Car manufacturers face a range of challenges globally as they strive to move towards sustainable manufacturing. Central to this is ensuring that production processes remain as clean and efficient as possible while maintaining product quality and reducing wastage. Digital transformation is underpinning this as cloud-based technologies such as artificial intelligence (AI) and machine learning (ML) play pivotal roles.
Accuracy and timeliness
One area where ML is supporting car manufacturers is in reducing production line interruptions. Automotive industry specialist, Richard Felton, explains that ML systems can help avoid unplanned maintenance by analysing data to improve predictive maintenance schedules. “If you avoid unnecessary maintenance, you reduce costs, increase productivity, and do not have unplanned downtime,” he says. “ML not only handles the sheer scale, breadth and accuracy of the data, but also the timeliness.”
The technology can also help manufacturers navigate current global component shortages. “Manufacturers are using ML to anticipate shortages and how to handle those shortages in components more efficiently,” Felton adds.
Efficient component inspections
Additionally, ML supports component quality inspections using data from camera inspections to check assembly processes and sequences in terms of complexity, speed and accuracy. “The machine learning can spot anomalies that human operators might miss across millions of data points,” he adds.
This digital transformation is supported by companies such as Amazon Web Services (AWS) which, as a cloud service provider, enables customers to access and manage data, scale globally and make data-driven decisions in real time using AI, ML and other advanced services.
AWS helps with sustainability, digital manufacturing and supply chains, and improves overall equipment effectiveness by capturing, analysing and visualising plant floor data. The service brings all this together as an holistic solution to support the automotive industry.
Felton, who is the senior practice manager (automotive) at AWS, says the platform has purpose-built capabilities, drawing on expertise from across the automotive industry, and offers the “broadest partner ecosystem of any cloud specifically for automotive customers to help them transform their businesses.”
Automating processes with AI
In one example, the company supported a digital production platform for Volkswagen (VW), which has 12 brands operating across more than 120 sites, 1500 suppliers and 200 million parts a day entering its factories to make 11 million cars a year. “We helped VW tackle that very complex operation with the digital production platform, with analytics in the cloud to help them achieve efficiency, quality and sustainability,” Felton explains.
During production, Volkswagen Group brands apply 25 different labels that contain country-specific safety, usability and specification data with over 2000 variants. To automate this process, VW’s Porsche brand developed a solution using the services, which saw a manual label inspection programme replaced with an AI-driven programme to conduct the process automatically and with greater speed and accuracy.
Providing integrated solutions
Manufacturers are increasingly looking for integrated solutions that combine manufacturing systems with those operated by their supply chain partners to reduce transport costs and lead to more sustainable ways to move millions of parts. In this instance, the digital platform configures route optimisation and ensures correct demand forecast to reduce component waste.
With a trend towards electric vehicles driven by sustainability goals and emissions regulations, computer-aided engineering can support durability, crash protection, safer use of batteries and thermal modelling of advanced cooling systems. The services can help customers predict and understand battery health, capacity and failures, range and weather conditions impacting battery performance.
The company also works with Rivian Automotive on high-performance computing for design engineering for crashworthiness, aerodynamics and durability. “There is a significant time and cost saving by doing simulation in the cloud, as it gives access to a scale that you do not have on present systems; you have almost limitless capacity to do simulations,” says Felton.
© Technews Publishing (Pty) Ltd | All Rights Reserved