A Digital Twin for Safety and Reliability By Dr. Timothy Smith

Photo Source: Wikimedia Commons

Digital twins aim to reduce the unexpected malfunctions that can pop up at the wrong times. Do breakdowns often seem out of the blue and at the most inopportune time? Yes, you need to be at work by 9:00, but the battery died in your car without anyone around to help you jumpstart it. You need all your clothes cleaned for that trip to LA tomorrow, and the washing machine stopped spinning and will not drain out the dirty, soapy water. You invited twenty people over to watch the big game, and your TV shows only half the screen. These scenarios highlight inconveniences, but more life and death scenarios, such as the failure of a jet engine in a fighter plane on a mission or of a jet full of passengers or the brakes on an eighteen-wheeler tractor-trailer barreling down a mountain can happen with far more serious consequences.

To avoid mechanical or electrical disasters, engineers develop maintenance schedules to replace or repair machine parts in a preemptive process. Anticipating when a component will fail and replacing it with a new part involves active maintenance, but people do not want to incur unnecessary costs by replacing parts too soon. Additionally, most machines have sensors and instruments that tell the operator of essential malfunctions such as loss of tire pressure, an overheating engine, or low oil levels. The operator can respond to the warnings in real time and take appropriate action. However, it would save the unexpected, inconvenient, and even dangerous malfunctions if each machine had a personal maintenance monitor that could predict malfunctions far ahead enough to avoid bad situations.

The advancement of artificial intelligence, telecommunications, and sensor technology have come together in a technology called digital twins that can predict failure and suggest the proper course of action to prevent failure. Digital twins consist of a computer simulation of a real machine running alongside the real machine. The simulation also collects sensor data from its real twin. It combines that data with additional information such as weather conditions, time since maintenance, and expected performance from other similar machines. The continuous stream of data flows through telecommunication networks to the digital twin that lives in a computer outside the real machine. Such a digital twin will inform the machine operator of anticipated failures.

Digital twins factor significantly in the aerospace industry, and jet engine maintenance and performance have improved considerably with this technology. The US Air Force applies digital twins technology to monitor and maintain airplanes in its fleet. For example, two F-16 fighter jets decommissioned to the ‘Boneyard’ in Arizona were selected for a complete digital twin development. Engineers will completely take apart and reassemble the aircraft digitally to complete a model of the entire airframe and all the mechanical and electrical systems. These digital F-16 twins will help the Air Force to keep its pilots as safe as possible and optimize the planes for every mission. The information, such as weather and stress on the airframe, will enhance the digital twins and serve to advise other F-16s in the fleet. (defencenews.com)

Digital twins do not only serve the highly demanding technical machines of the military. Tesla, the manufacturer of electric vehicles, creates a digital twin for every car it manufactures. Each Tesla on the road continuously sends information from the vehicle’s sensors back to Tesla computers. (industryweek.com) These computers maintain a digital twin of the actual car. When problems appear imminent, the Tesla computers may automatically update software in the vehicle or adjust settings in the suspension, motor, or battery to keep the car running optimally. Additionally, the digital twin may call the car in for physical maintenance.

The power of digital twins will find its way into more aspects of our lives with the expansion of artificial intelligence, increasing computational capabilities, and the availability of better, more affordable sensors. Apple makes digital twins of its phones to keep these complex devices working smoothly. Ultimately, digital twins will make their way into healthcare with each person from birth living with their digital twin that can based on biological information and environmental conditions from weather and food consumption to disease exposure and genetic information. A human digital twin could anticipate degenerative diseases and propose different paths to maintain optimal health. There is potential for digital twins to make life less prone to unexpected physical failures of our machines and bodies. Still, the question will become whether the digital twin knows what ‘optimal life’ actually means and whether it goes beyond the body running as expected.

Dr. Smith’s career in scientific and information research spans the areas of bioinformatics, artificial intelligence, toxicology, and chemistry. He has published a number of peer-reviewed scientific papers. He has worked over the past seventeen years developing advanced analytics, machine learning, and knowledge management tools to enable research and support high-level decision making. Tim completed his Ph.D. in Toxicology at Cornell University and a Bachelor of Science in chemistry from the University of Washington.

You can buy his book on Amazon in paperback and in kindle format here.