How to drill a better hole with analytics
Introducing 4 newly patented machine learning and IoT technologies from SAS
By: Alison Bolen, SAS Insights Editor
Who knew that analyzing IoT data could help discover a way to drill a better hole? Or a faster way to stop healthcare fraud? At SAS, these are just a few of the new technologies we've patented for analyzing real-time data and streaming data from the Internet of Things. Let’s hear more about them from the inventors.
1. Oil and gas: How to drill a better hole
What if you could optimize the speed of drilling a deep hole in the ground depending on surface materials and other environmental factors? Using SAS, data scientists discovered a way to feed data back to the drilling machine as it drills, automatically adjusting drilling speeds to increase efficiency and reduce breakdowns.
David Pope, Senior Manager of Pre-Sales Support at SAS says, “This patent is tied to IoT, analytics and the SAS® software. Without any one of those things, this patent would not have been possible.” Pope shares the patent with three other colleagues at SAS.
“It’s neat to be recognized for solving a problem that nobody else has ever thought of,” says Pope. The patented technology uses SAS to analyze the environment and optimize the speed of a drill bit when breaking through rock and deepening a borehole.
Pope holds 12 patents overall and works as a mentor to colleagues who share his interests in innovation.
- Patent name: Control Variable Determination to Maximize a Drilling Rate of Penetration
- Inventors: Moray Laing, David Pope, Keith Holdaway, James Duarte
- Key technologies: IoT, analytics, event stream processing
- Primary industry: Oil and gas
2. Financial services: Reducing risk in banking and beyond
Handling streaming data is still a new concept to a lot of applications that were designed to analyze data that gets processed by the system as a single static batch. Instead of rebuilding all of those systems, what if there were a way to simply trick the system into thinking a new streaming data source was actually a static one.
Katherine Taylor, Senior Data Scientist at SAS, has been issued a patent for a system that does just that. While she initially developed the streaming data analysis method to enable real-time risk calculations at a large bank, she soon realized the same method could work for any system receiving streaming data.
“It was developed to help banking, but the patent itself used electricity load management for its example. Any industry can use this and benefit from it,” explains Taylor. She’s been issued two patents, has a third one accepted and a fourth one pending.
- Patent name: Management of Real-Time and Historical Streaming Data
- Inventor: Katherine Taylor
- Key technologies: Data streaming, distributed file storage, in-memory analytics
- Primary industry: Cross industry
Who else is innovating with analytics?
From curing diseases to studying the sun, innovative uses of analytics are changing the world. Download this e-book to see where tomorrow’s technologies are happening today.
3. Manufacturing: Monitoring complicated systems and processes
Today’s manufacturing facilities are loaded with complicated, interconnected systems and processes. When one small piece of the system fails or slows down, the downstream effects can be monumental.
A new system collects data from these complex systems, as they’re operating, to prevent failures and optimize processes. This helps manufacturers meet day-to-day and year-to-year goals.
“Our method is an improvement on what’s been done before in manufacturing, and it’s perfect for IoT data,” says Anya McGuirk, Distinguished Research Statistician Developer at SAS. Her innovation is designed to analyze data from more than a thousand sensors that send data at sub-second rates from many different parts of the plant or facility.
This newly patented method helps detect when these machines or processes are starting to fail. It identifies a shift in the process mean or a change in the process variability. “Our method is not just an improvement on previous methods," says McGuirk, "But perhaps more importantly, it is easier to implement and especially well-suited for cases where data reads are coming in at a high rate-IoT type data."
- Patent name: Monitoring System Based on a Support Vector Data Description
- Inventors: Deovrat Kakde, Sergy Peredriy, Arin Chaudhuri, Anya Mcguirk
- Key technologies: Analytics, IoT, event stream processing
- Primary industry: Manufacturing
Our method is not just an improvement on previous methods, but perhaps more importantly, it is easier to implement and especially well-suited for cases where data reads are coming in at a high rate—IoT type data. Anya McGuirk Distinguished Research Statistician Developer SAS
4. Health care: identifying risky claims in real time
When fraudsters file fake health care claims, it doesn’t only cost the insurance company, it raises the cost of insurance for everyone. But catching fraudsters in the act can be challenging and costly too. Their schemes are constantly changing, and the complexity of medical billing practices gives fraudsters a lot of leeway for making false or overinflated claims.
Emily McQuiston, SAS Consulting Manager, devised a new algorithm that detects fraudulent claims almost immediately. Using real-time processing and machine learning, the new technology can identify suspicious claims as soon as the claim is filed. “We work directly with clinicians and understand the difficulties they face. We want to help them improve fraud detection and make it easier and faster,” says McQuiston.
This new technology is just one way SAS is using machine learning to improve decisions in health care. “We continue to innovate in the realm of helping clients in the health care industry improve quality, process and administrative efficiencies using machine learning and plan to submit patents for those innovations” says McQuiston.
- Patent name: Techniques to provide real-time processing enhancements and modeling for data anomaly detection pertaining to medical events using decision trees
- Inventors: Steven Enck, Emily McQuiston, Daniel Kelly
- Key technologies: Machine learning
- Primary industry: Health care
The future of innovation at SAS
Overall patent numbers across the industry show that artificial intelligence and the Internet of Things remain two of tech’s hottest areas for innovation.
Likewise, many of SAS’ recent patents have focused on AI and IoT. And if the responses from the patent holders interviewed here are any indication, there will be many new inventions coming in these and other areas from SAS.
“I’ve recently changed roles to focus on artificial intelligence and machine learning,” says Taylor. “There are still a lot of opportunities to innovate and improve on methods in this field.”
McGuirk is excited about how these and future innovations will change the way problems are solved. “We’re not just implementing methods that people have come up with before but we’re actually innovating and pushing the envelope,” she says.
Recommended reading
-
Fraud detection and machine learning: What you need to knowMachine learning and fraud analytics are critical components of a fraud detection toolkit. Discover what you’ll need to get started defending against fraud – from integrating supervised and unsupervised machine learning in operations to maintaining customer service.
-
Model risk management: Vital to regulatory and business sustainabilitySloppy model risk management can lead to failure to gain regulatory approval for capital plans, financial loss, damage to a bank's reputation and loss of shareholder value. Learn how to improve model risk management by establishing controls and guidelines to measure and address model risk at every stage of the life cycle.
-
frtb: a wait and see strategy could be riskyFRTB, fundamental review of the trading book, is a regulation that changes how banks analyze market risk in the trading book to address systemic challenges.
Ready to subscribe to Insights now?
Subscribe to Insights newsletter