Automakers began to reinvent themselves as digital companies a few years back, but now that they are emerging from the business trauma of the pandemic, the need to complete the digitalization journey is more urgent than ever. They will have no choice as more technology-focused competitors adopt and implement digital twin-enabled production systems and move forward with electric vehicles (EVs), connected vehicle services, and eventually autonomous vehicles. Car makers will make some tough decisions to bring software development in-house and some will even start building their own vehicle-dedicated operating systems and computer processors or partner with some of the chip makers to develop next-generation operating systems and chips to run on-board systems for future autonomous vehicles.
How AI is changing production operations
Automotive assembly areas and production lines are using artificial intelligence (AI) applications in several ways. These include new generations of smart robots, human-machine interaction, and advanced quality assurance methods.
While AI is being used extensively in vehicle design, car makers also currently use AI and machine learning (ML) in their manufacturing processes. Robotics in assembly lines is nothing new and have been used for decades. However, these were robots in cages that operated in strictly defined spaces and did not permit any human incursion for safety reasons. With AI, smart collaborative robots can work with their human counterparts in a shared assembly environment. Collaborative robots use AI to detect and sense what human workers are doing and adjust their motions to avoid injuring their human co-workers. Painting and welding robots, when powered by AI algorithms, can do more than just follow a pre-programmed routine. AI allows them to identify defects or irregularities in materials and components and to adapt the process accordingly, or to issue quality assurance alerts.
AI is also being used to model and simulate production lines, machines, and equipment, and to improve the overall throughput of the production process. AI enables production simulation to go beyond a one-time simulation of a pre-determined process scenario to dynamic simulations, which can adapt to changing conditions, the state of materials and machines, and alter the simulation. These simulations can subsequently adapt production processes in real time.
The rise of additive manufacturing for production parts
Using 3D printing to fabricate production parts is now an established part of automotive production, and this industry is second only to aerospace and defense in the use of additive manufacturing (AM) for production. Most vehicles produced today have a wide assortment of AM fabricated parts incorporated into the overall assembly. This includes a range of automotive parts from engine components, gears, gearboxes, brake components, headlamps, body kits, bumpers, fuel tanks, grills, and fenders, to frame construction. Some car makers are even printing complete bodies for small EVs.
For the burgeoning EV market, AM will be especially significant in terms of weight reduction. While this has always been desirable for traditional internal combustion engine (ICE) vehicles for improving fuel efficiency, this concern is more important than ever since lower weight can mean much longer battery life between charges. Additionally, battery weight itself is a downside to EVs where the batteries can add more than a thousand extra pounds to a midsized EV. Automotive components can be designed specifically for AM fabrication to be much lighter with a vastly improved weight-to-strength ratio. Almost every part in every type of vehicle can now be made lighter through AM fabrication rather than the use of metals.
Digital twin optimizes production systems
By using a digital twin in automotive production, it is possible to plan the entire manufacturing process in a fully virtual environment before physically building production lines, conveyance systems, and robotic work cells, or installing automation and controls. Because of its real-time characteristics, a digital twin can simulate a system while it is operational. This allows manufacturers to monitor the system, create models for adjustments, and make changes to the system.
The implementation of a digital twin enables the optimization of each phase of the production process. Capturing sensor data throughout the functional components of the system provides essential feedback, enables predictive and prescriptive analytics, and minimizes unplanned downtime. Additionally, virtual commissioning of the automotive production line works in conjunction with the digital twin process by validating the operation of the controls and automation functions and providing the baseline operation of the system.
Recommendations
The automotive industry is entering a new era, challenged by the imperative to move to an entirely new product based on a complete change in propulsion used for mobility. The change from ICE vehicles to EVs is being mandated because of the clear need to reduce carbon emission output and mitigate the increasing warming of our planet. The automotive industry is meeting the challenge to design and manufacture the next generation of EVs by embracing emerging the science and technology of AI and AM along with implementation of the digital twin to meet these challenges. Other industries would be well served to follow the example of the automotive industry in using technology and science to move their industries into the 21st century.
