Data quality
When two teams pull the same figure and get two different numbers, the meeting stops being about the plant and starts being about the data. Consistency means one shared definition of every asset, every measurement and every unit, so the number on the control room screen is the number in the monthly report.
What it means
Consistency is agreement. The same asset carries the same name, the same measurement carries the same unit, and the same term means the same thing whether it appears in an operations report, a maintenance log or a board pack.
It is the quietest of the six dimensions, because inconsistent data is rarely wrong on its own terms. Each system is internally correct. The damage happens at the seams, when two correct answers refuse to agree.
What it costs
Inconsistency rarely triggers an alarm. It shows up as delay, rework and a slow drift away from trusting the numbers at all.
When production, finance and maintenance each arrive with their own figure, the first half of the meeting goes to working out whose spreadsheet is right. The decision the meeting existed to make gets pushed to the next one.
A reading in one unit compared against a limit set in another does not look broken; it looks like an ordinary number. Those are the mismatches that survive review and end up driving a shutdown, a purchase or a safety call.
If every report starts by translating one system's names into another's, that translation lives in one person's head or in a file only they maintain. When that person moves on, the analysis stops being repeatable and the work is done again from scratch.
Our approach
Consistency is not a cleanup project that gets funded once and then decays. We treat it as a design decision — one place where assets, measurements and definitions are described a single time.
Every asset, and what it does, is described once and referred to everywhere rather than re-entered system by system. Teams stop maintaining private translations between their own naming and everyone else's.
A value is never held as a bare number. It carries what it measures and what it is measured in, so nothing that uses it later has to guess or rely on a convention that was only ever written down in a handover note.
Terms like availability, downtime and batch are settled in the model itself, where everyone can see them, instead of being redefined in each spreadsheet. When a definition has to change, it changes in one place and every view follows.
The other dimensions
Quality is not a single number. Each dimension fails in its own way, and each one is worth understanding on its own terms.