Do we need inside shoulders on freeways?

[kernals12]

A 1987 study looked at 10 cases from California where inside shoulders were removed to make space for extra lanes (generally in combination with narrowing the existing lanes from 12 feet to 11) and found no increases in accident rates and sometimes a *decrease* due to impact of reduced congestion.

And having no inside shoulders is the norm in most of the world.

There are very few scenarios where a disabled vehicle can make it to the inside shoulder but can't get over to the outside. And as cars become more reliable and electronic safety aids reduce rear-end collisions, the need for breakdown lanes in general is falling. So perhaps it's time to consider getting rid of inside shoulders.

[Rothman]

A 1987 study looked at 10 cases from California where inside shoulders were removed to make space for extra lanes (generally in combination with narrowing the existing lanes from 12 feet to 11) and found no increases in accident rates and sometimes a *decrease* due to impact of reduced congestion.

And having no inside shoulders is the norm in most of the world.

There are very few scenarios where a disabled vehicle can make it to the inside shoulder but can't get over to the outside. And as cars become more reliable and electronic safety aids reduce rear-end collisions, the need for breakdown lanes in general is falling. So perhaps it's time to consider getting rid of inside shoulders.

What in the ChatGPT is this?

[hotdogPi]

What in the ChatGPT is this?

It sounds convincing on the surface. While there might be something fundamentally wrong with the claim, I don't know what it is. Can you elaborate?

[StogieGuy7]

A 1987 study looked at 10 cases from California where inside shoulders were removed to make space for extra lanes (generally in combination with narrowing the existing lanes from 12 feet to 11) and found no increases in accident rates and sometimes a *decrease* due to impact of reduced congestion.

And having no inside shoulders is the norm in most of the world.

There are very few scenarios where a disabled vehicle can make it to the inside shoulder but can't get over to the outside. And as cars become more reliable and electronic safety aids reduce rear-end collisions, the need for breakdown lanes in general is falling. So perhaps it's time to consider getting rid of inside shoulders.

Well, our interstates are the among the best designed freeways on Earth - something I notice when traveling overseas. The upkeep may be inconsistent and the drivers that use them are often crappy, but the design (with two shoulders) tends to be very safe and generous. If you lose the shoulder, breakdowns have nowhere to go and end up blocking the lane....clearly, a safety hazard. Not to mention that you lose the use of that lane and traffic immediately ties up due to the bottleneck (and yes it would be worse than simply having one less lane).  And there are more scenarios than you think where a disabled vehicle cannot make it from the left lane to the right shoulder. It happens all the time.  Sometimes, it's just bad driving; but that's going to exist in your shoulderless freeway too.

[Scott5114]

I personally like not having a Jersey barrier up my ass when I'm in the left lane, thanks.

[roadman65]

I personally like not having a Jersey barrier up my ass when I'm in the left lane, thanks.

Then stay off I-70 between New Stanton and Washington, PA; I-78 from Frystown to Allentown; and all of I-83 in PA.

[Max Rockatansky]

I personally like not having a Jersey barrier up my ass when I'm in the left lane, thanks.

You mean to say that you don't enjoy the sensation of being in a Mario Kart game at 80 MPH but with actual consequences coming from impacting the barrier?

[Bruce]

A good shoulder is worth far more than an extra lane to squeeze more traffic. Collisions blocking actual lanes are a huge issue.

[J N Winkler]

I don't know that we need inside shoulders, but especially when they are of the same 12-foot width as a traffic lane, they have non-negligible incident management benefits.  PennDOT is able to maintain six lanes of traffic while rebuilding the I-95 bridge in Philadelphia only because that segment has full 12-foot shoulders both left and right.

Interstate minimum standards call for left shoulders just four feet wide, so there is not much to be gained from eliminating them in any case.

[jeffandnicole]

A good shoulder is worth far more than an extra lane to squeeze more traffic. Collisions blocking actual lanes are a huge issue.

Which is something completely missed in this study. It focuses solely on the number of crashes, not the length of time to return traffic to a normal state after the incident and the lengths of congestion, and if congestion forced traffic to utilize other options, creating or adding to congestion elsewhere.

