Why Do Engineers Do So Poorly with Intersection Sight Distance ?

[Brian556]

Deficient intersection sight distance is the most common engineering error that I see, and is by far one of the most dangerous. Why do engineers fail at this? Are colleges not teaching or emphasizing it enough? Another part of the problem might be that a lot of people pick engineering as a major eenie-meenie-miney-moe style, and don't really have the knack for it.

The two major types of this are 1) Where drivers on a side street or driveway cannot see traffic on the major street, 2) Where drivers turning left from a divided street or highway cannot see oncoming traffic if the oncoming left turn lane is occupied. This problem exists on every divided street there is, and is well-known, yet, they continue to do new construction the same way.

On a newly-opened section of FM 2499 in south Denton, they have had to keep the right lane closed because they put a stone wall right up against he road, and traffic comting out of a side street cannot see through it. https://www.google.com/maps/@33.1467893,-97.0974741,3a,42.1y,10.62h,84.42t/data=!3m6!1e1!3m4!1s2GNTTUFWz19amAaDEk4bQg!2e0!7i16384!8i8192Flash back to 2010, when they opened the previous section of FM 2499, they made the same mistake at another intersection, and obviously they didn't learn from that.

How can people capable of obtaining an degree in engineering be so stupid?

[silverback1065]

I don't think it's them being stupid. Sight distance is taught and is required to pass the PE exam it's on a lot of problems. In fact most of the questions on the transportation PE exam are about situations you describe. I believe it's even on the EIT exam too. In Indiana it is required to check sight distance before you begin design.  In some situations it can be the project owners fault in terms of bending the rules in my experience.  And in some situations it's not even an engineer designing a project, just a land developer doing it and a PE just rubber stamping it (bad practice).

[silverback1065]

Deficient intersection sight distance is the most common engineering error that I see, and is by far one of the most dangerous. Why do engineers fail at this? Are colleges not teaching or emphasizing it enough? Another part of the problem might be that a lot of people pick engineering as a major eenie-meenie-miney-moe style, and don't really have the knack for it.

The two major types of this are 1) Where drivers on a side street or driveway cannot see traffic on the major street, 2) Where drivers turning left from a divided street or highway cannot see oncoming traffic if the oncoming left turn lane is occupied. This problem exists on every divided street there is, and is well-known, yet, they continue to do new construction the same way.

On a newly-opened section of FM 2499 in south Denton, they have had to keep the right lane closed because they put a stone wall right up against he road, and traffic comting out of a side street cannot see through it. https://www.google.com/maps/@33.1467893,-97.0974741,3a,42.1y,10.62h,84.42t/data=!3m6!1e1!3m4!1s2GNTTUFWz19amAaDEk4bQg!2e0!7i16384!8i8192Flash back to 2010, when they opened the previous section of FM 2499, they made the same mistake at another intersection, and obviously they didn't learn from that.

How can people capable of obtaining an degree in engineering be so stupid?

Picking a major, especially engineering is very hard if you don't have a particular passion for a specific aspect of it. It was easy for me, I've loved roads since I was a kid. But most people I've met seemed to have just fallen into it. I mean civil has at least 11 sub majors you can choose from at least at Purdue. I'm not saying all engineers are like this, but I feel a lot lack the passion and at times it can be reflected in their design. I put a lot of thought into the aspects that most seem to think are afterthoughts. This sight distance thing you bring up I feel is a bit of an oversight in some circumstances.

[yand]

On a newly-opened section of FM 2499 in south Denton, they have had to keep the right lane closed because they put a stone wall right up against he road, and traffic comting out of a side street cannot see through it. https://www.google.com/maps/@33.1467893,-97.0974741,3a,42.1y,10.62h,84.42t/data=!3m6!1e1!3m4!1s2GNTTUFWz19amAaDEk4bQg!2e0!7i16384!8i8192

That is bad.

[silverback1065]

Who built that? That probably has nothing to do with the road design, a property owner likely did that. And if that's true I highly doubt they know what sight distance is

[jakeroot]

To address the specific issues, from a non-PE perspective:

1) Where drivers on a side street or driveway cannot see traffic on the major street

I think this is compounded in the US and Canada because most stop lines are pulled back from the main roadway, to allow pedestrians to cross in front of waiting cars. "Stop line visibility" is far too often quite poor, but room for pedestrian crossings have to be provided. So, in practice, drivers stop, then pull forward and yield (while stopped on the crosswalk).

Other countries usually place the stop/yield line right along the edge of the main roadway, so visibility is excellent. Other countries also have laws that allow the local councils to trim back hedges and trees to improve visibility, without acquiring ROW. There may be some states that do this, but I would assume this is more unusual than not. I highly doubt Texas has any rules for this. It would have helped this intersection near Tacoma, WA, but they decided to close off the street instead due to crashes (from poor visibility, I assume), and because it's a single-track road.

2) Where drivers turning left from a divided street or highway cannot see oncoming traffic if the oncoming left turn lane is occupied. This problem exists on every divided street there is, and is well-known, yet, they continue to do new construction the same way.

Texas is particularly bad at this. I recall reading about some double left turns that used FYAs somewhere near Fort Worth, that had to be turned off because of too many crashes. The left turns had negative offset, instead of positive offset, so visibility was terrible no matter where you positioned your car. With a single left turn, it's not too bad (as long as you creep all the way forward). Not ideal, though.

