[Torkel]

Yes, please! To me it feels like 90% of road works I pass actually has nobody there working at all. Some go on for years it seems.

Another option is what they do in Japan - just get 10X the number of people on it, with all the tools for the entire job and then do it all overnight.

Also, nice to see good old "dude with shovel" in there, a tool that would have looked the same back in the 19th century.

[mike_d]

> To me it feels like 90% of road works I pass actually has nobody there working at all. Some go on for years it seems.

This is for simple road resurfacing, which is usually done in a day or two anyway.

There are a lot of factors as to why road construction takes a long time, but the biggest reason is safety. They aren't completely closing the road in most cases so while a large section might be coned off, they are moving the worker protection (vehicle barriers) around from section to section.

Lots of specialized expensive equipment is involved where you might be waiting a week just for the thing you need to become available. Same with specialized workers and sub-contractors. Add in time you might just need to wait for concrete to harden to the point its safe to move a big piece of equipment onto it for the next phase.

Construction is a scheduling nightmare. You can throw money at the problem, but it is in the end tax dollars and rarely is there a good reason.

[utensil4778]

People say this a lot, but I really don't think it's a very satisfying explanation for having miles of one lane of a highway barricaded for weeks or months without any workers or equipment in sight, ever.

Everyone can name a stretch of highway they've seen treated this way, and this explanation just doesn't cover it.

[spenczar5]

This is a form of sampling bias. The quick projects are only there briefly, so you're less likely to see them. Everyone sees a slow project that has some scheduling disaster occur, like a strike or a supply shortage or a government failure of some kind.

You don't know just from driving by whether 99.9% of projects are quick and efficient.

[anonymousab]

Or they might live in a place with an enriched and proliferate construction mafia, such as Quebec, and so having multiple roadway construction projects ongoing with no actual work done for months would actually be a frequent, intentional occurrence.

[kelnos]

> without any workers or equipment in sight, ever. [...] Everyone can name a stretch of highway they've seen treated this way

I think there's an inherent bias here. Each individual is not seeing that section of road 24/7. At most they see it for a few minutes (or perhaps a little longer, if traffic slows enough), maybe once or at most a few times per day. That's a very small fraction of the day, and work could be occurring during other parts of the day, possibly even at night. Safety might require that they do the work when there are few cars driving by, but the nature of the work might be such that they can't unblock the lanes during the day.

I get that this isn't satisfying, but often reality isn't. And I have no doubt that there are plenty of mismanaged projects around the world where there are lots of delays, and long stretches of time when nothing is getting done. But I think it's incorrect to believe that nothing is getting done just because a few people you know drive by the area a few times a day and don't see anyone working.

[jcrawfordor]

Very common on US freeways for work to be performed at night even though it's barricaded during the day. The barricades are to keep traffic off of work in progress, actual work may require additional lane closures, traffic disruptions as trucks enter and exit, and safety scales with traffic volumes. Keep in mind, for example, that the permanent closure area is not typically large enough to stage equipment. That means that equipment and supplies need to be moved in and out of a laydown yard during work, which is very dangerous during normal traffic volume.

It also tends to be the case that price is reduced by scheduling work across multiple contractors with independent scheduling... so the cost savings come at the expense of idle periods while waiting for the next contractor to be available. Not a totally unaddressable problem but ultimately fast and cheap are, as usual, opposing requirements. Funding politics can also play a role here, very common that larger projects don't have all of their funding available at once, so they may sit idle while waiting for the next set of funds.

[kbrosnan]

Roadbed settling and concrete curing are two long term inactive construction projects.

Practical Engineering videos on road construction and public works https://www.youtube.com/watch?v=PIK6I6Q58Ec https://www.youtube.com/watch?v=22W5tRWbUVI

[mschuster91]

> People say this a lot, but I really don't think it's a very satisfying explanation for having miles of one lane of a highway barricaded for weeks or months without any workers or equipment in sight, ever.

The problem is that both the workers and the equipment are short in supply.

The cause of that is the lack of continuity in politics. Construction machines and staff training can easily reach dozens of millions of euros in cost - particularly when it's rail related. It would be financial suicide for any company outside of extremely large conglomerates to take on that risk without politics providing the guarantee of at least 20 years worth of projects to recoup that investment. Instead, the US is down to "we can barely plan for the next fiscal year" timeframe, and Europe to "we can barely plan 5 years until the next EU fiscal cycle, add national election cycles and you're down to 1-2 years as well". On top of that come government accounting clusterfucks - basically, the norm is that it is very difficult to transfer budgets from one year to the next, and when you don't use all your budget for whatever reason, next year's budget will be cut back.

[bagels]

Sometimes the work is done at night. Sometimes, they just tear up the road and do nothing for a year.

[jefurii]

Stretches of freeway in Bakersfield and Santa Clarita have been under construction for decades.

[akira2501]

You're looking through the lens backwards.

For any one stretch of road what percentage of the time is it closed in this way? Or in other words, what is the SLA of the road?

[Vaslo]

If this is the case then move the cones barrels where you haven’t done any work yet. In the states you often see miles cordoned off by barrels with no work has been done yet, well beyond where any person would say there is a safety issue.

[fbdab103]

I once heard an unsubstantiated claim that federal funding was dependent upon length of the road work. Over-provisioning blocker barrels was a loophole to extend the distance and receive more funding.

[groestl]

> it feels like 90% of road works I pass actually has nobody there working at all. Some go on for years it seems.

That's because the company that wins the tender knows they're competing on the lowest price, and nobody really cares for time to completion. So they're keeping the construction site as a fallback for other construction sites, that are more important to them and have a higher price tag.

It's the equivalent to preemptive pricing for VMs.

