Open kielejocain opened 7 years ago
Is reapplying this the OSU simulator the official route we're taking? One consideration is our data has less resolution than the OSU study because we're obscuring incident locations to fire blocks. I don't expect it to be an issue, but I'm not sure of the sensitivity of the OSU model.
Based on the call notes (https://github.com/hackoregon/emergency-response/wiki/2016-12-22-Call-Notes) I believe the intent is to float the simulator as a parallel, longer-term track we could pursue. This would definitely be the lower priority track right now, though.
Due to our abbreviated schedule, we'll need to make sure we prioritize producing at least an MVP for the March demo day, and the simulator+research track is not a short one.
As for model sensitivity, it's hard to say, and I expect would be part of the research. It's also possible that eventually PF&R would agree to something similar to what Corvallis did, where call locations are rounded to the nearest 200 feet (or some other value), if we could express a need for greater resolution.
We need to define what the simulator will take in and return, probably in coordination with the OSU professors.