Hi Caner, OK, you did not mention the different seeds in your first post. Now it is clear.
Best regards, Michael Am 10.03.2016 um 17:38 schrieb CANER İPEK: > Thanks for reply, > In my design, insertion of vehicles are far away from the intersection region > (around 500 m. away) and rate of vehicles are low. > I think that the difference is because of the probablity definitions in the > route.xml files. When the parallel simulations are created,different seed > numbers are produced for both simulations hence the results are changing. > Regards,Caner > > > > > > >> Subject: Re: [sumo-user] route data problem in parallel simulations >> To: [email protected]; [email protected] >> From: [email protected] >> Date: Thu, 10 Mar 2016 16:36:33 +0100 >> >> Hi Caner, >> it depends on your insertion parameters (departSpeed, -Pos and -Lane) >> how much the downstream traffic will affect vehicle insertion so even >> with a completely deterministic setting a jam which reaches back from >> the traffic light to the insertion edge will affect the insertion >> process because the insertion point may be blocked. Furthermore >> determining the best lane for insertion (or continuing the road) depends >> on occupancy as well. So it is not really feasible to have an identical >> insertion pattern unless the frequency of vehicles is very low and they >> are very far from the traffic light. Maybe you should rather compare >> more aggregated measures like the flow. >> >> Best regards >> Michael >> >> Am 09.03.2016 um 18:20 schrieb CANER İPEK: >>> Hi, >>> I have used TraCI with C++. I make 2 parallel intersection simulations such >>> that signal timings are different in both simulations. >>> The route.xml files are same for both simulations. But, they includes some >>> definitions with probabilities. (probabilistic flow or probabilistic routes >>> etc.) >>> I observed that the insertion of vehicles to the simulations are different. >>> Is it because of the probabilistic definitions in the route file although >>> they are same for both simulations? >>> I am not sure but in the previous studies i used older sumo version and i >>> did not observe this problem. In 0.25 version, creation of vehicles are >>> different as time or route in the both parallel simulations. >>> To handle this problem, should i create each vehicles as certain time and >>> route parameters ? Or any advice.. >>> Thanks, >>> Caner >>> ------------------------------------------------------------------------------ >>> Transform Data into Opportunity. >>> Accelerate data analysis in your applications with >>> Intel Data Analytics Acceleration Library. >>> Click to learn more. >>> http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140 >>> _______________________________________________ >>> sumo-user mailing list >>> [email protected] >>> https://lists.sourceforge.net/lists/listinfo/sumo-user >>> >> > > ------------------------------------------------------------------------------ Transform Data into Opportunity. Accelerate data analysis in your applications with Intel Data Analytics Acceleration Library. Click to learn more. http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140 _______________________________________________ sumo-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/sumo-user
