A wrinkle in time ....

[Anonymous]

Sioux City Journal, December 5, 2004

Clocks to post estimated times it will take trains to cross intersections

By Michele Linck, Journal staff writer

Vehicles line up on Friday on northbound Dakota Avenue in South Sioux City while waiting for a train to pass a crossing near 28th Street. (Staff photo by Tim Hynds)

You're driving down Dakota Avenue, approaching 27th Street, when the red lights begin to flash and the railroad crossing arms descend.

You can only guess how long you'll have to sit there this time. But in the near future, you'll know. An electronic clock posted at the crossing will count down the estimated time it will take the train to clear the intersection. Motorists waiting on the other side of the tracks will have a clock, too.

It's all part of a national Intelligent Transportation System, or ITS, pilot project, the first of its kind in the nation. The study will try to determine whether motorists, when given information on how long the intersection will be tied up, will change their behavior by choosing an alternate route, by just being more relaxed about the wait, or even by deciding not to try to beat the train.


Chad Kehrt, city engineer, said South Sioux City is buying the clocks. The rest is being done under a federal research grant by the state Department of Roads, the engineering department at the University of Nebraska-Lincoln and by Eva Signal Corp. of Omaha, with the cooperation of the Burlington Northern Santa Fe and Union Pacific railroads. The tracks are owned by BNSF and UP has joint use of them.

Eva Signal will place censors along the railroad track about a mile from Dakota Avenue in both directions. The censors will detect the presence of a train, as well as its length and speed, said Cliff Shoemaker, an Eva consultant for the project. It will send that information by radio signal to an electronic "logic center" at the crossing. The center will do the math and continually tell the sign how much longer the train will be in the intersection.

Shoemaker said most of the 16 trains per day which travel the route are going 15 to 20 mph. Given that speed, plus the time it takes for the crossing arms to come down and go up, each train takes six or seven minutes to go through. So, the sporadic traffic back-ups they cause on Dakota Avenue can be a problem.

"They think there are a lot of applications (for this software) around the country," Shoemaker said. "A lot of communities have more than one crossing or access. In South Sioux City, this is the only major crossing. It's being tested here because there is some delay and driver frustration. And, it's a fairly simple system at that location, so it's easy to test."

He said South Sioux City was chosen for the pilot largely because of its willingness to participate.

The countdown clocks won't change the number or speed of the trains, but researchers will find out if they change people's attitudes or behaviors.

Relax, wait almost over

Dr. Aemal Khattak, an assistant professor of civil engineering at the University of Nebraska-Lincoln, is the project's principal investigator. Khattak, who does research in Intelligent Transportation, said he is designing a method to test what effect knowing the duration of the tie-up has on drivers.

"Unfortunately, there aren't a whole lot of alternate routes available whereby drivers could be diverted," Khattak said. "But we're hoping if they know the crossing is going to be closed for 5 or 6 minutes they might be able to run an errand or use the time more productively. Also, we feel once you let them know they're going to have to wait five minutes, they'll be a little more relaxed.

"And," he added, "there's a chance drivers who try to beat the train might be less tempted if they know the crossing's going to be closed for 3 or 4 minutes, so there's a safety aspect."

Khattak said he will do before-and-after studies of driver behavior to see what difference having the information makes. He said the "after" studies will done awhile after the clocks are installed so that drivers are used to them.

[Overmod]

My prediction: more, and wilder, attempts to get across before the gates come down, or around them before the train gets there. Especially since the average delay is as long as it is... and the train speeds as slow.

I'm just a tad depressed that academics can be this shortsighted. Watch the hand-washing and buck-passing when the trial lawyers start apportioning blame for the crossing accidents! "Safety aspect", my tail!

A relatively simple "cure" -- fire up the clocks only AFTER the gates are down and locked, and provide some 'legal' means for people at the crossing to make U-turns or whatever when they see what is going on. Look for interesting and somewhat ad hoc attempts to make this gobbler fly once the early results are "interesting"... ;-}

One approach to this same question involves the use of different kinds of alert. I noticed years ago (in Bergenfield, New Jersey) a strobe light in an adjacent parking lot which activated a couple of minutes before trains on the West Shore reached a crossing... gave me adequate time to find a parking space and get near trackside before the actual lights and gates came on. Something like this, as part of the crossing-protection displays, would be a preferable option to something that starts only when the "legal" crossing protection is activating... and it could be seen and recognized from further away than the clocks will probably be 'legible' to drivers, perhaps far enough away to allow drivers to divert to side roads or safely make turnarounds. Might help to have some indication which way the train in question is coming from, too, particularly since there aren't any 'unblocked' alternative crossings (according to the report as I understand it) so time is of the essence in getting to an open crossing by going the 'right' way...

Then you have colored indicators that tell the length of time remaining -- say, an orange light that indicates more than 5 minutes delay, and a blue one that comes on to show less than, say, two minutes remaining. (And that coincidentally only shows when no other train is approaching that crossing). You could accompli***he same thing with slow, then faster, blinking of a single light. But NOT with a digital display.

