What Is Clash Detection - and What It Doesn't Catch

How clash detection works and what it catches well

Clash detection works by comparing modeled objects in three-dimensional space. Each wall, duct, pipe, beam, conduit rack, sprinkler main, ceiling, door, or piece of equipment has a position and size in the model. Clash software evaluates those objects against one another and flags conditions that violate the rules set by the project team.

The most familiar category is the hard clash. A hard clash occurs when two physical elements occupy the same space. For example, a supply duct may pass through a steel beam, a sanitary pipe may run through a shear wall, or a sprinkler main may intersect a cable tray. These are direct physical conflicts. If built as modeled, something has to move, be rerouted, be sleeved, or be redesigned.

Clash detection also catches clearance clashes, sometimes called soft clashes. These are not necessarily physical overlaps. Instead, they occur when there is not enough space around an element for access, maintenance, code-required working space, or installation. An air handling unit may have too little service clearance on one side. An electrical panel may be modeled with piping in its required working space. A valve may be located above a hard ceiling with no access panel nearby.

In 4D applications, clash detection can also support workflow or sequence checks. These look at construction activities over time, not only final geometry. A crane path may conflict with stored materials, a temporary scaffold may block access needed by another trade, or a prefabricated rack may be scheduled for installation before the supporting structure is ready.

For fully modeled projects, this is genuinely valuable. It gives design and construction teams a shared spatial view of problems before crews are standing in the field with material already fabricated.

What clash detection requires to work

Clash detection depends on the quality of the model being checked. To work properly, every relevant discipline needs a sufficiently detailed, accurate, and current 3D model. The architectural model has to reflect the current layout. The structural model has to include the beams, slabs, openings, and embeds that affect routing. The MEP models need enough detail to represent actual duct sizes, pipe elevations, equipment footprints, hanger zones, access requirements, and service clearances.

When that prerequisite is met, clash detection can be precise. When it is not met, the results become incomplete. A partial model cannot identify a conflict with an element that was never modeled. A coordination model that lags behind the latest issued drawing set may show a clash as resolved even though the drawings still contain the problem, or it may flag an issue that has already been fixed in the current documents.

Many projects also remain hybrid. One discipline may work deeply in BIM while another still issues primarily 2D drawings. Specialty consultants, low-voltage designers, food service vendors, fire protection contractors, elevator vendors, and equipment suppliers may provide information in schedules, PDFs, submittals, or diagrams rather than complete 3D models. Specifications and many schedules are often not modeled at all. They define requirements that do not exist as geometry in the clash model.

In those cases, clash detection is doing exactly what it was built to do: checking modeled data. The limitation is that not all construction information lives inside the model.

What clash detection cannot catch even with a perfect model

Even with a complete and well-coordinated model, clash detection is still primarily a geometric and spatial test. It cannot evaluate every obligation in a construction document set, because many obligations are not geometry.

Specification-to-drawing mismatches are a common example. A model may show a door, wall assembly, fixture, or piece of equipment in the right location, but the written specification may call for a different fire rating, material, finish, acoustic performance, manufacturer requirement, or installation standard. Since specifications are not usually modeled as clashable objects, a geometric clash test cannot know whether the drawing and spec agree.

Schedules create a similar problem. Door schedules, finish schedules, equipment schedules, lighting schedules, and hardware sets often live outside the model or contain information that is only partially represented in it. A door may appear correctly in 3D, but its schedule may list the wrong rating, frame type, hardware group, or access control requirement. A room finish plan may indicate one floor finish while the finish schedule assigns another.

Clash detection also does not catch broken cross-references and callout errors in the way a drawing reviewer would. A model does not inherently know that a detail bubble points to the wrong sheet, that a section marker references a missing detail, or that a keynote sends the reader to the wrong specification section.

Some code compliance issues are geometric, but many are not. Occupancy classifications, allowable area logic, hardware requirements, signage, fire alarm sequence notes, rated assembly continuity, accessibility scoping, and adopted-code edition issues often require interpretation across plans, schedules, specs, and jurisdictional requirements.

Finally, clash detection cannot catch anything that exists only in a 2D discipline's drawings and never enters the model. That remains a major practical limitation, because many real projects still include at least some 2D-only design information.

Closing

Clash detection and drawing review are complementary, not substitutes. Clash detection is strongest when coordinated models exist and the question is spatial: do modeled elements intersect, violate clearance rules, or conflict in sequence? Drawing review is needed when the question sits in the documents: do the plans, details, schedules, specifications, notes, references, and code assumptions agree with each other?

A strong coordination process treats both as necessary views of the same project. The model helps teams resolve physical space. The drawings and specifications define what actually gets issued, priced, permitted, and built.

Clash detection catches modeled geometric and sequence conflicts, but it does not catch document-level problems that are absent from or outside the model.