The clash you find in the field costs 10x the one you find in the model
Every project team has lived the same failure mode: a duct main lands exactly where a transfer beam needs to pass, a sprinkler branch fights a structural brace, or a sanitary line wants the same vertical chase the column already owns — and nobody catches it until the subcontractor calls from the field. Now you are issuing an RFI, holding the trade, eating rework, and absorbing a change order against a schedule that had no float. The root cause is almost never a single bad drawing. It is that structure, mechanical, electrical, plumbing, and fire protection were each modeled to their own logic, in their own files, and never reconciled against one another on a disciplined cadence. The coordination process exists to force that reconciliation to happen on paper, on a schedule you control, before 'issued for construction' (IFC) locks the documents and the cost of being wrong multiplies.
- The expensive clashes are not random — they cluster in tight plenums, shafts, and at transfer/footing conditions
- A clash caught in a coordination meeting costs an hour; the same clash caught in the field costs an RFI, a delay, and a change order
- Coordination is a managed process with owners and dates, not a one-time clash report you run at the end
The end-to-end workflow, placed on the production schedule
Coordination is a loop, not a milestone. It starts the moment the architectural background and reflected ceiling plan (RCP) are stable enough to model against, and it does not close until every discipline signs off ahead of IFC. The sequence below is the spine; on a fast-tracked job the loop simply runs more often and against partial backgrounds. The critical scheduling judgment is where in the DD-to-CD arc the federated clash runs begin — too early and you are chasing design churn; too late and you have no time to resolve hard clashes before the CD deadline.
- 1. Background + RCP issuance — architecture freezes a coordination background and ceiling heights/plenum depths so every consultant models to the same datum
- 2. Discipline modeling — structural frames the system; mechanical, plumbing, fire protection, and electrical model their distribution to real sizes (not single-line placeholders)
- 3. Federation + clash runs — models are combined into one federated model and hard/soft clashes are detected; first meaningful runs land late DD, then intensify through CD
- 4. Coordination meetings — clashes are reviewed live, each assigned a single accountable owner and a resolution path (not 'the team will look at it')
- 5. Resolution — the owning discipline reroutes, resizes, or requests a structural accommodation (sleeve, opening, drop); the architect adjusts ceiling/soffit if clearances demand it
- 6. Re-run — the federated model is rebuilt and clashes are re-detected to confirm the fix did not create a new collision downstream
- 7. Sign-off before IFC — disciplines formally accept the coordinated condition; only then do the construction documents go out for construction
Name the hard clashes — and sequence the services that cause them
Most plenum and shaft conflicts are predictable, which means they are preventable if you assign a service hierarchy up front. The governing principle: rank by what is hardest and most expensive to move, and let the flexible services route around the rigid ones. Large gravity-driven and rigid systems win; flexible, pressurized, small-diameter systems yield. The recurring offenders are worth naming explicitly so the team hunts for them on purpose rather than discovering them by accident.
- Ductwork vs. structural beams/joists — the classic tight-plenum fight; large rectangular duct has little flexibility, so route it first and resolve beam penetrations through coordinated openings, not field-cut holes
- Sprinkler mains vs. structure and primary services — fire protection mains are run early in the hierarchy because relocating heads later disturbs coverage and hydraulic calcs
- Plumbing vs. footings and grade beams — underground sanitary and storm lines that conflict with foundations are among the costliest to miss; coordinate the below-grade model against the structural footing layout, not just the above-ceiling plenum
- Service-priority rule of thumb in a tight plenum: gravity sanitary/storm (fixed slope, no flexibility) → large supply/return duct → sprinkler mains → domestic water and small pressurized pipe → cable tray and conduit (most flexible) → low-voltage last
Anchor the process in the BEP and hold owners accountable
Coordination only works when the rules are written down before modeling starts. The BIM Execution Plan (BEP) is where the team agrees on model origin and shared coordinates, level of development per phase, clash tolerances, the federation cadence, and — most importantly — who owns what. Pair the BEP with a clash matrix that defines which discipline yields to which, and a standing meeting with a tracked issue log where every open clash has a name and a due date next to it. The PM's job is to keep that log shrinking on a curve that hits zero hard clashes before the IFC date, not on the day of it.
- Define in the BEP: shared coordinate system, per-phase model maturity, clash tolerances, federation/run frequency, and the discipline-yield matrix
- Track every clash to a named owner and a date in a single living log — 'unassigned' is the status that kills schedules
- Distinguish hard clashes (physical collisions — must resolve) from soft clashes (clearance/access — judgment calls) so the team spends its hours on what actually stops construction
- Gate IFC on a clean federated re-run and documented discipline sign-off, not on the calendar alone
| Phase | Model maturity | Coordination activity | Clash posture |
|---|---|---|---|
| Schematic Design (SD) | Massing / single-line systems | Establish BEP, shared coordinates, service-priority matrix; reserve plenum and shaft space | Spatial reservation — proactive, not yet detecting |
| Design Development (DD) | Sized systems, real geometry emerging | First meaningful federation and clash runs; resolve major routing conflicts and structural penetrations | Catch the big, expensive clashes while there is room to move |
| Construction Documents (CD) | Fully sized, detailed distribution | Intensive run/meet/resolve/re-run loops; drive open hard clashes toward zero | Tolerance tightens; resolve and document, not discover |
| Pre-IFC sign-off | Coordinated, frozen | Final clean re-run, discipline sign-off, issue resolution log closed | Zero outstanding hard clashes before release |
Frequently asked
When should structural and MEP clash detection actually start?
Begin spatial coordination in SD by reserving plenum and shaft space, but the first meaningful clash runs should land in late DD once systems carry real geometry and sizes. Running detection against single-line placeholders only produces noise; running it too late in CD leaves no time to resolve hard clashes before IFC. Late DD into early CD is the window where you can still move major routing without blowing the schedule.
What is the difference between a hard clash and a soft clash?
A hard clash is a physical collision — two solids occupying the same space, like a duct passing through a beam. It must be resolved. A soft clash is a clearance or access violation — a valve too close to a wall to service, or insufficient room around a panel — where the elements don't physically overlap but the condition is unworkable. Separate the two in your log so the team spends its limited hours resolving collisions that stop construction before debating clearances that are judgment calls.
Which service gets routed first in a tight ceiling plenum?
Rank by rigidity and cost to move. Gravity-driven sanitary and storm piping comes first because it has a fixed slope and almost no flexibility. Large supply and return ductwork follows, then sprinkler mains, then pressurized domestic water and small-diameter pipe, with cable tray, conduit, and low-voltage last because they bend most easily. Agree on this hierarchy in the BEP so the flexible trades route around the rigid systems instead of fighting them.
Why coordinate plumbing against footings and not just the ceiling plenum?
Below-grade conflicts between sanitary or storm lines and foundation footings and grade beams are some of the most expensive clashes to miss, because resolving them in the field can mean relocating runs, deepening foundations, or core-drilling structure. The above-ceiling plenum gets most of the attention, but the underground model deserves an equal clash run against the structural foundation layout before the slab is poured.
What document governs the coordination process?
The BIM Execution Plan (BEP) governs it. The BEP defines the shared coordinate system and model origin, the level of model maturity expected at each phase, clash tolerances, how often models are federated and run, and the discipline-yield matrix that decides who moves when two systems conflict. Without it agreed before modeling starts, teams coordinate to different assumptions and the clash log never converges.