[jq-r]

Exactly! And it doesn't even have to be a fallback. The same company could have won (or "won" if you know what I mean) many/majority of those tenders and are intentionally spread thin. Because once you win it, you leave a token crew there, and move to win another one, leave token crew again and again. So they're just doing everything all at once very slowly while being paid a lot.

Also this doesn't even have to be a public works type of job. They do this kind of stuff for private works like house construction. 30 people show up and start working, and after couple of days you have 2 guys working for weeks while others are somewhere else.

I was very surprised when I saw repaving of couple of street intersections in Chicago. Those guys did it like they were on a competition. Really fast and really nice. They were done in couple of days. I couldn't believe it. In my part of the world, the same job would take months, and I'm not even kidding.

[groestl]

All of that is very true, including the bit about private works.

[hiatus]

Are there no deadlines in these contracts or penalties where the company is charged X/day until completion?

[floxy]

There are/can be incentives/penalties to complete road construction on time. Although I'm sure it varies with jurisdiction.

https://www.fhwa.dot.gov/construction/contracts/t508010.cfm

https://transweb.sjsu.edu/sites/default/files/2908-highway-t...

[bombcar]

Many of the cases that I know of where a performance bonus existed (like $million for every day the estimate is beaten) they finished ahead of schedule and under budget.

[soco]

Ahh and I thought the forever-construction sites were a Swiss thing. Every 20 km or so you'd have 1-2 km of road construction signage where nothing happens for a couple of years. At some point it just disappears, to come up again a few km down the road. Nobody here to explain us the phenomenon?

[aidenn0]

There's an old joke that works almost everywhere:

Have you heard about the new invention that will allow <insert name of local department for handling road work> to lay off 90% of its workers?

A shovel that stands up by itself.

[blowski]

I had an eccentric Maths teacher at school who used to stare out of the window and make odd claims. One was that he was running a roadworks business using telekinetic powers, and all the people in hi-vis jackets were merely scarecrows.

[JaggedJax]

One of the largest time components can be due to soil compaction. This period can look like no work is being done, but is a vital step. That's not to say that sometimes no work is being done because of scheduling, equipment, funds, or any number of other reasons.

This video/article does a good basic job explaining why it can take so long: https://practical.engineering/blog/2020/6/1/why-does-road-co...

edit: I think someone linked this down below too.

[floxy]

>Some go on for years it seems.

Isn't some of that due to earth settling that you can't really speed up? I remember a fun fact from many years ago where I was growing up. They built a new interstate bridge/overpass, and it wasn't too long after that they noticed it starting to sink. Well it turns out they had drilled sample cores down to like 100 feet, and had good results, but after the sinking, they drilled new cores, and there was a layer of sandstone or something at like 105 feet. I don't really remember how it all got sorted out, but it was under construction for a long time.

[swores]

I wonder (and maybe someone here will know) whether that means they were foolishly cutting corners and should have sampled further than 100ft down, or if what they did was considered best practise and they were just unlucky to find themselves in a freak situation?

[bombcar]

Likely the second - they usually base drilling depth on the known area and a wide margin, so everyone probably thought 75 feet was fine, do a hundred to be safe.

[swores]

Thanks

[softgrow]

And in Japan it seems that every time you drive somewhere after 10pm there are roadworks everywhere (on the non-toll roads). Very pretty with kaleidoscopic lights but it takes just as long late at night to move around as earlier on.

[avar]

    > Also, nice to see good old
    > "dude with shovel" in there,
    > a tool that would have looked
    > the same back in the 19th 
    > century.

The guy using a flat broom at around 1:12 in the video has him beat.

While the I was surprised at how recently the flat broom was invented (the very tail end of the 18th century), the job he's doing with a broom goes back to antiquity.

[froddd]

In the UK roadwork sites usually remain much longer than it takes for the result to start deteriorating once they have finished.

[colechristensen]

>Also, nice to see good old "dude with shovel" in there, a tool that would have looked the same back in the 19th century.

I grew up occasionally using tools of the shovel-variety which sometimes were from the 19th century, and when not, not far off.

Every time I drive through particularly ornery construction I start to fantasize about writing software to optimize the critical path and driver disruption.

I also am left wondering having seen several iterations of "improvement" projects that were followed a number of years later on the same stretch of road with another project... if anybody analyzes the whole effect of the project and if the designed improvement made up for the considerable interruption executing it caused. Like on a "net-positive" basis, would folks have been better off if nothing at all was done and the project just skipped.

[stormdennis]

Where I live it goes crazy with road works when it's coming to the end of the year with the council trying to spend the unused budget rather than lose it.

[fennecfoxy]

A shovel is a shovel, a glass is a glass.

[eterevsky]

The hourly wages in Switzerland are at least 2x those in Japan, and Swiss people generally don't like working outside normal working hours, so getting 10x workers at night is not very feasible.

[michaelt]

> getting 10x workers at night is not very feasible.

You can get as many workers as you need, for a price.

For example, got a 2 day window at Christmas to perform a rail upgrade? And overrunning on time costs thousands of pounds per minute? Just show up with a large workforce and lots of spare parts and equipment.

The downside is you end up paying for things like having a couple of diesel mechanics on site just in case a backhoe breaks down. Maybe none of your backhoes broke down, and you paid for 2 guys times 48 hours times 3x their normal hourly rate, just to sit in their truck.

Of course, overnight work does tend to get noise complaints from locals - no amount of workers will solve that!

[lostlogin]

The same breakdowns happen with slow roadworks. It’s just that the cost of taking 3 months longer isn’t factored in. Me sitting in a jam is an externalised cost that they don’t care about.

[TylerE]

In my Southern state they prefer to work at night because of how hot it is during the day. Something that could reasonably be called 'Summer' runs from about March to September here. More in some years. It was in the 70s New Years Day here this year.