Blog Post
Analyzing Big Data

Analyzing Big Data

Stephen Reilly, ProSys

In the January / February 2017 edition of the ISA’s InTech magazine, Michael Risse writes about the subject of big data analytics, noting that as “generating and storing data has gotten cheaper, the quantity of data has grown. The numbers are staggering, with more data stored in just minutes now than during multiple-year periods in the 1960s.” He cites how manufacturing “is far and away the leader, with 1,812 petabytes of data produced: 1,072 from discrete manufacturing and 740 from process manufacturing.” It makes sense that the manufacturing industry would have the largest volume, as “manufacturing organizations also have the longest history of generating and storing large volumes of data.” However, as Mr. Risse notes, they “are considered as laggards in exploiting big data technologies.”

So why is that the case? Mr. Risse writes that “big data implementations can be expensive and resource intensive, and…”these solutions do not fit the needs of front-line engineers or analysts within manufacturing plants.” Incorporating big data technology with plant automation systems that were set in place decades ago and are still running also challenges vendors to bridge the gap to existing infrastructures. Some people are simply confused about the premise, thinking big data just means applying advanced process control, statistical process control, etc.

Luckily, there are innovations in the field overcoming those challenges. With an actively competitive cloud-computing market and open source software driving down costs, an increased range and depth to algorithmic approaches, improving data organization techniques, and more flexible analytic models unifying various data sources, it is now easier than ever before. And even if there is a perceived gap between the amount of data to review and the number of engineers available to sort through it, there are also software innovations in place to help complete the task, such as collaborative, web-based accessibility with lightweight deployments showing an interactive, easy-to-interpret representation of data and results.

With that in mind, one should note ProSys’s own take on the big data analytics behind alarm metrics in its EventKPI application. It can handle multiple control systems and analyze various measures by the minute, hourly, daily, weekly, monthly, and even yearly basis! Its reports can be presented in several ways and can be accessed by all levels of management for a company’s operations. And while some reports are standard for any EKPI installation, others can be very-easily configured by an end-user with minimal training. Thus, if your company has shied away from big data before, consider contacting ProSys to see how we can meet your big data needs today!