Thursday, March 19, 2015

Why Does Highway Construction Take So Long?

I live near the connection of two major interstates, 880 and 280, and frequently take it. Work has been going on to improve the connection for quite a long time; the project was approved in 2011 and under construction as least as early as 2013. Most of the time the resulting constriction is only a minor inconvenience, but during rush hour it can result in a significant delay in getting from one highway to the other.

It seems as though what they are doing should take no more than a few weeks. Four possible explanations of why it is instead taking years occur to me:

1. I may be missing important elements of the process that make it much more time consuming than I would expect. Perhaps there are many steps that have to be taken in sequence. Perhaps some part of the process, such as the drying and curing of concrete, takes much longer than I realize.

2. Perhaps doing it slowly is a little easier than doing it fast and the people making the relevant decisions, the construction firm and the California Department of Transport that employs them, have little or no incentive to take account of the inconvenience to drivers of having a major intersection under construction.

3. Perhaps the construction company has persuaded friends in the Department to agree to contract terms that pay by time rather than by project, making it in their interest to stretch out the process.

4. Perhaps the government officials and the politicians above them believe that longer is better, that the political benefits of keeping drivers aware of their highway dollars at work more than outweigh the political cost of continued delays.

There are probably other possible explanations that have not occurred to me. Does anyone know what the right explanation is? Any evidence on how long such projects take in other states or other countries?


Unknown said...

I sure you could google an answer but Michigan DOT says 7 years for construction of a new highway.

Highways however are well studied by project managers and used as case studies because they are generally simple projects that make good student examples for civil/mechanical engineering courses. In general it's about having manpower and resources when and where you want them. If you have an abundance of both the schedule is fast but an abundance is expensive. Instead most construction companies run very lean because they are in a commodity business. You never get a perfect man/machine loading so it's more cost effective to have enough projects that all of the resources are 100% allocated even if this means projects go long pauses without being worked on. especially as roads are durable goods. A good project manager can minimize these delays on one project but often by prioritizing resources from another.

Additionally states/towns sometimes benefit from deferred payments over the length of a project and the contractors have the ability to halt work on multiple projects if money is not received promptly. And then the state mandates certain things like inspections at fixed stages to insure that the contractors in compliance with code... and inspectors always seem to have more things to inspect than manpower to do so.

In short your inconvenience is not costing the contractors or the govt any money so neither has an incentive to change from their mutually negotiated price-point.

...and yes there are engineering based technical issues as well like cure-times and working temperatures but they tend to be less important than the volume of cement or asphault that can be trucked in and volume distribution is pure project management.

Joeleee said...

In my understanding the above poster is pretty much correct. The only thing I would add, is that it's not just other projects that add to the time, but also the geographical disparity of a long highway project. If you had extra men and machines,you could almost always add more, but over say 100km of road, you would only generally have one crew, because the capital costs of maintaining a few sets of machines is too high.

In China they build road and rail absurdly quickly, and my understanding is because they have a whole lot of idle capital goods and labour they can throw at it (as well as fewer safety regulations do doubt)

Anonymous said...

This underscores the importance of privatizing roads.

geogavino said...

Some overlooked technical issues of roadwork (and earthwork in general) often include the need to export unsuitable soil and import suitable soil, aerate and dry out soil, correct deficient areas by reworking and recompacting or installing subdrains or seepage drains, etc., controlling erosion and stormwater - rain can cause havoc in earthwork, setting projects back, and snow stops it completely. If it involves concrete work, esp. bridges, overpasses, etc., then you have to consider iron work, various methods of reinforcement, lengthy curing times before loads are applied to build the next section, etc.

Unknown said...

I'd be cautious with the plan to privatize roads. The other aspect which I didn't mention is that I'm fairly certain without proof that the mob has deep in-roads with the larger construction companies. I suspect that if you privatized them not only would service decrease to the lowest level possible in a commodity market... but you would pretty much hand whatever revenue generation the roads created to the mob as a side effect.

Anonymous said...

To Sean Powell's answer, I would add:
(a) funding sources are often spread out over several years
(b) accelerating the schedule often means more acute inconvenience during construction, even if the duration is shorter
(c) union rules make it more expensive to do night work or overtime work, so it's often most cost effective to work only during the day
(d) you sometimes find things in the way that you didn't know were there, like utilities, that force you to stop work and redesign part of the project

On item (b) note the risk incentive structure for an agency like Caltrans. You're not going to get fired for causing 3 years of minor inconvenience, but you might get fired for causing a 20 mile backup in San Gorgonio Pass like they did a couple years ago.

Public works contractors are paid by the unit, eg how much concrete they have placed, not by time.

Tibor said...

I don't know how this works in the US, but two main problems with the speed of highway construction in the Czech republic seem to be

a) Buying the land from the owners.

Often a road is planned and the construction started before all land is securely bought by the state. Then they have to stop because the land further down the road does not belong to the state, and the owner reasonably wants to drive the price up. There was one extreme case of a farmer who had a stripe of land through which a highway was supposed to go and it would take her about 10 years for her to agree on a buying price. Of course, this is the fault of the government, because its bargaining position would be a lot better if the highway were not already under construction and with no other alternate route.

b) Ecological activists.

