County plan review, answered

Two different things get called one thing. An engineering comment needs judgement, a second look at the design, or a visit to the site. The other kind is something a reviewer sees without opening a calculator, and it is what this product is built to catch, because the drawing file can settle it on its own.

The kind that are quick to fix while the set is still yours and expensive once somebody has written them down. That is the whole reason this exists.

It depends on the county and on the queue. What you control is how many rounds it takes, and the fastest route to a signature is a set with nothing left in it for a reviewer to write down.

Many jurisdictions publish exactly what a submitted file has to be: the format, the sheet size, the border, the seal. The particulars differ from one place to the next, and because they are published, your file can be measured against the ones your own names before anybody uploads it.

It is among the most common comments and among the cheapest to fix before it is written down. It arrives three ways: a sheet the index lists that is not in the set, a sheet in the set the index never lists, and an index title that disagrees with the title block on the sheet itself. Each is a separate item here, and each one names the sheet it is about.

Most of it is arithmetic that stopped agreeing with itself. A structure that discharges higher than it receives, or a rim at or below one of its own pipe inverts. The same structure listed with different elevations in two places. A stated slope that its own inverts and run length contradict, or a pipe that gets smaller as it runs downstream. Every one of them is settled from numbers the plans already state, without opening the design file.

It reads the whole set, the drawings as well as the words, and reports what would draw a comment. Every item it hands back points at the page it came from, so you can settle it in seconds. What it can see grows with every release.

Open the sample reportHow it decides

Yes, on every run, and it keeps getting better at it. AI is one of several readers working over the same set: the words on each sheet, the geometry drawn on it, and AI where the others cannot settle it, including sheets nothing else could number. Everything any of them reports is held against what the page prints, and every item points at the page it came from.

How it decides

By hand it is a pass with a printed index: tick every sheet against what is actually bound in, read the general notes again, add up every quantity table again, and check every storm structure rim against its own inverts. Set-Check runs that pass in minutes and hands back a list you review one item at a time. It is a second pair of eyes, not a replacement for yours.

See a recorded run

Minutes on a full set, and you can run it again after every plot.

No, and the difference is worth being exact about. A plan check is the jurisdiction's own review, by the published definition. This is a quality control pass the submitting engineer runs on their own set beforehand, before anyone at the county sees it. Nothing here predicts approval and a clean result is not a prediction of one.

Read the methodology

Every item points at the page and the place on it, with the figures behind it, and the whole set comes back marked up. You settle each one with the drawing open in front of you, and what the run could not read is said plainly rather than left to look clean.

Open the sample report