Question 1

How is Engineering Ground Truth different from a CDE or BIM database?

Accepted Answer

A CDE is mainly an environment for sharing, approving and distributing documents and models, and a BIM database is the model-information store of a particular authoring tool. Engineering Ground Truth is the reference state, across many sources, showing which engineering information was verified on what evidence and by which rules. It does not replace a CDE or BIM database; it connects the information inside and around them as traceable references.

Question 2

How is Engineering Identity different from a simple Object ID?

Accepted Answer

An Object ID is valid only inside a particular file, program or database, and can change during copying, conversion or merging. Engineering Identity keeps external IDs through mapping while maintaining the continuity of the engineering object. No single classification code, file name, coordinate or external GUID is treated as the identity on its own.

Question 3

Does part-level detail make the data too heavy?

Accepted Answer

Creating Parts does not mean duplicating and storing every piece of information. Only the minimum unit needed for the purpose is identified; geometry, materials, rules and evidence are linked by reference, and heavy views can be generated when needed. The appropriate granularity is set by project size and purpose, and performance is verified with real data.

Question 4

How is METOMOS's 5D different from ordinary BIM + cost integration?

Accepted Answer

Ordinary 5D BIM is often used to mean a 3D model linked to cost, but the definition varies by market. METOMOS manages 5D Cost & Value Information as its own perspective and, when needed, combines 3D, 4D and 5D into a Composite BIM Profile. What matters is not the number in the name but whether the source quantities, unit rates, schedule and evidence can be traced.

Question 5

How are generative AI, rules and deterministic engineering separated?

Accepted Answer

Generative AI is suited to summarizing, classifying and generating candidates from unstructured material. Rules state the conditions of application, constraints and grounds for judgment, and Deterministic Engineering reproduces the same calculation result from the same input. METOMOS separates the three roles so that AI proposes, and rules, calculation, verification and authority gates confirm.

Question 6

If an AI or engineering rule is wrong, won't all the connected information be wrong too?

Accepted Answer

That is why version, source, evidence and scope of application are attached to rules and results. When an error is found, nothing is silently overwritten: a Change Event and the affected objects are recorded so that only the necessary scope is recalculated and re-verified. Changes to high-risk rules must pass human approval and regression testing.

Question 7

What happens when source drawings or data are wrong?

Accepted Answer

A source error is first marked as a Candidate or Exception, and automatic confirmation stops. It is compared with other drawings, lists, specifications and site evidence to show the inconsistency, and an authorized user corrects and approves it. The principle is to trace the error and its scope of impact rather than hide a wrong input behind a plausible result.

Question 8

When drawing, BIM, specification and BOQ disagree, which is right?

Accepted Answer

No single one is given priority by default. The decision is based on the contract-document hierarchy, the latest approved revision, the authority of each document type, the revision, the point of application and the verification evidence. The system detects and links the differences, and conflicts that cannot be judged automatically are passed to the responsible person as a Resolution Issue.

Question 9

What is near-real-time EVMS?

Accepted Answer

Near-real-time EVMS is the idea of quickly updating PV, EV, AC and forecasts in step with the cycle at which schedule, cost and progress changes are approved and collected. It does not mean updating every second like a sensor; it means reducing delay and manual re-entry by connecting source information with execution information. The actual update cycle and indicator scope are confirmed according to the customer's systems, closing cycle and data quality.

Question 10

Which files and software are supported?

Accepted Answer

DWG, IFC, Excel/CSV, document formats, and schedule, cost and ERP data can be connected through adapters and mapping. But “the file can be opened” is not the same as “meaning, attributes and history round-trip completely.” Supported vendors, versions and the read, write and verification scope are confirmed per project through a Support Matrix and a PoC.

Question 11

How much does a customer have to change its existing way of working?

Accepted Answer

At first, existing tools and approval procedures stay as they are, and improvement starts where repeated entry, code mapping and verification are heaviest. Users can keep working in familiar screens while references, rules and evidence are added underneath, step by step. The scope of change management is assessed against both the business benefit and the organization's readiness.

Question 12

Can older engineering rules and software logic still be used today?

Accepted Answer

Yes, but not by copying them as they are. Existing formulas, classifications and program results are preserved together with their source, reproducibility is confirmed against a Golden Dataset, and then they are migrated to modern languages and services. Logic whose meaning is unclear or that has not been verified is isolated and goes through expert review and regression testing.

Question 13

Codes and construction methods differ by country — is global expansion possible?

Accepted Answer

The core architecture keeps common principles of identity, reference, state and evidence, while each country's codes, unit rates, classifications and construction methods can be separated into Profiles and Rule Packages. Global expansion requires local experts, translated regulations, certification and contract review, and project-level verification. One country's rules are never imposed as if they were a world standard.

