Drive times on VMS’s - how are they calculated, and how accurate are they?

Started by KCRoadFan, October 29, 2022, 09:29:33 PM

Previous topic - Next topic

KCRoadFan

I've always been fascinated by the Variable Message Signs, or VMS's, along the freeways of major metro areas, that purport to show the estimated drive times to upcoming major junctions, based on the speed of the traffic. Here's what I wonder about them: how, exactly, do they calculate those times? I suppose they're probably equipped with some sort of transponder or sensor that determines the average speed of the traffic at the VMS's location, at which point the onboard computer software takes the distance from the VMS to the selected reference point (an exit, etc.) and divides it by the calculated speed figure, displaying the resulting time figure rounded to the nearest minute. I suppose that's how it's done, but it wouldn't account for traffic jams that happen to start in between the VMS and the listed reference point. I would like to see if anyone knows how the displayed drive times are determined. Also, based on your driving experience, how accurate do you find the VMS's to be? I'm interested to hear more.


Rothman

Wonder if the NYC's times are still based on the TRANSMIT system (E-ZPass) or if it's all based on cell phone data now.
Please note: All comments here represent my own personal opinion and do not reflect the official position(s) of NYSDOT.

jay8g

There are several ways I've heard of to do this. I believe WSDOT's system (at least in the Seattle area) uses the very large number of speed-sensing loops they have in all of their freeways. Seattle used to use LPR cameras and wi-fi readers, but has since dumped them and just buys cellphone data.

MASTERNC

I don't think the ones in Philly are very accurate.  They often say there's only a few minutes of delay when the actual delay is much larger (based on Waze data).

Rothman

Quote from: MASTERNC on October 30, 2022, 06:54:54 PM
I don't think the ones in Philly are very accurate.  They often say there's only a few minutes of delay when the actual delay is much larger (based on Waze data).
On the other hand, I've suspected that Google/Waze manipulates data and travel time reporting to incentivize detouring.
Please note: All comments here represent my own personal opinion and do not reflect the official position(s) of NYSDOT.

Thing 342

I once had an internship at a company that develops software that centrally manages VMS and travel time systems for a few state DOTs (iirc VDOT and PennDOT were customers). The traffic analysis and travel time data were provided by INRIX IQ, which I believe sources data from cell phones and in-road sensors.

roadfro

In Las Vegas and Reno, there are speed sensors posted on the side of the freeway (example) at semi-regular intervals along the freeways (I seem to recall either a video, webpage, or conversation that indicated the sensors in the Las Vegas area are posted roughly every 1/3 mile). The network of sensors (including, in some cases, speed sensors along freeway-to-freeway ramps) feed into a traffic management center. Each sensor is within a known segment of freeway. So by taking the average speed reported in each segment, you get an approximate travel time to a specific destination as reported on the signs. With this approach, travel time can take into account a jam or other slowdown downstream of the sign where the time is displayed. This approach also allows display of travel times to exits not along that freeway (example from I-80 in Reno).

I've found that the travel time information in Las Vegas and Reno tends to be pretty accurate. Any time I've looked at the TT displays in free-flow conditions, it's been pretty reliable (assuming I'm not going much faster than the speed limit). I don't see jams too often in my daily travels, but the few times I have come across a slowdown and have thought to see how accurate the travel time info was, I can recall it being fairly close.

BTW: Some travel time displays in the Vegas area and most in the Reno area are displayed on a static TT sign with variable displays for just the travel time, not a traditional VMS, as linked above. I actually like this approach a little better, as it frees up the VMS for more important messages, and you don't lose TT info when there's an important message to put on the traditional VMS.

Roadfro - AARoads Pacific Southwest moderator since 2010, Nevada roadgeek since 1983.



Opinions expressed here on belong solely to the poster and do not represent or reflect the opinions or beliefs of AARoads, its creators and/or associates.