Overall quality, Forensic BIM score™ and interoperability risk: three questions about one file
Summary
Every forensicBIM report opens with three judgements on the Scores page: overall quality, the Forensic BIM score™ and interoperability risk. Each answers a different question about the same file, so a file can do well on one and badly on another. Read them together before you accept a model, pay for it or rely on it.
| Judgement | The question | What it looks at | How it is shown |
|---|---|---|---|
| Overall quality | Can others reuse this model with little rework? | All five dimensions: validity, structure, geometry, information and consistency | 0 to 100: 80 and above is good, 55 to 79 needs attention, 30 to 54 is weak, below 30 is critical |
| Forensic BIM score™ | Is the information on the objects worth relying on? | Only the data: completeness, richness, reliability and semantics | 0 to 100 with a grade: A 80 and above, B 65 to 79, C 50 to 64, D 35 to 49, E below 35 |
| Interoperability risk | Will the file arrive intact in someone else’s software? | What can be lost, moved or merged on import, in four steps | A traffic light: High risk: fix before import, Medium risk: import with care or Low risk: ready to import |
Overall quality: can others reuse the model?
Overall quality is the weighted average of the five dimension scores. Information counts double; geometry and consistency count three quarters. Each dimension score is in turn the weighted average of its checks, which you find on the dimension pages.
A model cannot score well on quality without usable data. The overall quality is at most 35 + 0.7 × the Forensic BIM score™. When that limit applies, the report says Limited by its data, with the score the five dimensions would have given and the score after the limit. The Forensic BIM score™ and its grades explains the limit in detail.
Use it for comparing deliveries with each other and with comparable models (the Benchmark page gives a percentile), for a minimum level in your requirements, and for tracking whether a supplier’s files improve.
Forensic BIM score™: is the data worth relying on?
The Forensic BIM score™ looks only at the information the objects carry, in four factors: Completeness (30 points), Richness (25), Reliability (25) and Semantics (20). It ignores whether the file opens cleanly, how the spatial structure is organised or how heavy the geometry is. It asks whether the data is there, says something, can be trusted and has a defined meaning.
Use it for judging whether the model can support what comes after handover: operations, asset management, analysis or AI. A model can be tidy and complete in shape and still carry almost nothing to work with. See The Forensic BIM score™ and its grades.
Interoperability risk: will the file arrive intact?
Interoperability risk is a traffic light, not a score. It follows the file through four steps (opens, lands in place, follows the alignment, arrives complete) and flags information that would be lost, moved or merged on import, often without anyone noticing. One blocker makes it red, warnings make it amber.
Use it for accepting a delivery and before you import a model into a system of record, such as an asset register, a GIS or a maintenance system. See Interoperability risk: what can go wrong when someone imports your file.
Why three, and not one number
Each judgement can hide what the others show. A model with rich, reliable data can still fail on import because of one alignment gap. A clean export that passes every validity check can carry almost no data. A simple geometry-only model imports without trouble and gives you little to reuse. One combined number would average these situations away.
The two scores are linked on purpose: the overall quality is limited by the Forensic BIM score™, because a model without usable data should not look good. Interoperability risk stays outside both scores, because a single blocker matters however good the rest of the file is.
Reading them together
| What you see | What it usually means | What to do |
|---|---|---|
| Good overall quality, grade A or B, low risk | A model others can reuse, with data you can rely on, that arrives intact. | Accept it, and use it as the example for future deliveries. |
| Good scores, high risk | A good model with at least one blocker, such as an alignment gap, duplicate GlobalIds or a missing length unit. | Have the blockers fixed before you import it into a system of record. The rest of the model is worth the effort. |
| Limited by its data, grade D or E | Tidy geometry and structure, but little usable information on the objects. | Fine for coordination. Before handover, ask for classification, materials, quantities and standard property sets. |
| Low scores, low risk | A simple model, often geometry only, that imports without trouble but has little to reuse. | Accept it for viewing or coordination, not as a source of asset data. |
| Medium risk, whatever the scores | Information that can drop out unnoticed: linked documents, risky shapes, textures or quantities. | Import, then check the items listed in the four-step card. |
The bridge example report combines several of these: its Forensic BIM score™ is a D, and its interoperability risk is high because two road alignments have gaps of up to 31.4 cm between segments.
Which one answers your question?
| Your question | Look at |
|---|---|
| Can we accept this delivery? | Interoperability risk first, then overall quality. |
| How does this model compare with others? | Overall quality and the Forensic BIM score™ on the Benchmark page. Interoperability risk has no percentile. |
| Will this data support operations, asset management or AI? | The Forensic BIM score™ and its four factors. |
| What should the supplier fix first? | The blockers in the interoperability card, then the Potential page, which ranks fixes by the points they add to overall quality. |
| What should we put in our requirements? | All three, for example: overall quality of at least 55, a Forensic BIM score™ grade of C or better, and no interoperability blockers. |
Which fix moves which judgement
| Fix | Overall quality | Forensic BIM score™ | Interoperability risk |
|---|---|---|---|
| Add classification, materials and quantities | Rises (Information) | Rises (Completeness) | No change |
| Use standard property sets or bSDD properties | Rises (Information) | Rises (Semantics) | No change |
| Give objects that share a GlobalId their own | Rises (Consistency) | Rises (Reliability) | Removes a blocker |
| Close the gaps between alignment segments | Rises (Structure) | Rises (Reliability) | Removes a blocker |
| Declare a length unit | Rises (Validity) | No change | Removes a blocker |
| Georeference the model | Rises (Validity) | No change | Removes a warning |
| Repair broken document links | Rises (Consistency) | No change | Removes a warning; the references themselves stay a warning |
| Replace booleans and advanced B-reps | Can rise (Geometry) | No change | Removes a warning |
When the overall quality is limited by its data, fixes outside the data only show in the overall quality once the Forensic BIM score™ rises.
Where you see them in the report
- Every page: the Quality and Forensic BIM score™ pills in the header.
- Overview: both scores, the benchmark percentile and a short verdict.
- Scores page: all three side by side, the quality profile and the dimension bars, what makes up the Forensic BIM score™, and at the bottom the four steps of interoperability risk.
- Benchmark page: where both scores sit against comparable models.
Frequently asked questions
Can a file have good scores and still be high risk?
Yes. The scores weigh many checks against each other; interoperability risk looks for the single finding that breaks an import. A gap in an alignment or a missing length unit is enough for red, however good the rest of the file is.
Why doesn’t interoperability risk have a number?
Because it is a weakest-link judgement. A percentage would suggest that a hundred things done well make up for one blocker. On import, they don’t.
Why is my overall quality lower than the dimension scores suggest?
Two reasons: the dimensions do not weigh the same (information counts double), and the overall quality is limited by the Forensic BIM score™. The Scores page says Limited by its data when the limit applies.
Which one should I report to management?
For a single delivery, the light and the overall quality: can we use it, and how much rework does it need. For what the data can do after handover, add the Forensic BIM score™ grade.