Recent data shows that industrial productivity is currently stuck in neutral. According to the US Department of Labor Bureau of Labor Statistics, labor productivity, during the current business cycle (which started in the fourth quarter of 2007), has grown at an annualized rate of only 1.1 percent. In essence, across many industries, a productivity ceiling has been reached. A new wave of industrial edge applications, however, that generate and process manufacturing machine data locally, now present a possible solution to this problem.
Productivity numbers have reached this plateau because traditional technologies are imposing fundamental limits on how effectively workers can perform their jobs in the workplace. Most organizations have reached a productivity limit that will require a different fundamental set of work tools and new, more collaborative work approaches to overcome. Now, the search is on for new ways to improve industrial productivity.
A recent trend which is helping to address this issue is driven by the Industrial Internet of Things (IIoT) phenomenon. New technologies are emerging that focus on capturing and analyzing industrial machine operational and production data. Companies are not only looking for new ways to exploit their own internal data but are also looking towards connecting to supply chain data sources upstream and downstream of their plants. This priority has, in turn, spawned increased interest in leveraging both cloud and edge computing (the processing power that is embedded either inside or close to industrial machines for the purpose of enabling analytics and control automation functions) to help collect, process, and analyze that new-found data.
Business benefits of better data access
Consider how IIoT analytics and industrial edge concepts are being applied to productivity improvements in discrete manufacturing. One of the key components of the “smart” manufacturing process is predictive analytics. IIoT devices and sensors on the manufacturing floor are measuring not only temperature and humidity but energy, motor and drive characteristics, vibrations and other variables. Now, this data can be collected and accurate predictions can be made regarding when machine components will start to fail. This improves the Overall Equipment Effectiveness (OEE) metric which impacts availability, quality and efficiency.
The other big application of industrial edge computing is video analytics. Such technologies have proven highly efficient in quality control environments. Sorting machines, for example can leverage high definition video observation to inspect items on a conveyor belt. These devices, at high speed, can identify which items on the conveyor belt are good or bad. The video analytics-driven inspection quickly answers questions like “Is this the right shape, consistency, finish, weight, length, density, color, or thickness?” Automated pickers take those items that are not meeting the pre-set quality standard down another conveyor belt. Those particular items may get sold as “seconds” at a somewhat reduced price or may find a secondary function or end use (like a group of discolored or undersized apples that would now be used to make juice instead of appearing as whole, shiny, equally-proportioned apples on the grocery shelf).
Local video analytics improves real-time productivity and output, but also uses the visual information gathered to direct changes to the process further upstream so that fewer defects make their way down the conveyor belt. All of these capabilities depend upon the industrial edge hardware and software computing solutions that support the plant. It’s all about speed, reduction of costs, improved quality, and the agility to address shifting consumer demands.
Method of technology deployment is critical
The way that companies deploy and combine these technologies into “hybrid” (cloud + edge) environments will depend upon factors such as where the industrial facility is located (e.g., are they subjected to local data laws and regulations), the criticality of the work being performed (i.e., how much latency can be tolerated when acquiring and analyzing data), the degree of security required, and the short and long term needs of the particular business.
As the deployment of these new technologies requires knowledge of both Operation Technologies (plant-based automation technologies such as SCADA), and Information Technologies (traditional computer-based operations and protocols), it is important to partner with vendors that possess expertise in both arenas. Companies like Schneider Electric, in addition to their deep expertise in process automation, also design and integrate plant cloud and edge computing solutions that can help businesses regain their productivity momentum.
Originally published on SE Blog by Jamie Bourassa.