Determining truck route roundtrip time from GPS

[TroyKechely]

I have exhausted my efforts at figuring this out on my own so I am reaching out to you all.  I'm running the ArcGIS Pro Advanced license with the GeoAnalytics tools and I have over 1.4 million GPS points for more than two dozen trucks over a 30 day period.  I have a single known starting location and multiple destinations that I have added as a text field to the table of the GPS data. Trucks leave the the start location, go to one of the destinations, remain there a varied amount of time (need to get average, max and min for this as well), and return to a location near the start point that is identified as the end. I need to calculate the average, max, and min travel time for round trips for each destination over the entire month.  I have tried the dwell and detect incident tools in GeoAnalytics with no success. Likely because of my not understanding how to use them.  Is this even possible and if so, please help me figure out how to best accomplish it keeping in mind I have minimal programing knowledge.  Thanks. @GeoAnalytics 

[DanLee]

Could you show a screenshot of your data/table? Do the points from each round trip have unique trip IDs? Are the date/time information in the point table?

[TroyKechely]

Here is that screenshot of the GPS points table. Unfortunately they do not have a unique trip ID.  There is date and time info.  The Dwell_Area field is defines if the points are in start/destination areas or traveling between sites (Route).

[DanLee]

Thanks for the screenshot and information.  This is a tough one without trip ID.  The key is to determine which point is the start of a trip, which point is the destination, and which point is the end of the trip.  I don't have a solution on top of my head, but if you could select points from two different trips by the same truck, i.e. trip 1: from location A to destination 1, then back to location A; trip 2 from location A to destination 2, then back to location A, I might test and play with some ideas.

[DanLee]

By the way, do you have the start and the multiple destination locations in a separate feature class or are they embedded in the GPS points?

[TroyKechely]

Both. I have a polygon feature layer so I could identify how long they were within the facility and I used the name of each facility to populate the dwell_area field in the gps point layer.  I was thinking about creating a just a point feature layer to represent the destinations thinking that might be best or even just one at the start (scale), and one at the end (tipper).  This might be what you are thinking with your prior response with point at start, destination, then end.  If so I will create that layer.

Thanks for your help.

[DanLee]

I am not sure yet that the start/destination points are necessary, given that you have the Dwell_Area field. I am imaging a potential workflow, but I would ask again - would you be able to share a sample data from a couple of trucks and a couple of days GPS points? 

Points in Dwell_Areas from different trucks or the same trucks in different days can be selected. I guess it's a matter of sorting out which ones of them belong to separate trips by analyzing their Time values and TruckNames. Once points of unique trips can be identified, calculating the total time spent in particular Dwell_Area would be possible.  The same logic might work for points of Route.

[TroyKechely]

Sorry I missed the request a sample.  What format do you want that in?

[DanLee]

A point feature class in a gdb would be great.

[TroyKechely]

I hope I did this correctly.  This should be three trucks over a couple days doing routes to multiple sites.  The truck route should start at the Landfill - Scale, go to the destination and swap trailers, then return past the scale and end at the Landfill - Tipper.