[Max Rockatansky]

A good shoulder is worth far more than an extra lane to squeeze more traffic. Collisions blocking actual lanes are a huge issue.

I seem to recall that being a common problem on I-5 north of Seattle. 

[Big John]

Another benefit of a wider inside shoulr is if there is a high concrete barrier, if won't block the view of the left lane continuation on a horizontal curve.

[Hot Rod Hootenanny]

A 1987 study looked at 10 cases from California where inside shoulders were removed to make space for extra lanes (generally in combination with narrowing the existing lanes from 12 feet to 11) and found no increases in accident rates and sometimes a *decrease* due to impact of reduced congestion.

And having no inside shoulders is the norm in most of the world.

There are very few scenarios where a disabled vehicle can make it to the inside shoulder but can't get over to the outside. And as cars become more reliable and electronic safety aids reduce rear-end collisions, the need for breakdown lanes in general is falling. So perhaps it's time to consider getting rid of inside shoulders.

You've never had the experience of your vehicle breaking down in the inner lane of a freeway before, have you.

[JREwing78]

I definitely have. Would not recommend. Of course, it was the tire facing 80+ mph traffic that had to blow out on me.

[Urban Prairie Schooner]

The inner shoulder comes in very handy for emergency vehicles on I-12 in BR during evening rush hour.  So benefit > costs.

[Bruce]

A good shoulder is worth far more than an extra lane to squeeze more traffic. Collisions blocking actual lanes are a huge issue.

I seem to recall that being a common problem on I-5 north of Seattle. 

We've got a special WSDOT crew (the Incident Response Team) that moves cars out of the lanes and it isn't as big of a problem nowadays. A lot of sections now have widened shoulders (one stretch also doubles as bus lanes during peak periods), but the lane expansion north of Everett has shaved away the shoulders and caused a lot of issues. Drivers refuse to adjust to the new conditions and there's near-daily collisions blocking the lanes.

[amroad17]

Yes. https://goo.gl/maps/7MGQnfCJk4cvdfs78

[Occidental Tourist]

What in the ChatGPT is this?

It sounds convincing on the surface. While there might be something fundamentally wrong with the claim, I don't know what it is. Can you elaborate?

The reduced accidents conclusion by the study's authors is a bit of a misnomer because it's tied to measuring increased freeway capacity. And I'd also argue the study doesn't support what OP was contending about safety.  First, the study's conclusions are based just on the accidents per million vehicle miles travelled observed with and without the "full-sized" inner shoulder.  The study compared the accident rates on the same segments of freeway at two different points in time, first with a wide inner shoulder and then later when that shoulder was reduced in size. The number of accidents was then divided into the total number of millions of vehicle miles traveled in that segment during the study periods.  The millions of vehicle miles number was calculated using segment length, average daily traffic counts, and total number of lanes.

But the process of removing the shoulder added a lane to each of those segments, increasing capacity. So you're going to get more vehicles traveling through that segment during the "no shoulder" study period than you would before the shoulder is removed.  One would expect there to be a drop in accidents per million vehicle miles traveled on a fixed segment of roadway when you increase its capacity by either 25 percent (going from 4 lanes to 5 in the segment) or 33 percent (going from 3 to 4). 

Think of it this way: Let's say you have a one lane road.  Along that road in an average day, there's one accident per day and 10 cars travel it.  Then you widen that road to ten lanes. Now you get 100 cars that travel it per day, but you get 7 accidents on average. Is that road safer than it was before? It depends on how you're measuring "safety." Based on total number of vehicle trips, it is safer (0.07 accidents per vehicle trip on the 10 lane road versus 0.10 accidents per vehicle trip on the one lane road). But based on total accidents per segment, it's not (7 accidents versus 1).

I'd argue that to conclude that the removal of the inner shoulder didn't significantly contribute to an increase in accidents, the data would have to show a corresponding statistical reduction in accidents per million vehicle miles reflecting the gained million vehicles miles of throughput achieved on that higher capacity segment during the study period.  But the study's data do not uniformly show that much of a decrease. Only segments 4, 5, and 9 show a decrease commensurate with the additional capacity. The other seven show less of a decrease that that, and, in one instance, show an increase.



