Page 1 of 1

linear interpolation of emissions?

Posted: Fri Sep 01, 2023 11:41 am
by Bernd Welter
HI there,

my colleague from Italy asked me this question in the context of PTV Developer:
Daniele wrote:A customer of us using PTV DEV asked how to retrieve back the emissions for each countries touched by the routing; my suggestion was to activate border events and divide the initial routing into subrouting from start waypoint to first coordinates of border events + first coordinates of first border events to second coordinates of second border events+...+n-1 coordinates of n-1 border events to end waypoint;

In a note an excel in which I did a test of a routing from Perugia (IT) to a location in DE.
You finde the PTV DEV requests and emisions total that is exactly the sum of the partial ones:
I obtained two border events, that contains coordinate of CH border and coordinates of DE Border.

So I created the routing request with same date/hour for start waypoint to first border coordinates event and obtained the
emission in IT (SUBROUTE1)
Then I created another routing from first border coordinates event and coordinates of DE Border events, so I obtained the emission in CH (subroute 2).
Last part is the routing from coordinates of DE Border events to end waypooints, so I obtained DE emission (subroute 3).
This is the values I obtained with a quick test..if you sum the values of subroute 1,2,3 of each value you obtain the correspondant values of the TOT routing:
ScopefuelConsumptionco2eTankToWheelco2eWellToWheelenergyUseTankToWheelenergyUseWellToWheel
TOTAL464.37..1490.22..1808.36..20037.122..23832.45..
Subroute 1187.40601.39729.77..8086.1359617.77
Subroute 2115.19369.66448.584970.385911.84
Subroute 3161.78519.19630.036980.998303.29
Is this the way to follow?
My response: NO. And here's why: whether this approach returns compareable results or not depends on different factors - take some moments to understand the underlying effects:
  • The emission calculation on a "route", a single "leg" or a "segment" may depend on more than just the "relative length": Some emission standards such as HBEFA based ones use more attributes of a segment, e.g. the elevation. That means that two segments with equal length could return different emission values.
  • Furthermore the loading status of a vehicle can change from one wayoint to another. If the standard you use uses the total weight ( or load weight) of a transport vehicle the linear approach fails, too.
Long story short:
:!: Under some cicrumstances you could apply the business logic described above - but not in general.

Thanks to Daniele for asking the question - thanks to Sebastien for guiding us through the obstacle!

Bernd