Source: Advanced Manufacturing
For many years, the knowledge economy and traditional industries like manufacturing appeared to be on parallel tracks. Now those tracks are converging.
Innovations like sensors and machine learning that are part of the Internet of Things (IoT) movement coupled with dramatically falling prices for data storage and the ease of cloud computing have prompted manufacturers to overhaul their standard ways of doing business. This combination—the Industrial Internet—has the potential to deliver up to $11.1 trillion a year by 2025, GE says, noting that B2B solutions would capture about 70% of this.
Predictive maintenance, quality control, resource optimization, supply chain management, remote monitoring and asset tracking are use cases that are spearheading the adoption of the Industrial Internet. The application of data has shifted focus for many manufacturers from raw numbers. These days, instead of trying to create more, manufacturers are trying to be smarter about the way they create.
As we all know, what gets measured gets managed. Since the birth of the Industrial Revolution though, much of the gritty details of manufacturing have been opaque. How often machines wear out, how much energy and resources each machine uses and the source of small problems that blow up into big ones—these have all been addressed in roundabout ways. Machines were put on maintenance schedules that may or may not have been effective. Fuel and electricity use was monitored but not fully understood, and so forth.
Increased data about operations has the potential to dramatically change all of this. Real-time machine data can help shed light on a facility’s energy use. That can help cut costs by targeting excess use and allowing plant managers to set informed and achievable targets for energy savings.
Unprecedented traceability at the unit level also identifies inefficiencies and potential problems at an early stage. As a recent McKinsey report illustrated, such data can also be visualized to spotlight both problems and opportunities. Plant managers can test hypotheses in real time to determine the root causes of a yield drop and focus on the most statistically significant factors for further exploration.
Overall, such data provides both a micro and macro view of plant operations.
In another real-life example, a steel manufacturer valued at $2.5 billion was able to centralize machine data; compare real-time and historical overall equipment effectiveness (OEE) and production visibility; set up real-time alerts and notifications to improve real-time monitoring of machines and then drive an accurate schedule for predictive maintenance.