Have you implemented machine learning on your process?

There’s a lot of talk about doing machine learning. Talk is cheap, results are important. Sitting on the fence is not an option.

Here’s a typical situation.

Your plant operator knows a piece of equipment is failing, he doesn’t mind because it’s a rest break until maintenance fixes it.

The maintenance crew knows equipment repeatedly fails, they don’t care, overtime is an easy bonus.

The production manager knows production stops regularly. The production and maintenance crews provide different stories, it’s too hard a problem.

The Business unit manager knows production is meeting sales requirements and making a profit, that’s enough to keep the board off his back.

The Board relies on their business unit manager to be their eyes and ears for problems and to provide accurate reports on key metrics.

Without accurate knowledge about what is happening and why, everyone is flying blind.

Just imagine if someone identify trends in the data. Then pinpoint a solution.

The reality is modern process plant control systems collect a massive volume of data. That data can ‘see’ how the plant operates and it’s ready to be harvested.

The solution is easy, engage a data scientist to analyse the data, discover savings and and reap the benefits.

Can you afford not to?

If you are an operating process plant with a data problem to be solved, contact me today.

bmatthewtaylor@gmail.com