The powerful fix of Big Data and IoT
Big data and the Internet of Things (IoT) may not seem an obvious couple at first glance, but they could in fact be a perfect match to handle the deluge of information that is coming our way. Improving product development, the supply chain, customer care – and also helping drive predictive and pre-emptive maintenance.
IoT is still relatively immature, but offers numerous opportunities for innovation, whilst big data brings with it many great prospects for technology development. Analytics tools, for example, have been created to handle and filter large volumes of data and then extracting insights from massive data sets.
One often overlooked, but potentially huge growth market is predictive maintenance. Maintenance analytics is forecast to generate US$24.7 billion in 2019, driven by predictive maintenance and IoT, according to ABI Research.
“Today, predictive maintenance is one of the commercially readiest forms of M2M and IoT analytics, possibly second only to usage-based insurance. It helps asset-intensive organizations transform their maintenance operations and eliminate waste, reducing costly downtime. Infrastructure, vehicles, and industrial equipment can all benefit from it,” says Aapo Markkanen at ABI Research.
The real-time data challenge
The digital universe is growing fast. By 2020, the data we create and copy annually will reach 44 zetabytes, or 44 trillion gigabytes, according to IDC. And all of this data must be made sense of in real-time and put it to good use.
Until now mining big data in real-time has been the preserve of large enterprises with big budgets. But real-time big data analytics tools are now going mainstream, enabling companies of all sizes to track customer behaviour and react quickly to transform both the business model and the customer experience. Talend 6 data integration platform is a prime example, offering native support for Apache Spark data streaming and enabling any size company to convert big data and IoT sensor information into immediately actionable insights.
“For years large web companies have been turning real-time data management and analytics into a major competitive advantage, with Apache Spark and intuitive integration platforms like Talend 6, the playing field is about to get a lot more even,” says John Tripier, head of business development at big data company Databricks.
Revolutionizing predictive maintenance
Predictive maintenance is a key application for big data and IoT, since together they give the ability to accurately diagnose and prevent failure in real time – a major boon to a host of vertical markets and vital to critical infrastructure applications.
Preventative maintenance is costly, and equipment failure causes expensive downtime. Typically technical staff are deployed to carry out routine diagnostic inspections and preventative maintenance according to fixed schedules, but this is costly, labor-intensive and does not guarantee that there won’t be a failure between inspections.
Predictive maintenance also helps reduce operating and capital costs by mapping out proactive servicing and repair of assets and pre-empting possible failures, whilst enabling more efficient use of repair resources. CGI and Microsoft, for example, have joined forces with Thysssen Krupp Elevator to create a predictive maintenance solution for more than 1 million elevators under maintenance worldwide. By connecting elevators to the cloud, gathering data from sensors and systems, and transforming the data into valuable business intelligence (BI), ThyssenKrupp is vastly improving its maintenance services. Now, instead of just reacting to a failure alarm, technicians can use real-time data to define a required repair even before a breakdown occurs.
Predictive maintenance is also being championed by likes of GE, Bosch and other large and increasingly software-centric manufacturers. But its key enablers are the technology suppliers that allow customers to employ similar approaches regardless of their OEMs, according to ABI Research. In this group are horizontal analytic vendors of various sizes, ranging from BI giants (SAP, IBM) to nimble start-ups such as RapidMiner and Blue Yonder. Predikto and Mtell are examples of vendors carving a niche for themselves by specializing in predictive maintenance in specific verticals.
Mtell’s Conscious Monitoring Agents have been designed to prescribe precise maintenance intervention that protects the machine and its ecosystem. In upstream drilling in the oil and gas industry, for example, Mtell’s solution reduces drilling downtime, avoiding delays in completion, rig drilling costs and production deferrals. Predikto predictive analytics tools can turn equipment-generated data into actionable predictions to improve operations, velocity and asset reliability. It is used in a number of vertical markets including the railway and aviation industries. Examples of its use include predicting which train doors on commuter trains will fail for a Canadian rail company and providing a three hour potential warning for the failure of an Air Cooled Condenser (ACC) at a power plant in Saudi Arabia.
Shaping the future
The idea of prediction in technology is nothing new. What is changing is the massive amount of data being gathered by sensors within IoT which makes continual prediction possible without any manual intervention. Predictive maintenance is relevant to all major industries as it has the power to drive efficiency by providing higher levels of safety and quality at way below current costs, whilst improving customer satisfaction.
Big data and IoT will be at the center of the biggest opportunities around connected innovation. “Analytics is where much of the money in IoT will be ultimately made. This means that application platforms need to facilitate big data if they want to gain a competitive edge,” commented Dan Shey, practice director at ABI Research.
Learn how to mine your treasure trove of data here.