Traffic analysis statatistics

[chrisday]

Hello everyone,

new to the forum and traffic engineering in general. I was hoping for a point in the right direction with a traffic study problem.

I have automatic traffic counter data for for a 2-way road. I am seeing if a pedestrian crosswalk is warranted at the location.
My output data from the traffic counter is arranged by the quantity and length between vehicle gaps occurring over each one hour counter interval for 24 hours. I am trying to figure out the average wait time for a pedestrian that needs a full 9 seconds to safely cross the road during peak hour.

Thanks,

Chris

 

[IRstuff]

Do pedestrians have the right of way in your location?

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[IRstuff]

Oh, and what speed are the cars?

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[chrisday]

Peds do not have right of way at this point and the average speed of vehicles is 31 mph. The highest number of gaps for pm peak is 557 gaps per hour or 558 vph. 2-way peak hour traffic volume = 1218 vph. Divided into 3600 seconds/hr gives 3 seconds between vehicles but vehicles obviously are not perfectly spaced out in real life. A pedestrian would never be able to cross if a vehicle went by every three seconds. This is why I thought having the actual veh. gap intervals might be helpful.

 

[IRstuff]

So, you need to have concurrent gaps of at least 125-m in both directions, right? Is that sort of information available in the data? If you throw in some hesitation time, then you'd need more like 152 ft in both directions. The hesitation time would be needed for the pedestrian to look both ways and still be confident that they won't get run over. I threw in 2 seconds, but that might still be low.

BTW, in California, pedestrians have the right of way regardless of the existence of a crosswalk.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[chrisday]

Right, I am using a crossing time of 9 seconds. Do I need to calculate a weighted average from the gap data? Ideally I am shooting for wait time of less than 30 seconds for an adequate gap to cross, (per mutcd). I am trying to either justify yea or nay for a cross walk to be installed. It seems like there should be some statistical formula to figure out the average wait time, or at least the probability of a 9 second traffic gap in any given 30 second period.

 

[IRstuff]

I assume crossing time is just the time it takes to move from one side to the other? Otherwise, the time looks to be too short for a 3.5 ft/s pace crossing a 30-ft gap. But, I think you need to include the pedestrian's time to determine whether he can safely cross. That's got to be a couple of seconds to do that alone.

Your counter data presumably has each occurrence tied to time. Are the two directions separately tallied?

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[DRWeig]

Peds do not have right of way at this point

Is this in a state in the US?

Best to you,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[ACtrafficengr]

What is your goal? Do you just want to give the pedestrians the right-of-way, or make the crossing substantively safer? Is there a pedestrian crash history there? Are there any risk factors like restricted sight distance, a wide crossing, etc? If the average speed is 31, what is the 85th% speed?

It's a little stale now, but the seminal research paper on uncontrolled (no signal or stop signs on the vehiclular approaches) crossings found:

The study results revealed that on two-lane roads, the presence of a marked crosswalk alone at an uncontrolled location was associated with no difference in pedestrian crash rate, compared to an unmarked crosswalk. Further, on multilane roads with traffic volumes above about 12,000 vehicles per day, having a marked crosswalk alone (without other substantial improvements) was associated with a higher pedestrian crash rate (after controlling for other site factors) compared to an unmarked crosswalk.

http://www.fhwa.dot.gov/publications/research/safety/04100/index.cfm

The authors recommend more than just preformed tape across the road if you want to make it safer. Actuated signals or beacons and raised median refuges were recommended.

@Goober Dave, considering his units, he's probably in the US. I believe we're the only country using that archaic measuring system other than Liberia and Myanmar.

 

[TerryScan]

Is there an excessive amount of data? If one is only looking at a day or two, it appears the simple way to evaluate the data is to manually review the data:

1.Highlight all gaps 9 seconds or greater. (crossing potential)
2.Calculate each time segment between THESE gaps. (waiting periods)
3.Compute statistics on the waiting periods. (This should give the average waiting time you were seeking in original post).

It should be easy to do in a spreadsheet.

 

[chrisday]

Hey thanks for the input. I think I got it figured out this morning.

Chris

 

[DRWeig]

Hey chrisday, fill us in! We appreciate follow-up.

Best to you,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[chrisday]

Sorry!

I pretty much did the steps that Terryscan mentioned. I got an average peak pm wait time of 45 seconds. Not too good for pedestrian patience I suppose.

 

[IRstuff]

But, that wouldn't be that different that the wait time at a traffic-light controlled intersection, would it?

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[DRWeig]

Thanks. I'm not a traffic guy, but having lived in Atlanta -- 45 seconds would be a dream.

Best to you,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[chrisday]

45 seconds probably wouldn't be that much longer than signalized crossing but according to the MUTCD peds get antsy and might take unsafe crossing chances if the wait is too long at a busy unsignalized crossing.

Chris