How companies are already using AI in manufacturing

Manufacturers worldwide are actively integrating AI into their operations for automation, optimisation and innovation. They are harnessing AI's power to enhance efficiency and gain a competitive advantage. Indeed, 93% [1] of industry leaders express excitement about the potential of AI in manufacturing.

Several factors combine to make manufacturing an ideal candidate for artificial intelligence implementation. Production often involves complex yet variable processes and generates vast amounts of data from diverse sources. Reports indicate that the average factory generates 1 TB of production data each day [2]. However, less than 1% of this data is being analysed.

There is a strong focus on quality control, product consistency and safety. This is especially within heavily regulated sectors like food and drink, healthcare, construction, transport and chemicals. In global markets, the level of competition is extremely high. This, along with the need to respond swiftly to market fluctuations, adds to their pressures.

AI is proving invaluable in addressing these challenges by analysing, enhancing and optimising processes to boost efficiency and productivity. The technology is helping prevent costly mistakes and setbacks, and facilitating more informed decision-making. Through artificial intelligence implementation, manufacturers are able to react faster, minimise disruptions and seize new growth opportunities.

Instead of solely relying on our claims, look at these three companies. They showcase the thrilling potential of AI in the manufacturing industry.

AI drives the future of quality control

Audi

Audi has began to roll out an AI system for quality control of spot welds in car body construction. This follows the success of a pilot project. The AI was developed and tested at Audi’s manufacturing plant in Neckarsulm, close to Stuttgart. Over the course of each shift, it analysed 1.5 million spot welds on 300 different vehicles.

Until recently, production staff used ultrasound to manually monitor the spot-welding process based on random analyses. The ultrasound method checked around 5,000 spot welds per vehicle. By applying AI, employees can now focus on spotting possible anomalies. Audi says this new approach will enable them to control quality more efficiently and in a more targeted way.

The Volkswagen Group's other plants are already in the process of making preparations to implement the technology into their operations. This includes Audi Brussels and the Volkswagen plant in Emden, which sits near the Germany-Netherlands border. The data that is generated by AI can be used in a wider variety of contexts. Its potential for predictive maintenance is something the Audi group is looking into at the moment.

The project also serves as a use case for Audi’s Automotive Initiative 2025 (AI25). This group- wide programme aims to establish a common framework and partner ecosystem for smart factory transformation and innovation. Audi’s overarching goal is to leverage digitalisation to make production and logistics more flexible and efficient.

AI in CNC machining

CloudNC and the CAM Assist

A plug-in for the Fusion 360 software platform that is powered by AI is called CAM Assist. It has quickly gained a reputation as the most significant change in precision manufacturing in decades.

CloudNC developed it in response to the repetitive, tedious, and manual nature of computer numerical control programming. The company refines CAM Assist in its production facility located in Chelmsford, England. It simplifies the process of generating 3-axis machining strategies. With just a click of a button, it can automatically produce professional toolpath instructions in seconds.

User feedback indicates that CAM Assist is capable of completing 80% of the machining strategy. The remaining 20% is left for manual review and adjustments. According to CloudNC, this streamlined technique has the potential to save hundreds of hours of manufacturing time. It enables engineers to do jobs in a matter of seconds which would normally take them hours or even days to complete.

AI in CNC machining that is both quicker and more efficient is one of the perks that manufacturers appreciate. In addition, the process of estimating and quoting will be sped up, and the personnel will be able to acquire new skills more quickly. “CAM Assist helps manufacturers win more deals, complete jobs faster and create the capacity to take on more work in a virtuous cycle,” says Theo Saville, CEO and Co-founder of CloudNC.

The CloudNC team had a successful early access launch in 2023. They are currently refining their AI-powered solution for generating 4 and 5-axis machining strategies.

AI helps deliver the clean energy revolution

Siemens

Siemens Gamesa Renewable Energy is leveraging AI-powered virtual twins. This technology helps to optimise the layout of wind farms. As a result, it increases overall production and reduces operating costs.

The company operates thousands of turbines globally. These turbines generate a combined total of 100 gigawatts of wind power, which is sufficient to supply energy to nearly 87 million homes. Analysts project that the global demand for wind power will quadruple from 2020 to 2025. Therefore, maximising the power produced by each turbine is more important than ever.

Analysing wind farms involves a complex process. It requires minimising and optimising for various wind and weather scenarios in real time. This task demands hundreds or even thousands of iterations and simulation runs. Time constraints and costs previously prohibited this.

Siemens Gamesa overcame these limitations by leveraging physics-informed machine learning and a cutting-edge design simulation platform. Researchers can now simulate the behaviour of an entire wind farm quickly and accurately. They incorporate data such as wind direction, terrain features, and the positioning of individual turbines.

This allows engineers to make more informed decisions about turbine placement, control strategies and future performance. The team can now accurately simulate the effects of adding a turbine next to another. This allows them to anticipate any changes in wind flow that could lead to a decrease in electricity generation.

Sources:

[1] https://www2.deloitte.com/cn/en/pages/consumer-industrial-products/articles/ai- manufacturing-application-survey.html

[2] https://www.ibm.com/downloads/cas/QRKNQ4A