Source: The Wall Street Journal
Volkswagen is among a growing number of firms experimenting with quantum computing to push their businesses forward
Companies have started to tap into quantum computing, like this D-Wave 2000Q System, in an effort to gain a competitive edge.
Quantum mechanics has fascinated, confounded and even alarmed scientists for nearly a century with the notion that particles can exist in two states at once and communicate with each other across vast distances. The underlying science that Albert Einstein famously called “spooky” could soon become one of modern computing’s core tenets.
Computers that utilize quantum mechanics are moving beyond pure scientific research and inching toward the commercial sector, with companies such as Volkswagen AG beginning to harness their unprecedented power to solve complex problems in nanoseconds.
“This technology is not futuristic,” said Martin Hofmann, Volkswagen chief information officer, who oversees information technology for the group’s 12 brands including Audi , Porsche and Bentley. “It’s a question of years until it’s commercialized, and investing right now in the technology is a big competitive advantage.”
Companies including D-Wave Systems Inc. and International Business Machines Corp. have been pioneering quantum computing, and experts say that within five years the technology could be powerful enough to solve new classes of problems that are currently beyond the grasp of even supercomputers.
While traditional computers use binary digits, or bits, which can either be 0s or 1s, quantum computers use quantum binary digits, or qubits, which represent and store information in both 0s and 1s simultaneously. This means the computers have the potential to sort through a vast number of possibilities within a fraction of a second to come up with a probable solution.
Volkswagen put that speed to use for a recent traffic-optimization project. Working on a $15-million D-Wave quantum computer over the cloud, a team of five in-house data scientists took GPS data from 10,000 taxis in Beijing and simulated specific routes that would allow each car to travel from downtown Beijing to the nearest airport, about 20 miles away, in the fastest time possible without creating a traffic jam.
After six months and several attempts, Dr. Hofmann and his team in March came up with an algorithm for the computer that optimized the routes for each taxi within a fraction of a second. A normal computer would have taken about 45 minutes to complete the same task, he said.
Dr. Hofmann’s next project is a navigation-based mobile app for drivers in Barcelona that will harness the power of quantum computing and machine learning, a part of artificial intelligence, to predict traffic jams and immediately send alternate routes. The first iteration is expected to debut by the end of the year, though Dr. Hofmann declined to go into more detail about the project. “If our project succeeds and in six to eight years the technology is where it should be, then traffic jams won’t happen anymore,” he said.
IBM has also developed a cloud-based quantum computing service, which became available in March for commercial business use. A similar service for individual experiments launched last year. Executives say that in less than five years IBM’s system could begin solving a range of computations for businesses related to chemistry and supply-chain optimizations. Those could include the ability to simulate significantly more molecular interactions within the human body than is now possible with traditional computers, which could accelerate new drug discoveries, said Bob Picciano, senior vice president of IBM Cognitive Systems.
The lab at D-Wave’s headquarters. The company has been pioneering innovative quantum computing solutions, but commercializing that technology is complicated.
But commercializing the technology is a complex task. Qubits can’t yet maintain their quantum mechanical state for more than a fraction of a second, said Dario Gil, vice president for science and solutions for IBM Research. They are delicate, easily disrupted by changes in temperature, noise or frequency. The number of qubits created by companies to date remains relatively small, meaning experiments are currently limited to a narrow swath of information. There also isn’t yet a programming language for quantum computing like there is for classical computing.
“This is phenomenally difficult from a tech perspective from every aspect,” including materials, devices, theory and error correction, Dr. Gil said.
Still, companies experimenting with quantum computing now stand to reap benefits if and when the technology becomes advanced enough to perform computations that classical computers can’t, said Paul Warburton, a professor of nanoelectronics at University College London. “From the commercial point of view, it’s a question of not wanting to be left behind when the real machine arrives,” said Dr. Warburton.