|
|
|
Go
![]() |
New
![]() |
Find
![]() |
Notify
![]() |
Tools
![]() |
Reply
![]() |
|
|
Junior Member |
Hello,
I am working on SimTraffic Simulation for a project. I found an interesting issue at couple of the intersections with the 95th %ile Queue Lengths greater than the Maximum Queue Lengths for couple of movements. The difference between the Maximum queue length and the 95th %ile Queue Length is approximately 100 feet. I understand that the 95th %ile is just based on statistical calculations and is not necessarily ever observed in the field. Still, I could not really understand how this is possible to have higher 95th %ile than the Maximum value. Any ideas, Radha |
||
|
Member![]() |
Rahda,
I would reference you to previous discussions on this topic. Here is one: http://trafficware.infopop.cc/eve/forums/a/tpc/f/57.../334102757#334102757 Sincerely, Jeff Gerken |
|||
|
|
Member |
the 95 percentile is adding up the average with multiplle of standard deviation. from any basic statistics book you can find that the value of standard deviation depends on the deviation of the distribution. so 95th percentile can be higher than maximum value. this is not just for queues, for pretty much any statistical distribution. hope it helps.
|
|||
|
|
Junior Member |
Jahid,
Didn't I teach you anything? The 95th percentile is based upn a calculated queue, i.e. 1.65 standard deviations added to the average queue. This is not possibly ever observed. The max. queue is the maximum observed queue for the analysis period. One is seen, the other calculated. Let me know if you need anything further. Scott Casey & Godfrey Engineers |
|||
|
| Previous Topic | Next Topic | powered by eve community |
| Please Wait. Your request is being processed... |
|

