Clash detection that pays for itself

Any coordination tool will produce a clash report in minutes. The trouble is that a raw report on a real project can list thousands of clashes, most of them noise. Clash detection only pays for itself when it's set up to find the problems that matter and turn them into decisions before steel is cut and ductwork is made.

Federate first

Coordination starts with a federated model: each discipline's model (structure, architecture, mechanical, electrical, drainage) brought together in one place, on the same coordinate system. If the models don't share an origin and orientation, every clash test that follows is meaningless. Agreeing that at the start is the cheapest coordination work on the job.

Test with intent

• Run discipline pairs, not everything against everything: steel against ductwork, steel against drainage, and so on.
• Separate hard clashes from clearance checks. A duct touching a beam is a clash; a duct 20mm from a flange may be too, if the insulation, fire protection or access needs the space.
• Group results by location or system, so one misplaced duct run is one issue, not two hundred.
• Filter out the known and the acceptable, so each run shows only what's new.

Turn clashes into decisions

A clash is only resolved when someone decides who moves. Track each issue with an owner and a due date, in a shared format such as BCF so it opens in everyone's software at the right view. Then review the open issues together at a regular coordination meeting: short, visual and decision-focused.

Where steel sits in this

Steel is usually the first trade to be fabricated, so its clashes are the most expensive to find late. Service penetrations through beams, bracing through a riser, a plant platform that fouls a pipe run: all cheap to fix in the model and expensive to fix with a grinder on site. That's why structural coordination belongs before the detailing is issued for fabrication, not after.