Artificial intelligence (AI) is not new. In fact, the ideas have been around since the 1950s when Enigma code breaker, Alan Turing, discussed the building of intelligent machines in a scientific paper titled ‘Computing Machinery and Intelligence’. Turing’s mind was ahead of the technology though, so the ideas languished until semiconductor devices burst onto the scene to galvanise the field of digital computing in the 1970s.
In today’s world of digitalised manufacturing, readers of this magazine are no doubt aware of the symbiotic relationship that exists between people and machines in a modern processing plant – each depends on the other for optimum performance. AI algorithms provide the cognition for perceptive interpretation, while process-related data provides the ‘fuel’ that powers smarter executive decision making.
AI is also changing the way new products get designed. One of the perceived advantages of such an approach is that an AI algorithm starts with no preconceived idea of how best to solve the design problems. The design brief is simply fed to the algorithm complete with restrictions on material types, available production methods, budget limitations and time constraints. The algorithm then explores the merits of every possible production arrangement, before narrowing its focus to the best possible solution.
The manufacturing sector is perfectly suited to the application of AI. Even though the Fourth Industrial Revolution is still in an early phase, we’re already witnessing the benefits of applied AI methodology. Perhaps the most obvious is the progress being made in the area of maintenance strategy and management.
Over the lifetime of a facility, maintenance of plant machinery and equipment represents one of the major operating expenses, having a significant impact on the bottom line performance of any asset-intensive production operation. Recent studies show that unexpected asset failure is the cause of about 42% of all unplanned downtime.
For this reason, maintenance managers are under pressure to deliver more from less. Predictive maintenance has replaced the traditional preventative maintenance approach thanks to its ability to deliver better overall equipment effectiveness with greatly reduced spares holding requirements. After all, manufacturers able to predict the next failure of a machine or system stand to gain significant production advantages. Predictive maintenance uses pervasive sensing in combination with machine learning and AI to formulate assessments of equipment health and overall remaining useful life before overhaul becomes necessary.
Technews Industry Guide: Maintenance, Reliability & Asset Optimisation
Predictive maintenance strategy has become such an important component of competitive manufacturing that some years ago we took the decision to publish a separate handbook on the subject – posted every year with the April issue.
Maintenance is a fact of life, so it’s better to have the ability to plan for it than be caught up in the frenzy of an unexpected late-night breakdown. We hope this year’s copy of the Industry Guide is the perfect accompaniment to a more efficient maintenance strategy implemented within the requirements of your own facility.
Steven Meyer
Editor: SA Instrumentation & Control
| Tel: | +27 31 764 0593 |
| Email: | [email protected] |
| www: | www.technews.co.za |
| Articles: | More information and articles about Technews Publishing (SA Instrumentation & Control) |
© Technews Publishing (Pty) Ltd | All Rights Reserved
printer friendly version