Hello,
Among all the different simulations I have run, even with a lane change mode of
1557 (which activates everything but the drive/keep right changes to avoid any bias between lanes and treat them all equally; 2^10 + 2 ^9 + 2^4 + 2^2 + 2^0
). For some reason, simulations end faster (in lesser steps) when a blockage is placed on the lane with index 0, i.e., when the lane with index 0 is blocked - vehicles seem to be finishing their route faster. The term "free passage" just means that there is no stationary blockage (stopped vehicle) on that lane.
Some quick results to prove this case in vanilla SUMO runs (without TraCI control):
Steps in seconds - number of lanes, blockage on lane 'x'
262 - 2 lanes, blockage on 0
286 - 2 lanes, blockage on 1
176 - 3 lanes, blockage on 0
186 - 3 lanes, blockage on 1
185 - 3 lanes, blockage on 2
147- 4 lanes, blockage on 0
148 - 4 lanes, blockage on 1
156 - 4 lanes, blockage on 2
154 - 4 lanes, blockage on 3
The length of the entire route (and road/edge) is 1000m. The end of the blockage (or stationary vehicle) is placed at 750m, with a size of 300m. So the blockage is from 450 - 750m on any given lane.
<vType carFollowModel="Krauss" color="0,1,0" id="veh" latAlignment="center" length="5" sigma="0" speedDev="0" speedFactor="1.0" />
<flow arrivalLane="current" arrivalPosLat="center" begin="0" departLane="random" departPosLat="center" departSpeed="random" id="f" number="100" route="r0" type="veh" vehsPerHour="10000" />
Is there something inherently different about a lane with index 0? Why are the experimental results always proving this as the case even though the vehicles are being inserted on random lanes at random speeds? Please advise.
Thank you.
Sincerely,
Hriday