The article is very high level, looking at root causes and (correctly) pointing out the flawed incentives in how America builds infrastructure. Here are some lower level observations:

I've been watching construction and similar projects at a government site lately. It seems that there are always more people watching than working. But that's not too bad, as often construction projects will languish with no one working on them for months (meanwhile, the site gets dirty due to rain water filling the site, etc) due to red tape requiring certain people to do certain parts of the job (such as cost exceeding a certain amount and thus the work needs to be put out to bid instead of using the on site contractor).

The exception is if it's something simple, like erecting a temporary meeting tent rented by some company. In that case, it seems like EVERYONE is busy and is doing something. Stuff gets done rather quickly. Maybe they have a lot more practice? I bet they also do multiple types of job, i.e. a guy doesn't just tie the tent down, but also helps unroll the fabric from the truck or moves chairs to the seating area, etc.

Regular road projects seem to be not so bad, more people working than during a building construction. But I noticed another thing: huge machines which must cost almost a million dollars sitting idle with no one using them.

I think the secret might be just to ensure that work starts and finishes on a project as soon as possible, that both people and expensive equipment have work ready for them so they can keep busy constantly until the job is finished. This probably would require a responsive, competitive, well-managed, and very well practiced company to do this. And maybe it'd help if people were cross-trained to do multiple trades, so that after one task is completed, they could immediately go work on another. But if people & machines were simply kept busy constantly on the project, I've gotta think the project could be done for far, FAR less money (and time).

And the author is correct that the way to establish those is via the proper incentives. Constructing a building by contract with a government agency is not a very dynamic environment. Renting a meeting tent, on the other hand, is something very amenable to competition even if the customer happens to be the government.

I'd really like to see some sort of study that simply observed construction projects and crews, on-site, in multiple locations in the world and multiple types of projects. We could get some decent answers.