Question 14

Is METOMOS a software company or an engineering service company?

Accepted Answer

METOMOS combines specialist software packages with an Engineering Data Foundry service. The software provides repeatable product functions, and the Foundry organizes and verifies the customer's rules, data and legacy so that the products can use them. Over the long term, the aim is a structure in which service results accumulate into product functions and standard datasets.

Question 15

As the number of Metomers grows, won't QA become a new bottleneck?

Accepted Answer

It could, which is why QA is never concentrated on a single person. Automated checks, separation of roles, multiple reviews, sample audits, evidence records and escalation by difficulty are combined. Pass rate, rework rate, review time and agreement rate are measured before the number of Metomers (distributed specialist workers), and capacity is expanded only where the quality indicators are stable.

Question 16

Is a 2–5× productivity improvement already proven?

Accepted Answer

It should not be said that 2–5× has already been proven universally across every project. Some technical experience and automation potential can be reviewed as evidence, but productivity depends heavily on scope, staffing and data quality. Figures on this website are presented, with their conditions, only where a Same-Scope Benchmark and reviewable results are available.

Question 17

What if a benchmark result is worse than the existing method?

Accepted Answer

An unfavorable result is kept as important verification evidence too. We analyze whether the cause lies in data preparation, interfaces, missing rules, learning cost or product performance, and may narrow and improve the scope or stop the adoption. A benchmark is not a procedure for justifying a sale; it is a Go/No-Go decision tool.

Question 18

What is the fastest way to validate METOMOS?

Accepted Answer

The fastest way is to choose one task from real customer material whose scope and correct answer are relatively clear. Perform it with the existing method and the METOMOS method on the same input, and compare time, quality, traceability and change response together. Agree the deliverables and review criteria in advance, and expand to the next task once the success conditions are met.

Question 19

If generative AI or AGI advances greatly, won't METOMOS become unnecessary?

Accepted Answer

The more AI advances, the more the source, applied rules, authority, verification, responsibility and change history of what it generates matter. METOMOS's role is not a particular AI model itself but providing the meaning, references, evidence and gates that let AI take part safely in engineering work. Even if the runtime is replaced to suit future models, the reference information and lineage are kept.

Question 20

How can past projects and technology experience be verified?

Accepted Answer

Verification does not rely on contracts or introductory material alone; original drawings, quantities and bills, schedules, program results, registered patents, project deliverables and change histories are checked. Where possible, results are regenerated from past inputs and compared with a Golden Result, and reviewed by a third party or the customer's experts. Experience that is hard to reproduce is classed as “history,” and reproducible results as “verification evidence.”

Technical answers, by topic.

Engineering Data Architecture

AI, Rules & Validation

EVMS & Project Controls

Integration & Existing Systems

Legacy Technology & Modernization

Foundry & Scalability

Performance & Benchmark

Future AI & Evidence

Validate the scope that matters.

Can’t find the answer you need?

Technical answers, by topic.

Engineering Data Architecture

Q1-1How is Engineering Ground Truth different from a CDE or BIM database?+

Q1-2How is Engineering Identity different from a simple Object ID?+

Q1-3Does part-level detail make the data too heavy?+

Q1-4How is METOMOS's 5D different from ordinary BIM + cost integration?+

AI, Rules & Validation

Q2-1How are generative AI, rules and deterministic engineering separated?+

Q2-2If an AI or engineering rule is wrong, won't all the connected information be wrong too?+

Q2-3What happens when source drawings or data are wrong?+

Q2-4When drawing, BIM, specification and BOQ disagree, which is right?+

EVMS & Project Controls

Q3-1What is near-real-time EVMS?+

Integration & Existing Systems

Q4-1Which files and software are supported?+

Q4-2How much does a customer have to change its existing way of working?+

Legacy Technology & Modernization

Q5-1Can older engineering rules and software logic still be used today?+

Q5-2Codes and construction methods differ by country — is global expansion possible?+

Foundry & Scalability

Q6-1Is METOMOS a software company or an engineering service company?+

Q6-2As the number of Metomers grows, won't QA become a new bottleneck?+

Performance & Benchmark

Q7-1Is a 2–5× productivity improvement already proven?+

Q7-2What if a benchmark result is worse than the existing method?+

Q7-3What is the fastest way to validate METOMOS?+

Future AI & Evidence

Q8-1If generative AI or AGI advances greatly, won't METOMOS become unnecessary?+

Q8-2How can past projects and technology experience be verified?+

Validate the scope that matters.

Can’t find the answer you need?