calculate x Minutes response time

817
3
10-09-2018 06:46 AM
HushamMohamed
Occasional Contributor

I received an Excel file with data on previous 3 years fire and emergency response times. I was able to geocode 85% of the data,  I have the following fields

ReceivedCallDateTime,  DispDateTime, EnrDateTime, ArrDateTime, CLRDatetime,

I asked to to map the response times without the Network Analyst since we dont have well configured road network,

 which to me looks more like a sort of heat map. They'd also like maps for each of five  fire stations. They're sort of unclear on exactly what they want; they may not know. I think essentially what they want is a map for each station showing 3 and 6 minute response time zones based on their actual data. I could do a multiple ring buffer based on an average speed, but that doesn't take into account actual streets.

Or just if I can run the IDW spatial analyst tools and get the response head map.

All help appreciated, thanks.

Tags (2)
0 Kudos
3 Replies
DanPatterson_Retired
MVP Emeritus

You would require the network analyst to use vector streets to produce the spatial allocations along the network.

This can be emulated using the spatial analyst extension using a cost path constrained to a raster representation of the road network.  The difficulty with this approach is maintaining continuity of the road in raster format

Curtis Price‌ did this sort of thing recently 

https://community.esri.com/message/779960-can-i-use-cost-distance-tools-to-model-travel-times 

0 Kudos
HushamMohamed
Occasional Contributor

Thank you Dan,

That  approach will not work for me,  because I don't have a reliable street network, besides I already have the response time in minutes,  I think I will end up with just a heat map for my purpose.

Thanks

0 Kudos
curtvprice
MVP Esteemed Contributor

Seems to me what you want to do is measure distance using cost path and calculate a regression with your dataset to estimate what you'd get everywhere (as apposed to just at your points. But if your point density is high enough, just building an interpolated (eg IDW) surface from your response time points would probably meet your purpose. 

Heat maps count density of points so I don't know if that's what you are looking for.

0 Kudos