PDA

View Full Version : Route Calculation and Timing Problems



Roger Allen
12-19-2011, 02:16 PM
Hi:

I have a series of about 40 locations.

In VB.net I'm creating an instance of mappoint (2011) and calculating the drive time between each point and every other point.

I calculate a route that optimizes the number of locations that can be achieved within a fixed time period.

HERE'S THE PROBLEM

I then import the waypoints and route into another instance of mappoint, recalc the route and display it.

For a 10 hour route, it will typically show that it will take 11 to 12 hours.

If I then reverse the route twice, (making no other changes), it will reduce the route time to 10 hours!

This is completely reproducable and the number of waypoints, and length / max time of route makes no difference.

If this a known issue?

Thanks

Roger

Mattys Consulting
12-21-2011, 11:28 PM
I was wondering if the whole route is exactly the same when you import?
Does it recalc and still come out with the same route, but different overall trip time?
I would think that the average speed would increase over a known route, but 1 to 2 hours less is a stretch ...

Roger Allen
12-22-2011, 12:34 PM
I was wondering if the whole route is exactly the same when you import?
Does it recalc and still come out with the same route, but different overall trip time?
I would think that the average speed would increase over a known route, but 1 to 2 hours less is a stretch ...

I'm not importing a specific route.
As a small example of the process, if Ihave waypoints A, B C, D, & E I calculate the driving time for:

A,B
A,C
A,D
B,A
B,C
B,D
etc

Lets say that I find the drive times for A,D then D,B, then B,C add up to an acceptable value, then I start a new instance of mappoint and import waypoints A,D,B,C and calculate the route.
The problem is that the total drive time does NOT equal the sum of the individual leg values, it is usually 10-20% more.
By revesing the route twice, the total does APPROXIMATELY equal the sum of the legs.

I would have thought that the sum of the individual leg calculations would have exactly equaled the total route, but this is never the case.

I have already found that I have to calculate both direction on the individual legs, in the case above B-A drive time does not calculate the same as A-B, sometimes its a small difference and sometimes it's significant.

Mattys Consulting
12-22-2011, 04:32 PM
It sounds like there is a random multiplier in the calculation of times per leg and also for the total.
If one were to do a something like Monte Carlo simulation, perhaps one would obtain best, worst, and average cases.
That's simply speculation - I don't find the DriveTime Zones any more accurate than route times.