I came across an interesting article from NIST about work they’re doing with OIML on digitalizing metrology standards.
It gets pretty technical, but the basic idea caught my attention.
Think about a calibration procedure in the software you use today. It might tell you what measurements to take, what tolerances apply, what conditions are required, or how to determine whether something passes or fails.
You can read those instructions. The software can display them. But the software doesn’t necessarily understand what those instructions mean.
Now imagine the concept of making those kinds of requirements understandable to software.
Instead of simply displaying a requirement, software could potentially recognize that a particular test needs to be performed under certain conditions, which parameters apply, and what the acceptance criteria are.
That’s similar to the direction NIST is exploring.
Their current project with OIML is focused on OIML R60 and load cells, with a longer-term goal of representing specifications in a way that software can interpret and use in testing and evaluation.
That’s where I think this starts to get interesting for those of us working in calibration and metrology.
We’re also seeing work on Digital Calibration Certificates, where calibration results can be exchanged as structured data rather than information that someone has to pull out of a PDF and enter somewhere else.
So there are some interesting developments happening on both sides: making requirements and specifications more understandable to software, while other efforts are making calibration results more understandable to software.
Where all of this eventually leads remains to be seen, but it gives us an interesting glimpse at how metrology information might be handled differently in the future.
If you want to get into the technical details, NIST’s article goes much deeper:
Read the NIST article:
Digitalized Standards – Progress Towards the Next Generation of Standards Delivery
I’m curious what the metrologists here think.
Where could you see machine-readable requirements being useful in your work?
And would you be comfortable with software interpreting and applying requirements automatically, or would you still want a human reviewing those decisions?