IBM quantum computing: From healthcare to automotive to energy, real use cases are in play
Companies including Anthem, Daimler-Benz and ExxonMobil, have big plans to deploy IBM quantum computers this decade.
Quantum computers have been receiving a lot of attention because of their potential to solve computationally difficult problems that classical computers cannot. Among those problems are the abilities to help companies reduce their carbon footprint and protect the world from the next pandemic.
SEE: The CIO’s guide to quantum computing (free PDF) (TechRepublic)
Since announcing the IBM Quantum Network in 2017 with 12 initial organizations, today IBM said its commercial quantum computing program has grown to more than 150 Fortune 500 companies, academic institutions, startups and national research labs. Some 260,000 users have run nearly a trillion circuits, according to the company.
Last spring, IBM rolled out Qiskit Runtime, and the ability to speed up quantum programs on the cloud by 120x, as well as IBM’s to deliver a 1,000+ qubit system by 2023.
In addition to increasing speed, Qiskit Runtime changes how IBM is able to offer quantum computing to clients and make it more widely available, the company said.
“Before we got to Runtime, clients were doing research using simulators,” and now they can investigate applications for finance, machine learning and chemistry using real hardware, said Jay Gambetta, IBM fellow and vice president of quantum computing.
“To me, this is fundamentally important because a simulator can never mimic quantum computing, so you need to do your research and development on the hardware and that’s what’s getting enabled,” Gambetta said. “I see this year as when this fully comes out of beta and will be the new way of using quantum” to ask questions such as whether quantum will scale in the way clients can use apps.”
Identifying anomalies in healthcare
In the meantime, customers are incorporating quantum into their plans for the future. At healthcare provider Anthem, quantum computing is “an integral part of our digital platform for health,” and is being used for “computationally intense and expensive tasks such as identifying anomalies, where there’s tons of data and interactions,” said John Utz, staff vice president of digital product management.
Quantum computers are better at that than classical computers, Utz said. Anthem is running different models on IBM’s quantum cloud. Right now, company officials are building a roadmap around how Anthem wants to deliver its platform using quantum technology, so “I can’t say quantum is ready for primetime yet,” Utz said. “The plan is to get there over the next year or so and have something working in production.”
SEE: Expert: Now is the time to prepare for the quantum computing revolution (TechRepublic)
A good place to start with anomaly detection is in finding fraud, he said. “Classical computers will tap out at some point and can’t get to the same place as quantum computers.”
Other use cases are around longitudinal population health modeling, meaning that as Anthem looks at providing more of a digital platform for health, one of the challenges is that there is “almost an infinite number of relationships,” he said. This includes different health conditions, providers patients see, outcomes and figuring out where there are outliers, he said.
“There’s only so much a classical system can do there, so we’re looking for more opportunities to improve healthcare for our members and the population at large,” and the ability to proactively predict risk, Utz said. Quantum computers are better at driving outcomes from the models Anthem is building, he said.
Developing energy-dense batteries
Daimler AG, the parent company of Mercedes-Benz, is studying how to develop energy-dense batteries such as the lithium-sulfur battery. But going from the drawing board to a commercially viable Li-S battery is “essentially a mammoth chemistry experiment,” the company said.
Engineers are testing quantum systems to distill some very abstract physics theory into a new kind of computing power that can handle what IBM calls “once-insoluble complexity.” Using quantum bits known as qubits, the performance doubles, giving a substantial boost to the ability to run algorithms to speed the simulation process and test the feasibility of the battery, the company said.
Overcoming shipping challenges to meet energy needs
Energy challenges are expected to increase as the global population grows from 7.5 billion today to a projected 9.2 billion by 2040, according to ExxonMobil. This has created what the company refers to as the “dual challenge” of providing reliable and affordable energy to a rising population while also reducing environmental impacts and the risks of climate change.
One way to tackle that challenge in the near term is to use natural gas, which emits up to 60% less greenhouse gases than coal, according to Dr. Vijay Swarup, vice president of research and development at ExxonMobil, in a statement. This creates issues with production and transportation, he said.
It requires efficient liquified natural gas shipping, but finding optimal routes for a fleet of LNG ships to transport critical fuel supplies is a “mind-bendingly complex optimization problem.” It involves accounting for each ship’s position every day of the year along with the LNG requirements of each delivery site.
This type of problem cannot be solved exactly with classical computing, IBM said. So ExxonMobil, in tandem with IBM Research, is using a combination of classical and quantum computers to address the complexity. Teams are modeling maritime inventory routing on quantum devices, analyzing the strengths and tradeoffs of different strategies for vehicle and inventory routing, and laying the foundation for constructing practical solutions for their operations, IBM said.
Swarup said ExxonMobil’s goal is to increase its ability to tackle more complex optimizations and previously insoluble routing problems as IBM’s quantum hardware scales from small prototype systems to larger devices.
This post was written by and was first posted to www.techrepublic.com