rate is calculated for that structure so that we can then check to see how much closer we’re getting to that realistic maximum occupancy rate.

I also think it’s reasonable that these various transportation demand management strategies be scrutinized based on their actual statistical impact. If letting parking prices vary (by time of day and geographic location, or location within a structure) doesn’t have some measurable (positive) impact on the efficiency of the parking system, then it’s reasonable to expect that these various transportation demand management strategies should be revised for another try, or eventually eliminated — if we discover we can’t achieve what we want with respect to efficiency.

One value of this metric is that it’s not tied to revenue. Transportation demand management is supposed to be about encouraging people to park in a pattern that’s best for everyone, not about squeezing as much possible revenue out of the system. But if transportation demand management strategies result in vastly more revenue, there will be some who point to that alone as sufficient evidence that “it’s working.” It’s not. You have to show that you’re able to accommodate a greater number of people in your parking system in order to claim success. I think this occupancy percentage measure that the DDA will now be receiving from Republic Parking is at least a step in that direction.