If I recall correctly, there were some digital traffic-light displays tested years ago in Florida, where the yellow light counted down to give drivers an idea of how quickly it would turn red. I did not see this experiment repeated, and have not come across particularly detailed discussion of the outcome, so I suspect this had exactly the effect I expected it would: massive incentive for drivers to get their nose into the intersection before the last 'second', *while watching the location of the yellow light instead of traffic elsewhere in the intersection*. Makes you wonder what cereal box had the human-factors-engineer certificate on it that year... ;-}

Note that it is possible to pass this information to some types of modern car radio, complete with alternate routing information (e.g., directions to crossings that will be open more quickly). This might be a useful addition to traffic crews and radio stations that broadcast train information to listeners, including advice on alternate routes and crossings.

[CShaveRR]

"Censors?" Are they going to tell us which trains we can't watch, or photograph?

Maybe the censors will employ sensors to tell them who many railfans are going to be waiting how long for the approaching train.

[;)]

P.S. Timers that count remaining time at pedestrian walkways are fine, but people close enough to the crossing to read a digital display are close enough to see what kind of train it is, how fast it's moving, and figure for themselves whether they want to be stuck there. And how does this timer predict changes in train speed? I know some crossings where the computer would go up in smoke within a couple of hours!

[jeaton]

This is really interesting.

I will be honest and say that I don't know for sure just how the people of Souix City, Iowa tend to react to the approach of a long, slow train. My guess is that they are more inclined to observe the safety warning of dropping gates and not try to beat the train. I can think of many other places in the country where it would seem that the potential for even momentary delays to forward progress bring out the win at all cost side of behavior.

The findings of this experiment might be very enlightening.

Jay

[JoeKoh]

Somehow this new wrinkle will still not prevent some drivers for putting a new wrinkle in their cars as they get hit by the lowering gates or the passing train.by the way what effect will weather have on these gates?snow ice and road salt can effect the timers too.
stay safe
Joe

[Anonymous]

My thought, is if these drivers consistantly see a wait period of 4 minutes, then all of a sudden they see a wait period for 7 minutes, aren't they going to be wanting to try and beat the train even more? That's 3 more minutes than usual!

Seems to me that would be the case.

[espeefoamer]

I see this causing a rise in idiots trying to beat the train across the crossing if there is going to be a "long" wait.Some people would risk a quick trip to the morgue rather than wait 6 minutes at a crossing[:(!][xx(].

[dldance]

Texas A&M university has been doing some similar experiments in College State. See the following website:

http://railview.tamu.edu/rail_monitoring_xml.htm

They have put sensors on several miles of track and also have video feeds from 4 locations. They feed this data into a chart, updated on the web, the predicts the ETA/ETD and confidence in the estimates for several grade crossings -- one of which is across from their football stadium.

dd

[tree68]

I'll second holding the display until after the train occupies the crossing, unless the crossing is impassable once the gates go down (ie net, quad gates, other barrier).

It does occur to me that a secondary display should be located at the place where a driver can choose an alternate route, if it's possible, and if it's not too far back. Thus, if the time remaining is within the driver's acceptable parameters he/she heads for the crossing. If an alternate route would be faster, away they go!

Of course, that does nothing for a situation like we had when CSX dumped 8 cars north of Watertown. The crossing was blocked for an hour and a half.

[Mookie]

UNL should stay at home and watch the same crossings I do. 1st & 2nd and J. They could - as Brother Carl sez - burn up a lot of sensors watching the traffic that uses these two crossings. And they could also put a "ticket writer" there who could put a lot of $ in the city coffers for the people who go around the gates and some that not only go around the gates, but then use the railroad right-of-way to cut between the set of 4 and set of 2 tracks. Better yet, I would do it - cheap!

Mookie

[Anonymous]

My prediction: more, and wilder, attempts to get across before the gates come down, or around them before the train gets there. Especially since the average delay is as long as it is... and the train speeds as slow.

This is very true. It's my observation that most "A type" Americans see a caution light not as a signal to stop, but to go faster. All you do when you tell them how many seconds they have left to get into an intersection (or a rail crossing) is give them a challenge.

It would be more interesting to have a display showing how much longer one has to wait at a crossing before the train clears. This would keep people from doing the "turn around and finding an alternate route" thing.

Erik

[rrnut282]

I'll chime in with the majority on this one. If the display is activated before the gates, more will attempt to go around the gates. If the display is a countdown timer for the crossing to clear, it will give the average driver, (ie not a railfan) something to watch and make the time pass with less stress. If they insist on a predictive display before the gates go down, place it far enough away from the actual crossing to give drivers time to take the alternate route or make an informed decision about a pit stop prior to the crossing to avoid the fender-benders when too many are attempting a three-point turn at the same time.