Gathering data & analyzing it from industrial assets on the plant floor registers as cutting-edge not very often. For several years, both the plant floor managers as well as their counterparts have analyzed industrial machine data in order to not only get alerts to production snafus, but also identify quality glitches or as a guide for tweaking assets with an aim to boost performance. This is where
industrial analytics comes in to the picture.
As industrial assets are digitized with sensors & are connected through Industrial Internet of Things, manufacturers are still looking forward to analyzing machine data in order to build production efficiencies, reduce the downtime, & promote better decision making. However, the goal posts have moved significantly when it comes to scale. Rather than a plant manager or even a maintenance worker analyzing historical data from a specific asset in a spreadsheet to make a modest change down the road, the manufacturers at present are striving for wholesale transformation. Their aim is to build flexible as well as intelligent operations where networks of assets and systems can be automated & optimized completely in near real-time.
The engine for intelligent operations
Advanced analytics is the engine for such intelligent operations in areas that include predictive maintenance, real-time quality control, as well as scenario testing for root cause analysis & this is being powered by AI/ML. While the analytics category stays nebulous in the industrial world, the use cases are already sparking huge interest & growth. As per
IoT Analytics, Industrial AI as well as analytics market hit USD 15 billion. Spearheading the transition to Business Intelligence-based practices is the use of predictive analytics, which mines existing data for knowledge that can be used to forecast future events and
industrial analytics market trends.
The increase in the use of industrial analytics is rooted mainly in its desire for capitalizing on the sheer amount of data that is generated by plant floor of field-based industrial assets. This includes robots, automation cells, oil rigs as well as wind turbines. IDC Insights predicts the typical plant is generating more than a terabyte of data/day with expectations that the number will be multiplied by a factor between 5 & 10, based on the industry, over the next 5 years since manufacturers across industries ramp up digitization efforts.
“That’s what’s driving companies to develop or buy these [analytics] tools—without them, there’s nothing else to do with the data,” says Kevin Prouty, group vice president,
manufacturing insights and energy insights at IDC. “It’s too much data for an engineer with an Excel spreadsheet to analyze.”
Optimizing productivity & increasing profits
Most of the manufacturers agree that the bounty of data can be used to the core for bottom line impact, either it is increasing profits, or productivity or both. However, the problem is when the data is scattered across silos, in multiple formats, without any context, & with much of it stored as time-series data. This is not handled adequately by several
enterprise analytics & the big data tools that is designed for structured as well as unstructured data. Furthermore, Excel has been the analysis tool of choice in this space for years, however, it is inefficient & any insights stay isolated to a single engineer tackling one specific problem.
Also, there are huge differences in the approach that enterprise IT & operations Technology groups take for process such as data collection & analysis. Most of the enterprise analytics efforts include ingesting data, normalizing it, as well as putting it either in a centralized repository or data lake, and most likely in the cloud, for making it accessible to numerous business users for various types of analysis. However, from an OT perspective, the data collection & analytics efforts have been usually local & tactical.
“OT people know how to build an analytical model for a single machine in a single factory for a single problem and they do it in Excel or by hand with an open-source platform,” says Marcia Walker, global industry principal for the manufacturing practice at SAS. “When you ask them how they’re going to take that and apply it to all the machines in a factory and to their factories across the globe, they stare at us like deer in the head lights.”
Bridging the gap between industrial analytics & enterprise analytics
Another major gap between industrial analytics as well as general enterprise analytics is the understanding of the nature of the machines & its ability for providing context to the data that is collected in historians & supervisory control & data acquisition systems.
Time series data lacks context for understanding on how the raw data set are related either to a particular process or a condition, unlike structured content from the financial systems. For instance, what factors would be present in order to impact the operation of a pump and without any proper context, it is not possible to completely leverage data for driving operational performance, or predictive or prescriptive maintenance applications.
In a nutshell
There will not be any single analytics product which fits the bill for all use cases & all scenarios given the complexity of factory automation as well as the diversity of industrial data. Rather, experts state that firms should consider the long-term strategy & the questions they are looking to answer in order to line up the right tools for the job.
The Industrial analytics market is anticipated to reach USD 55.3 billion by 2029. The increase in investment in industry 4.0 and surging emphasis on real-time data analysis as well as predictive maintenance, rise in adoption of Internet of Things and Industrial Internet of Things devices are some of the major factors driving the
industrial analytics market growth.
