In the age of the Industrial Internet of Things, the physical world is converging with the digital world at a rapid pace. This brings unprecedented availability of data and data processing technologies to bear on industrial operations. Using sensor-based data, you can ask how your equipment is performing or which parts will need maintenance soon. This kind of insight is extremely useful and valuable (and only scratches the surface of what’s possible).
Our mission is to transform data into actionable insight that helps you make better, fact-based decisions for greater worker safety, operational efficiency, and revenue.
The key is of course in data. Today, you have thousands and even millions of tags, and more are added with each new piece of equipment. The sheer volume of tags can no longer be managed manually.
Additionally, you want to do more with data. Analyzing tag vs. tag is no longer the standard. You seek to compare assets fleet-wide, or compare start-up sequences and similar events. You want to perform predictions like the way iPhones predict the best time to leave for home to beat the traffic. Data should work for you, rather than you constantly working on the data.
When I was a DCS Technician, Historian Administrator, and Data Analyst, I mined data for answers, wrangled chaotic data sets and naming systems. Consulting for industrial companies, I saw many struggling with decades of compounded inconsistencies. Data has become dispersed and hard to use. It’s even hard to compare data from similar assets. In working with data, you spend 90% of your time wrangling data into a manageable form and 10% actually gleaning insights.
We flip this. It is crucial for technology to do the heavy lifting and automate data preparation legwork so you can shift from wrangling data to having data that is continuously cleaned, organized and rendered usable. It’s a new way of experiencing data and one that will empower the entire industrial organization.
Ready-to-go data is a huge enabler — it can be readily used by data scientists for analysis, readily aggregated for fleet-wide views by site operators, and readily consumed by analytics applications for visualizations and predictive modeling. If SCADA systems, data historians and sensors make up data infrastructure, then everything that goes into preparing data for analytics is analytics infrastructure. The abstraction of analytics infrastructure will bring self-serve data to the industrial world.
“The abstraction of analytics infrastructure will bring self-serve data to the industrial world”
Cloud makes computing infrastructure easy to provision (Infrastructure-as-a-Service, IaaS), as a result our applications are easy to deploy (Platform-as-a-Service, PaaS) and we can use software applications as a service (Software-as-a-Service, SaaS) — no need for managing a lot of IT, it just works. IaaS, PaaS, and SaaS have revolutionized the IT world. They empower people to do their jobs better and faster, and generate over a hundred billion in annual revenue. The idea behind these “as-a-service” layers is that you have exactly what you need, without having to worry about how it got there. It’s self-service.
The same can be accomplished with industrial data. You can have data ready for analytics at your fingertips, without worrying about how that data came to be collected and transformed into a consistently clean, organized, and usable form. At Element Analytics, we bring you closer to self-serve analytics. We sit between and we bridge the data infrastructure and data analytics layers. That is, we take collected data, transform it into a usable form, perform analytics and connect it to data applications that help you visualize insights relevant to your industrial issues. We believe this middle layer—the analytics infrastructure layer—is what will drive industrial analytics forward.
“You can have data ready for analytics at your fingertips, without worrying about how that data came to be collected and transformed”
Analytics is only as good as your data. This fundamental concept is reflected in the name of our company. We’re called “Element Analytics” because a big part of our focus is foundational aspects of analytics, what must be done in order for data to enable everything above it. We make data an enabler of your Industrial Analytics Journey from descriptive analytics with insight into what has happened, to predictive analytics with foresight into what will happen in the future.
Predictive analytics will fundamentally change the way your industrial organization operates. For example, predictions for turbine failures and associated costs enable you to see, prioritize, and complete fixes ahead of time to prevent failures. Once implemented by your service teams, this predictive maintenance reduces operational costs while improving asset uptime and performance. Prediction is not a data stream passing a threshold, triggering an alarm to prevent an issue — that’s still reactive. Predictive analytics provides actionable foresight based on the probability of anomalous events occurring in the future, which is proactive. The shift from a reactive to proactive is especially helpful during the Great Crew Change where the next-generation industrial workforce needs data to guide decision-making, since they don’t have the 30+ years of experience to make intuition-based decisions.
“Predictive analytics will fundamentally change the way your industrial organization operates”
By showing what will happen in the future, foresight helps you become more informed and better able to make faster, fact-based decisions ahead of time to shape the outcomes you want and need. Over time, predictive analytics helps your industrial operations consistently outperform, giving your organization a sustainable business advantage.
We’re excited to be here and share with you our thoughts, ideas, and updates. This blog is a direct line from us to you. It reflects our diverse perspectives and will cover topics from design to engineering to data science to culture and anything else that’s on our mind. Welcome to our blog!