They can block construction often for years by protesting against the road for ecological reasons...although I am not sure whether this can be done when the construction is already in progress or whether they have to raise their objections before everything is started.

Then probably the reasons you mentioned also apply. Government funded construction always take much longer to build than the private funded.

jimbino said...

Concrete cures to about 98% of its ultimate strength in a month.

Fred Mangels said...

Tibor wrote, "...although I am not sure whether this can be done when the construction is already in progress or whether they have to raise their objections before everything is started."

Here in California, yes they can. We have two freeway projects in Northern CA that are being held up by environmentalists- one, after the project was well under way. In that particular one- the Willits Bypass- the vast majority seem in favor of it. All it takes is a few environmentalists to file a lawsuit and the work is often stopped.

Same with a highway realignment further north. Funny thing is, they're not really doing much in that project, yet the usual suspects filed suit. The project had to go back to square one, for the most part, because CalTrans- the agency doing the environmental study- mislabeled a couple small trees that would be affected, among other things.

GregS said...

I recall there was construction on a bridge between Springfield and Chicago, Illinois that backed up traffic for miles, and construction lagged on for months. It took about an hour to travel five miles for traffic going in either direction. I remember being annoyed that nobody was working on the bridge on either of the days that I passed it. A fun exercise is to look at traffic patterns on these roads. IDOT collects this data, and with a pretty simple Google search I could find useful maps that showed how many travelers passed the bridge each day. You can do a back-of-the-envelope calculation, assuming each driver’s time is worth, say, $10 an hour (or more or less if you think some other figure is appropriate). I don’t remember exactly how the calculation went, but I concluded that the inconvenienced drivers should be willing to hire armies of laborers to finish the bridge sooner. Sean Powell’s answers all seem perfectly reasonable. But any answer should be checked against the kind of calculation described above. It might rule out some answers as implausible. I’m skeptical of answers that suggest the true cost of the project is cheaper if it drags on for a few months; those “savings” are probably swamped by the social cost of the inconvenienced passengers. I strongly suspect the true answer has to do with the construction firms’ (and the government’s) being insulated from the cost of delay. In other words, 2) is a good answer, probably not 1). Agnostic on 3) or 4).

Norm said...

There was a contractor with a cowboy persona whose company worked on the SF bay bridge and earlier on an earthquake damaged freeway in southern California These projects were completed incredibly fast I never heard of a comparison of his emergency price versus a typical slow price (useless press).
If the contracting agency valued public time at all they could include a small bonus for every day a project is finished before a fixed long away date.

Salli Weston, said...

I work for MNDOT. The issues have been covered pretty well. I know we calculate man hours lost in traffic and about 5 years ago moved to totally closing roads to more quickly finish projects when there are viable alternate routes. I can cost the state more but overall costs are cheaper. We also do more and more total closures on weekends and evenings when we cannot close the road. MNDOT rarely does incentives for early finish but when we do there are also fines for late finish.
Also there are large numbers of utilities of all types that parallel the roads and they need to be addressed.

J Scheppers said...

I am frequently involved in street and highway construction and have these observations:

1. Much of the work is underground and not seen in the finished product. The list of underground elements include: drainage, water, sanitary sewer, power, and 4 or 5 telecom companies.

2. In most states these underground utilities are independent companies and their permits to be in the road real estate don't always mandate fast movement for conflicts.

3. Contracts usually do have time requirements, but are costly to enforce and once a lapse occurs on one contract it is difficult to enforce the contract provisions on any contract. It is also difficult to write specific time incentives and the more rigorous the more costly.

4. There are some contractors who make their money by being fast. They plan and get in and out. There are other contractors who make their money applying cheaper less timely resources. The fast ones you likely don't remember the slow ones you cannot forget.

5. Permitting leaves you at the mercy of the regulator who does not have a deadline. Projects have been stopped because a spider has been found and five years later they are still discussing if the spider was endangered or a mutation. By the way the single spider they did find was promptly killed for analysis.

6. Competition. The person in the UK that built a private toll road detour to compete against the free 10 mile longer detour inspired the government to finish their work 20% faster.

Peter McCluskey said...

A section of 580 in Oakland was replaced 26 days after it was destroyed in April 2007. That speed seems partly due to the contractor being paid mainly via rewards for early completion.

Will McLean said...

I think public choice is part of the answer. let's suppose that a state has fifty districts, and twenty have major state-funded repair projects underway. Any of those districts is a concentrated interest, but if they want to shift the allocation of funds they are in conflict with the other 19. who are equally interested in getting their projects completed. To increase funding, they need to get votes from some of the other districts, where the constituents see no immediate benefit from increased spending. And some of the inconvenienced drivers will be from out of state, with no voice at all.

RP Long said...

I have heard - though have never personally verified - that some large-scale road construction projects run out of money and must sit idle for a while until the next batch of money comes in. I put it out there as another possibility.

stan said...

Sometimes it is a function of govt simply getting a project started whenever approval can be secured -- even knowing that there isn't the money to get it completed for a long time. Once started, it will, eventually, get done.