Fujitsu
Not logged in » Login
X

Please login

Please log in with your Fujitsu Partner Account.

Login


» Forgot password

Register now

If you do not have a Fujitsu Partner Account, please register for a new account.

» Register now
Apr 07 2019

The Power of Quantum-Inspired Computing: Journey of Digital Annealer (Part 4)

/data/www/ctec-live/application/public/media/images/blogimages/digital-annealer/Digital_Annealer_Slide.jpg

For most of us, the idea of quantum computing is akin to an exotic beast: Somewhere in the back of our minds, we know that it exists and that at some point in the future we might even be able to build proper machinery, which will be easily accessible and can solve real world problems that are unsolvable today. But what would you say if someone told you that it is possible to provide and utilize quantum-like capabilities and solve these problems in the here and now? Well, that would seem like this person had just claimed to have spotted Schrödinger's cat in the wild. This five-part blog will help us find out more.

In the previous chapters of our series, we discussed core principles of quantum and quantum-inspired computing as well as obstacles that hampered the development of adequate hardware and finally how Fujitsu's Digital Annealer can help to overcome these challenges and enable customers to solve complex combinatorial optimization problems on the spot. Thus far, we gave you quite a bit of abstract stuff to digest – now it's time to add more color to the mix so that you may assess which of your clients could benefit from the new technology.

Real-World Use Cases
Before we outline a few practical examples, let's first look at the very special category of tasks Digital Annealer was developed for: combinatorial optimization problems. Combinatorial optimization is essentially, the process of obtaining the optimal solution from a finite set of possibilities. While the general term may seem exotic, the problems themselves are pretty common not only across all types of industries, but in all spheres of life. For example, in the public sector cities and national governments are urgently addressing the issue of traffic optimization. The potential benefits are significant: better air quality means lower levels of respiratory and other diseases, and increased citizen well-being. Lower carbon emissions feed through to national targets, enabling governments to focus their spending on other vital policy areas. More efficient journeys with optimum travel time reduce frustration, encourage economic growth and productivity.

Let's assume you want to determine the optimal routes for five different pairs of starting points and destinations in a way that avoids traffic congestions. Right off the bat, you'd have to consider 10100 – or one googol – of possible combinations. Even if you were to work with today's most powerful computers, this would be an utterly futile venture sure to last for eons. By contrast, Digital Annealer can perform the same calculation in one single second.

Aside from traffic flow optimization, there's a plethora of other real-world scenarios that Digital Annealer can help with, including, to name a few, robot positioning, job shop scheduling, and car design optimization; factory optimization; material or life sciences; and finally portfolio optimization in the financial industry. Three ambitious real-world optimization problems that Fujitsu's Digital Annealer has shown exceptional results in are as follows:

    • One large automotive OEM company has solved numerous optimization challenges – including job shop scheduling, engineering design and robot positioning for chassis welding to accommodate just-in-time manufacturing – in collaboration with Fujitsu teams. The impact on the processes and its results have been extremely encouraging and overwhelming. For example, PVC seam sealing is part of the paint shop process in car manufacturing, which is one of the most expensive steps in car production that contributes an average 40% to the total manufacturing costs. Any form of optimization which could be done within the paint shop process will potentially reduce this cost significantly. Fujitsu along with OEM manufacturer chose to test the PVC seam sealing process in which robots must be able to move in two different directions and it must be ensured that no overlap occurs when dealing with a quintillion of possible combinations of welding. Using today's 8192-bit Digital Annealer, this automotive vendor was able to identify the optimal welding roundtrip, which means its staff is now able to fully handle 64 seams at once, resulting in more cars being produced with the same resources and within the same time frame – a major advantage in an extremely competitive market.

Image

Fig. 1: A comparison done to see how Digital Annealer performs vis-a-vis a traditional quantum computers. The potential of what the solution could achieve is truly game changing.

  • Fujitsu IT Products Limited, a key manufacturing facility of ours, uses Digital Annealer to optimize its warehouse operations. One key issue that had to be solved was to find optimum pickup routes for computer components that are then sent to the assembly lines. The existing ticket system did not include the order of part shelves to visit, and there was no way to simply deduce the best route. As a result, even experienced warehouse workers often spent quite some time gathering the required components. Today, Digital Annealer finds the shortest route and sends a map to the tablet of each individual clerk that they use for orientation. Thus, workers are no longer forced to memorize parts locations and figure out the best course for themselves. Fujitsu IT Products Limited was able to reduce the traveling distance for gathering parts by 30% each month, and further changing the shelf location proactively based on forecast could reduce the distance travelled by up to 45%.
  • Toray Industries. a Japanese corporation that once started as a basic materials manufacturer and now specializes in industrial products used in organic synthetic chemistry, polymer chemistry, and biochemistry, employed Digital Annealer in to improve its capabilities in protein structure prediction, a key area of drug discovery and biotech research. More specifically, they were able to identify the most stable protein side-chain structure among a set of 10100 candidates within minutes. Moreover, Digital Annealer can be used to improve stereochemical prediction accuracy in targeted protein research, thus helping to accelerate drug discovery.
  • In financial services, a UK-based retail and commercial bank and long-term Fujitsu project partner, NatWest, announced last September their plan to implement Digital Annealer after conducting successful tests on portfolio optimization, which previously counted among the most complex, time-consuming and expensive problems the company was facing. In the future, the bank's portfolio managers hope to use quantum-inspired computing power to determine the right composition of its £120 bn ($156.5 bn/€137.9 bn) worth of high quality liquid assets (HQLAs). HQLAs are assets such as cash and bonds that every UK bank must hold as a buffer in case it runs into financial trouble. In the initial tests the bank was able to optimize its portfolio through diversification 300 times faster and with higher accuracy than before.

Image

Fig. 2: Digital Annealer is able to provide a well-diversified, optimal portfolio option which gives a high return given keeping the volatility required by regulation.

Needless to say, this selection only represents a fraction of the potential use cases, as the technology itself carries an intrinsic potential to transform various industries and enable companies to create their own new disruptive market. In the next chapter, we will explore how Fujitsu is engaging with customers to help them do this. Join this journey of creating new disruptive markets with Fujitsu – a bridge to Quantum computing!

Image

 

 

 

Manju Annie Oommen

Dr. David Frith Snelling

 

About the Author:

Manju Annie Oommen

Sr. Manager – Product Marketing

About the second Author:

Dr. David Frith Snelling

Fujitsu Fellow and Program Director Artificial Intelligence, CTO Office, Fujitsu

SHARE

Comments on this article

No comments yet.

Please Login to leave a comment.