Trade capacity and collision forecasting model

What each trade can actually deliver in a week, derived from what they have already delivered for you, laid under the forward start schedule twelve to twenty-six weeks out. See the collision in September instead of paying the premium for it in March.

$3.1M

Reference-scale value

Data sets in the model

8 weeks

Build and deploy

Derived

Evidence grade

The decision this model makes

Your schedule says you are starting forty houses in March. Your framers can actually do thirty. Nobody discovers that until March, when a superintendent starts phoning around and pays whatever it takes to get somebody on site.

Roughly fifteen per cent of starts hit a trade that has nothing left to give that week. Each one costs premium and expedite labour to unstick, and adds slip days that carry real interest on the work in process.

None of that is a capacity problem. It is a visibility problem. The trades have not changed what they can do — you just found out on the wrong day.

The model measures capacity as what each trade has already delivered for you, not what they said on a call. Crew count times proven weekly throughput, read out of twenty-four months of your own task actuals, laid under the start schedule so the collision surfaces while it is still a phone call.

Want more detail?

The rest of this model — the data sets it runs on, the cause split, the working behind the reference figure and a sample output — goes out by email. Two fields, nothing else.