IIoT: Profiting from data analytics

Given the depth and breadth of digitalization in both discrete and process manufacturing, many companies have decided to adopt a gradual approach to the first phase of an IIoT implementation. With individual POCs or by selectively upgrading existing plant and machinery with smart sensors, manufacturers can assess the value of IIoT analytics for a specific use case while building justification for subsequent investment across other business units.

Once production has been upgraded to be IIoT-capable, and plant and machinery now stream operational performance data continuously, manufacturers can transition from a reactive to a proactive approach driven by predictive maintenance. By monitoring the machine pool, production output, and plant integrity in real time, performance degradation becomes clearly visible before a critical resource fails.  Repairs can also now be scheduled based on the actual condition of the piece of plant or machinery instead of estimating repair needs based on the mean time between failures (MTBF).

Thanks to the continuous monitoring of even the slightest variance in temperature, flow rate, pressure, and other parameters, technicians and other responsible personnel can detect and avoid production defects, and even make in-process adjustments during batch manufacturing. With real-time data visualization and analytics, faults are identified as soon as they materialize.

A single faulty manufacturing process can impact hundreds or thousands of products every minute, threatening to reduce first pass yield and drive up costs per unit. Fortunately, real-time alerts can be configured to notify personnel when certain thresholds, tolerances, or other specifications are exceeded, allowing them to take corrective actions to reduce output variation and limit non-conformance.

Manufacturers are now able to optimize the production lifecycle with measures ranging from increasing plant uptime, yield, and throughput to reducing scrap, waste, and re-work. However, comprehensive collection of production data only translates into efficiency gains if steps are taken to exploit the value in this data.

Smart sensors can be used to collect large quantities of data such as temperature, weight, timestamps, etc. in real time. The continuous analysis of this data is able to identify subtle fluctuations in plant processes and machinery that could cause persistent quality problems, production losses, or even equipment failure.

While much of IIoT focuses on productivity and efficiency gains, data from machinery and plant can also be used to improve safety and reduce the risk of workplace injuries. Operators and technicians can utilize this machine-to-machine (M2M) data to identify the location, duration, and frequency of unsafe working conditions within the facility, and take appropriate precautions to ensure compliance with occupational safety and health regulations.

In production, fault codes, time stamps, and system alerts can be correlated across machines on the factory floor to provide clearer insights into the root cause of work stoppages, equipment failures, and safety-related shutdowns.