The second issue is that, although the study authors somewhat downplay the data, there seems to a statistically significant increase in accident severity along some of the segments with no inner shoulders. If the purpose of the study was just to see if you can erase the inner shoulder to add another travel lane and get more traffic through without causing too many more accidents, then the study sort of supports that (although the authors bury until the end that the data show that the accident reduction wasn't statistically significant).



But if I understand OP's premise, it is that you can remove inner shoulders without making a highway less safe. To reach that conclusion in a meaningful way based on the study, you don't just have to have to show that accident rates did not increase for the segment, but that accident severity rates also did not increase as well.  From my reading of the data, quite a few of the segments show statistically significant increases in "injury" and "fatal" accidents on the no shoulder versions of the studied freeway segments.



Third, this is 50 year-old data, so it does not take into account what improvements in vehicle technology would have on increasing or decreasing safety when a shoulder is decreased in size.  The study was done on vehicles that were generally lighter and slower than today's vehicles as well as lacked anti-lock brakes and other modern car features that might introduce variables as to whether an accident occurs or is avoided, and its severity, in a "left lane, no shoulder" collision situation.

That's not to say this study's data can't be considered. But given what the data show, I don't think it supports the conclusion that freeways without inner shoulders are just as safe as those with shoulders.  To draw that conclusion, you would have needed to see both the accidents per million vehicle miles decrease in direct proportion to the increased throughput along the segment AND no statistically significant increase in severity of accidents.  This study's data shows neither. At best, it supports the conclusions that if you remove a shoulder to add capacity, the accident rate along that segment won't greatly increase, and that on segments with high average daily traffic volumes, the accident rate will increase less than on segments with lower ADT.

[machias]

Judging by the number of broken down vehicles I see on the inside shoulder along I-10 between Tucson and Phoenix (because Arizona folks really like the left lane), we definitely need an inside shoulder.

[ran4sh]

Yes. This isn't really a valid question.

[kernals12]

The inner shoulder comes in very handy for emergency vehicles on I-12 in BR during evening rush hour.  So benefit > costs.

Why couldn't they use the outside shoulder?

[DriverDave]

Definitely on wide freeways where a vehicle that lost power may not be able to make it over to the right in time. Inner shoulders should be wider than outer ones as they are next to lanes with generally faster traffic, which is more dangerous.

[webny99]

There are very few scenarios where a disabled vehicle can make it to the inside shoulder but can't get over to the outside. And as cars become more reliable and electronic safety aids reduce rear-end collisions, the need for breakdown lanes in general is falling. So perhaps it's time to consider getting rid of inside shoulders.

Have you ever driven considerable distance in the inside lane of the PA Turnpike? It's very tedious, and will make you rethink the value of wider shoulders. The presence of shoulders improves the overall driving experience, not just the safety of the roadway.

[vdeane]

There are very few scenarios where a disabled vehicle can make it to the inside shoulder but can't get over to the outside. And as cars become more reliable and electronic safety aids reduce rear-end collisions, the need for breakdown lanes in general is falling. So perhaps it's time to consider getting rid of inside shoulders.

Have you ever driven considerable distance in the inside lane of the PA Turnpike? It's very tedious, and will make you rethink the value of wider shoulders. The presence of shoulders improves the overall driving experience, not just the safety of the roadway.

Now I'm wondering if the inside shoulders on US freeways is a contributing factor in our lack of lane discipline here.

[triplemultiplex]

The inner shoulder comes in very handy for emergency vehicles on I-12 in BR during evening rush hour.  So benefit > costs.

Why couldn't they use the outside shoulder?

Because most drivers are idiots and will panic in case of a mechanical problem and stop at the closest available spot rather than limp a bit further to find a safer spot to stop.  Especially if there's any decent amount of traffic.  If they're in a left lane, they'll use the left shoulder rather than try and merge their way over to the right shoulder where it's wider.  They call it the "breakdown lane" for a reason.