When we think of high technology on the shopfloor, it is tempting to picture the high-value manufacturing environments of the automotive sector or the sleek operations of robotic warehousing businesses.

McKinsey argues that generative AI could also be employed here to great effect. It says, “In a manufacturing setting, for example, gen AI could organise data from maintenance logs, equipment images, and operational videos. A digital twin could analyse data, identify patterns or anomalies that might not be evident from unstructured data alone and inform decision making and predictive maintenance strategies. Additionally, gen AI tools can supplement data training sets used by digital twins by creating synthetic data.”

#2. Predictive maintenance

Around a plant, IIoT devices will collect information about machine states, performance and usage, including vibration, temperature, current, voltage, number of units produced, etc. This information can be used to help predict subsequent machine performance, including when a failure or problem is likely to occur.

AI is used to understand these patterns and make predictions about machine performance and likely failure or downtime. In this way, manufacturers can take preventative action to avoid disruption to operations and the associated costs of lost production time. Further, they can replace parts in good time before their failure creates further damage to other components or product.

#3. Enhanced quality control

The combination of sensor data to track quality issues and automated monitoring through the use of AI can power quality improvement programmes. For example, optical recognition systems combine cameras and other IIoT technologies with machine learning and AI capabilities to spot problems in quality control or output in real time. 

In this way, the two technologies combine to help spot problems earlier and, thereby, reduce the cost of malfunction or product defects. Improved quality control enhances product quality and reduces waste and, as a result, improves productivity, sustainability and profitability.

#4. Optimised production

AI can also be used to make sense of the vast quantity of sensor and machine data generated across the plant. To achieve this, an integrated data platform or data lake in which data can be consolidated and combined is the underpinning technology requirement.

AI can then be applied to monitor performance, look for anomalies or unintended impacts of particular operational decisions or methodologies and underperforming machines or parts of the plant. In this way, plant managers can make sense of the vast quantities of data being generated and spot opportunities for improvement which might otherwise have been overlooked. This will help to drive forward OEE and continuous improvement initiatives.

#5. Improved sustainability

As energy prices spiked following Russia’s war in Ukraine, it became a more pressing imperative for manufacturers to control and optimise their energy use. Always an essential part of profitability calculations, energy use suddenly became a decisive factor. While crippling and inflationary in the short term, this had the benefit of raising sustainability and Net Zero initiatives higher up the corporate agenda. 

Information is the first step to improved energy efficiency. Gathering data about energy use and assessing this against plant operations, environmental conditions and productivity metrics helps to identify opportunities to reduce energy use. However, these are complex calculations and AI has a role to play here in making sense of the vast quantity of data involved.

In this and so many other ways, AI is the perfect partner to IIoT – offering manufacturers a streamlined and effective way to turn plant data into actionable insights.