Vancouver has a ton of roads with medians, many with very wide medians. Virtually none of the left turns along these roads use protected phasing, so traffic ends up positioning itself like this (much better visibility than if stopped at the stop line). But, many left turns along these wide roads have been either built properly, or modified later, to make the left turns more pointed directly at each other (no offset or positive offset), so left turning traffic has excellent visibility.

All Texas needs to do is narrow the median a bit, and have the left turn lanes pull away from the through lanes maybe five car-lengths back, so they are pointed more at each-other. Like this (overhead view of previous link).

[Brian556]

To address the specific issues, from a non-PE perspective:

1) Where drivers on a side street or driveway cannot see traffic on the major street

I think this is compounded in the US and Canada because most stop lines are pulled back from the main roadway, to allow pedestrians to cross in front of waiting cars. "Stop line visibility" is far too often quite poor, but room for pedestrian crossings have to be provided. So, in practice, drivers stop, then pull forward and yield (while stopped on the crosswalk).

Other countries usually place the stop/yield line right along the edge of the main roadway, so visibility is excellent. Other countries also have laws that allow the local councils to trim back hedges and trees to improve visibility, without acquiring ROW. There may be some states that do this, but I would assume this is more unusual than not. I highly doubt Texas has any rules for this. It would have helped this intersection near Tacoma, WA, but they decided to close off the street instead due to crashes (from poor visibility, I assume), and because it's a single-track road.

2) Where drivers turning left from a divided street or highway cannot see oncoming traffic if the oncoming left turn lane is occupied. This problem exists on every divided street there is, and is well-known, yet, they continue to do new construction the same way.

Texas is particularly bad at this. I recall reading about some double left turns that used FYAs somewhere near Fort Worth, that had to be turned off because of too many crashes. The left turns had negative offset, instead of positive offset, so visibility was terrible no matter where you positioned your car. With a single left turn, it's not too bad (as long as you creep all the way forward). Not ideal, though.

Vancouver has a ton of roads with medians, many with very wide medians. Virtually none of the left turns along these roads use protected phasing, so traffic ends up positioning itself like this (much better visibility than if stopped at the stop line). But, many left turns along these wide roads have been either built properly, or modified later, to make the left turns more pointed directly at each other (no offset or positive offset), so left turning traffic has excellent visibility.

All Texas needs to do is narrow the median a bit, and have the left turn lanes pull away from the through lanes maybe five car-lengths back, so they are pointed more at each-other. Like this (overhead view of previous link).

The concept that fixes the left-turn visibility problem is called offset left turn lanes. TxDOT used that at this intersection in Marshall. https://www.google.com/maps/@32.5269068,-94.3518117,169m/data=!3m1!1e3

[jakeroot]

To address the specific issues, from a non-PE perspective:

1) Where drivers on a side street or driveway cannot see traffic on the major street

I think this is compounded in the US and Canada because most stop lines are pulled back from the main roadway, to allow pedestrians to cross in front of waiting cars. "Stop line visibility" is far too often quite poor, but room for pedestrian crossings have to be provided. So, in practice, drivers stop, then pull forward and yield (while stopped on the crosswalk).

Other countries usually place the stop/yield line right along the edge of the main roadway, so visibility is excellent. Other countries also have laws that allow the local councils to trim back hedges and trees to improve visibility, without acquiring ROW. There may be some states that do this, but I would assume this is more unusual than not. I highly doubt Texas has any rules for this. It would have helped this intersection near Tacoma, WA, but they decided to close off the street instead due to crashes (from poor visibility, I assume), and because it's a single-track road.

2) Where drivers turning left from a divided street or highway cannot see oncoming traffic if the oncoming left turn lane is occupied. This problem exists on every divided street there is, and is well-known, yet, they continue to do new construction the same way.

Texas is particularly bad at this. I recall reading about some double left turns that used FYAs somewhere near Fort Worth, that had to be turned off because of too many crashes. The left turns had negative offset, instead of positive offset, so visibility was terrible no matter where you positioned your car. With a single left turn, it's not too bad (as long as you creep all the way forward). Not ideal, though.

Vancouver has a ton of roads with medians, many with very wide medians. Virtually none of the left turns along these roads use protected phasing, so traffic ends up positioning itself like this (much better visibility than if stopped at the stop line). But, many left turns along these wide roads have been either built properly, or modified later, to make the left turns more pointed directly at each other (no offset or positive offset), so left turning traffic has excellent visibility.

All Texas needs to do is narrow the median a bit, and have the left turn lanes pull away from the through lanes maybe five car-lengths back, so they are pointed more at each-other. Like this (overhead view of previous link).

The concept that fixes the left-turn visibility problem is called offset left turn lanes. TxDOT used that at this intersection in Marshall. https://www.google.com/maps/@32.5269068,-94.3518117,169m/data=!3m1!1e3

Yes, I'm aware. I mentioned them twice (bolded above), and alluded to them more than that. Did you read my post?

You don't need positive offset left turns to improve visibility. No offset is acceptable, assuming there is room in the intersection for cars to maneuver enough to improve visibility. Hell, negative offset is OK too, if the intersection is long enough.