Evaluate Weigh the pros and cons of technologies, products and projects you are considering.

Data context vs. content in an IIoT environment

There is a tremendous amount of raw data being generated from machines, machine sensors and robots on the factory floor as part of the new industrial internet of things revolution. Some of this data will have intelligence and value that can improve operational efficiencies, foresee maintenance requirements and deliver smarter and faster business decisions. But most of it will be somewhat wasteful, not that interesting and certainly not worth saving for any extended period of time. The challenge that many industrial companies face — or will soon face — from their IIoT systems is how to extract value and intelligence from machine data?

Data content vs. data context

To answer this question, it is important to understand the difference in data between content and context relating to an IIoT environment. Automated production machinery and built-in sensors record endless hours of unfiltered operational machine data. That entity of recordings is content. There is a lot of information within the recordings, but nothing that can extract value, or intelligence, or enable process improvement decisions.

From the data that a machine collects, wouldn’t it be valuable to know more about the machine’s operation and whether it has deficiencies, is running off-center or requires maintenance or immediate attention? How about information that includes peak hours of machine use, associated yields, product rejects, operator effectiveness, congestion areas, material use estimates and so forth? These entities are context. Interesting or valuable data which needs to be stored for further use, analysis or mining for the unexpected, at some point in the future.

As this raw content resides in machines, sensors or robots at the edge of the network, compute processing is required to add that intelligence to the data. As data changes, more real-time processing at the edge of the network will be required. Historically, humans wrote algorithms in an attempt to transform the data from content to context. These simpler, fixed algorithms were processed over a small data stream, at near real time, local to the raw data in the devices. Nowadays, the sensor-created data streams are enormous and high fidelity.

Artificial intelligence is an example of massive real-time processing at the edge, enabling machines to perform human-like tasks. The in-machine sensors read, compare and physically map machine or robotic data to its environment, and include analysis and intelligent algorithms that look for patterns in the data, and will alert operators to anomalies and opportunities for process improvements that can save a manufacturing operation significant time and money.

Machine learning is the “teacher” of localized AI and developed from learning patterns within very large data sets so, when applied to machines or robots, will analyze behavioral machine patterns and interpret real-life operational scenarios that the machine can learn from. The more it learns, the more it can improve the localized AI algorithms to be even more accurate and effective.

IIoT edge-to-cloud storage strategies

The abundance of data generated from IIoT systems and AI-supported machine applications is creating new challenges for industrial operations. Either the systems respond to developing situations or review and analyze historical data, looking for areas where the process can be improved so the artificial intelligent agents can be trained to monitor the system. The ability to respond to real-time data changes requires that the data be immediately accessible and locally available (edge storage), while data worth saving for future use, analysis or process training purposes will be moved to the cloud.

Data analytics at the edge has become more of a requirement as well, as the growth of “smart video” in today’s surveillance systems creating business value and intelligence. Once placed in a commercial or retail setting, the smart surveillance cameras can perform real-time analytics that can recognize customer facial expressions, identify the number of people in the store at a given time, where they go, how long they stay, the effect of sign placements and a host of other possible options. The analysis is performed locally at the edge, in real time, as the results reside in the cloud for archival, or deleted and re-recorded.

In order to generate content locally that will lead to valuable context, either the compute and storage elements will reside directly within a sensor, and sometimes as part of the machine, or the data will be sent to a local wired or wireless network, using an edge gateway but located on the production floor. The IIoT storage strategy is to not funnel all of the data to the cloud, but instead, use a combination that stores data locally at the machine-level, as well as an edge gateway at the factory-level so that data can be aggregated locally, not exposed to the outside world, analyzed at the edge for intelligence and translated into a common cloud format for long-term data storage.

Forward-looking statements: This article may contain forward-looking statements, including statements relating to expectations for Western Digital’s embedded products, the market for these products, product development efforts, and the capacities, capabilities and applications of its products. These forward-looking statements are subject to risks and uncertainties that could cause actual results to differ materially from those expressed in the forward-looking statements, including development challenges or delays, supply chain and logistics issues, changes in markets, demand, global economic conditions and other risks and uncertainties listed in Western Digital Corporation’s most recent quarterly and annual reports filed with the Securities and Exchange Commission, to which your attention is directed. Readers are cautioned not to place undue reliance on these forward-looking statements and we undertake no obligation to update these forward-looking statements to reflect subsequent events or circumstances.

All IoT Agenda network contributors are responsible for the content and accuracy of their posts. Opinions are of the writers and do not necessarily convey the thoughts of IoT Agenda.