Greater manufacturing efficiency: it’s the golden egg we’re all looking for. But it’s getting increasingly hard to find. The solution could be smarter automation, which involves lots and lots of data (‘big data’) and data collection and data-driven modelling. The smart machine then uses the models to automatically adjust its own behaviour (i.e. machine learning).
The discussion about ‘Artificial Intelligence’ (AI) in manufacturing is now gaining momentum. This is mainly due to the availability of greater processing power and increasing volumes of data (the ‘sensorisation’ of the industry). A key factor that will help manufacturers to gain the maximum benefit from these recent advances is the use of adaptive algorithms. This in turn is a major stepping stone towards the development of ‘the factory of the future’.
Taking advantage of adaptive algorithms
In terms of the advancements required for Industry 4.0 (such as predictive maintenance and networked, efficient production), adaptive algorithms offer enormous potential. They can help you to create new ways of optimising your production. Many manufacturing companies are realising that AI represents a golden opportunity for increasing their Overall Equipment Effectiveness (OEE) and combining reduced costs with higher productivity.
However, there’s still something of a chasm between the desired status and the cold reality of the situation. Many of the AI solutions that are advertised on the market, which are often cloud-based, still have significant requirements in terms of infrastructure and IT. These solutions also work with an overwhelming amount of data – and the preparation and processing of this data is both laborious and time-consuming.
When it comes to cost-effectiveness, the question of added value often remains a somewhat murky issue for providers, who can’t determine whether or how any investment in AI will provide them with a return. Another contributing factor to this confusion is the fact that system designs for the mechanical engineering sector tend to be both complex and unique.
As a result, it isn’t simply a matter of transferring learnt experiences from other machinery, as might be the case for mass-produced products in the consumer goods industry. The majority of mechanical engineering systems are so complex that it isn’t possible to map out the entire system mathematically (as a ‘white box’) and maintain costs at an acceptable level. A ‘black-box’ approach is often more common. The available data in these systems for typical AI algorithms is under-determined, and a reliable operation can only be confirmed through testing, optimisation and, frequently, over-dimensioning.
Creating value with AI
Given these conditions, how can we design and integrate AI solutions that create tangible added value for the production process? Instead of laboriously searching a huge volume of data for patterns, in addition to the processes that are running, at Omron we tackle the issue from the opposite direction. We integrate the required algorithms into the machine control system, creating the framework for real-time optimisation — at the machine, for the machine. In contrast to edge computing, where individual manufacturing lines or sites are analysed using limited processing power, the AI controller that we use features adaptive intelligence. It’s closer to the action and learns to distinguish normal patterns from abnormal ones for each individual machine.
Our Sysmac machine automation platform is a complete solution for factory automation, featuring modules for control, motion and robotics, image processing and machine safety. The AI controller integrated within the Sysmac platform is primarily used in the manufacturing process at those points where the customer experiences the greatest efficiency problems (the ‘bottlenecks’). The processes gain intelligence based on previous findings and the improvements that have been made and subsequently drive the holistic optimisation of the entire manufacturing process.
Although OEE values of 80% and above have been achieved in isolated cases (particularly in the automotive industry), many of the systems currently in live usage have been generating figures of around 50%. However, if quality is improved and predictive maintenance is used to prevent machine downtime, it’s possible to make significant efficiency gains.
The AI controller provides optimisation in these specific areas. It’s driven by practical requirements and aims to noticeably improve the OEE. It’s important to note that an improvement of just a few percentage points can result in significant efficiency gains and cost reductions.
With taking the next step in implementing AI in manufacturing, Omron hopes to drive added value and practical improvements. This in turn will take us towards the creation of the factory of the future – the key ingredient in creating a smarter industry.
Interested in learning more about implementing AI in manufacturing? Contact Omron.