Does your company have a big data strategy? If they don’t; read on because NCR is imagining the unimaginable and finding real business value in big data analytics. If your company does have a big data strategy – stop reading here and now. No, seriously, keep going because NCR is at the forefront of the super-hyped world of Big Data. NCR is taking telematics data from devices globally and performing predictive analytics; NCR fixes devices before they fail. Because of analytics they have the right technician with the right part at the right time so the downtime for the device is planned. If you’re in manufacturing or in retail you KNOW how important that is.
This isn’t the end of the story, but it’s pretty far down the road, so let’s back up a bit.
The Teradata Customer Engagement Team sat down with NCR’s Brian Valeyko, Director of Data Warehousing, Business Intelligence and Big Data Analytics. Brian explained how NCR’s big data strategy has enabled them to find the right people, technology and processes to change their business.
“We provide services to many different customers worldwide and being able to mine that data effectively so we can support that business and do a much better job at it than anyone else is hugely important to our business.” – Brian Valeyko
A self-described global tech company, NCR commits to “making everyday transactions easier” – all 300 million of them. Every second NCR and its global customers perform 3500 transactions with ATMs, POS and self checkout in retail stores, on mobile devices and kiosks in every industry. NCR is getting their big data telemetrically from the Internet of Things. So, under the strategy of “Imagine the Unimaginable” NCR’s Big Data team (that’s right team) set out to find the right technology for Big Data.
“The way we built the case for big data was we listened to our Teradata representatives. We listened to the individuals who help us every day with the business that we do and we've read the stories, we understand the possibilities, and we've taken advantage of that knowledge and relationship that Teradata has provided to us so we're able to understand how Hadoop and Aster technologies help us. We were able to understand that we should pick Hortonworks Hadoop because it allowed us a much quicker time to market without having to train as many people to leverage that standardized release of a Hadoop package. Super important to us to be able to do that. So, I think, we're listening to the advice we're getting from Teradata and making sure that we take advantage of that and use it to enhance our business.” - Brian Valeyko
NCR employed Unified Data Architecture with Hadoop, Teradata Aster and its existing Teradata warehouse and started the project; predictive analytics on all those devices. The team took a process for creating algorithms to allow NCR to send the right technician with the right part before the device actually failed. Previously NCR analysts manually created these algorithms – it took 6 months. With the Aster platform NCR analysts reduced that time to 3 weeks!
“We were able to reduce that time to having an effective algorithm to less than three weeks and that's from harvesting new data, to creating the algorithm, to testing it, to releasing into our production application. So, that's a huge improvement. The other thing that's very interesting about that is not only did it improve the speed to market for that solution; it was also much more effective at finding those faults. So, we were able to find a much higher percentage of the actual faults than we had done manually in the past. So, that kind of success breeds a lot of other success as well. It gives us the opportunity to expand that capability beyond a single instance, a single product. If we can do that for every product that we will release we will have saved a lot of time and effort on our part and improved the service to our end customers.” – Brian Valeyko
NCR’s team is also able to better trace why device failure may occur - is it a manufacturing supplier problem? An installation training problem? Or a maintenance problem?
The team is combining the telemetric data with other data from multiple lines of business. Now in many cases NCR doesn’t have to send a technician at all- they send the right commands and the right components and are able to fulfill customer agreements with zero unplanned downtime.
This also allows NCR to cross reference supplier parts and commonalities with their many business units. The biggest benefit – NCR no longer looks at supplier parts within siloed business units but rather they look and react to supplier parts as a whole across the global enterprise. This affects supplier contracts and inventory levels.
Brian tells us they are still at the beginning of this Big Data journey. If the journey is a ‘lifecycle,’ NCR is barely a teenager. But wasn’t that the time of life when we learned and experimented the most?
Thank you, Brian, and NCR for sharing your experiences with Big Data and for “Imagining the Unimaginable.”
Want to learn more about exactly how NCR's Big Data team made their case for UDA as their big data strategy. Below is NCR's Brian Valeyko... in his own words.
Latest posts by TCSET (see all)
- RBC: Doubling Success & Earning the Right to be the Client’s First Choice - September 17, 2015
- LCL: Meeting & Exceeding Customer Expectations with Integrated Data & Analytics - September 4, 2015
- Dell: Maximizing Spend, Efficiency & Profit Globally with Project “MaSCoT” - August 10, 2015