The explosion of the internet-of-things (IoT), along with sensor, social, and other streaming data associated with Industry 4.0, is driving organizations to use edge computing to provide them with real-time analytics. Edge computing—edge servers that control one or more field devices—provides unprecedented time-to-insight and time-to-action while reducing the amount of data being transmitted to central data centers. Applying prescriptive analytics “at the edge” optimizes operations.
Edge Analytics in Action
Automated embedded artificial intelligence, predictive and prescriptive analytics can all be found in retail, logistics, security and cyber-security, energy, fleet maintenance, manufacturing, insurance, industrial, and other segments. Edge intelligence stems from artificial intelligence-based analytics over data collected from devices located in the edge server’s smart objects and other passive or semi-passive devices—such as sensors and RFID tags—but also from edge computers and routers. As a result, edge processing can still discover intelligence patterns in near real-time at a low latency and without any essential loss of bandwidth.
Edge analytics is appropriate for use in cases involving fast, close to the field processing with multiple devices. Here are several examples of edge analytics in action.
- Performing prescriptive maintenance
Developing maintenance plans based on mean time between failure (MTBF) statistics for reducing costly downtime periods. Improving product quality and process productivity by ensuring an unmatched degree of utilization to serve customers reliably.
- Managing and optimizing operations
Performing real-time anomaly and root cause detection; profiling behavior; improving resource allocation; aiding decisions on productivity and automating routine to optimize smart factory, facility, asset, fleet, and personnel management.
- Reducing risks
Developing risk models, minimizing potential for human bias, detecting fraudulent activities, refining processes with consistent, rules-based decision-making, and continuously monitoring environmental conditions.
- Preventing customer churn
Preventing customer churn by proactively resolving detected performance issues, intelligently learning usage or consumption patterns, and effectively targeting offers and services to keep customers happy.
- Enhancing competitive advantage
Better handling of market challenges in an ever-changing global business. Adapting to windows of opportunity, and scaling and automating processes while reducing costs. Recognizing variations and forecasting responses to changing conditions.
For example, you could use edge intelligence to proactively provide predicted customer churn information to customer success and sales teams to take actions. In service or contract-driven organizations, customer acquisition is expensive. Every lost customer directly impacts your bottom line. Most unhappy customers will never tell you about performance issues. With simple, smart churn detection analytics, you can improve customer experience and retention.
Low-latency decision making is also ideal for remote asset monitoring. Edge intelligence enables monitoring operations to rapidly decide when to take needed actions, to repair or replace an asset. Based on the deployment of edge intelligence in the field, directly on the machine or parts, it is possible to identify asset degradation patterns. This can lead to proactive versus reactive replacement of the part to avoid downtime or disruption in production operations. Edge intelligence also isolates the asset location or causes of problems making it much easier for staff in the field to find and resolve issues.
Another benefit of edge intelligence is the collection of related conditional data. Operators can learn more about situational factors or alterations of an environment that may have caused an issue. That contextual knowledge can be used to better prepare for or prescriptively prevent issues in the future.
At the organizational level, richer datasets can be collected and persisted, including historic datasets from multiple sensors, parts, machines, and edge devices. Hence, learned intelligence can be reused across the enterprise enabling proactive support services such as “Asset Condition Monitoring as a Service” and “Maintenance as a Service.” It also can be continuously updated through measurement of model drift.
The emergence of “smart,” connected things coupled with existing subject matter expert knowledge creates an opportunity for a new era of analytics. Modern analytical tools can not only predict what is likely to occur but also offer “what-if” analysis of alternatives to better guide decision-makers. Edge analytics solutions today augment humans – they do not replace them. It is crucial for humans to guide and monitor these systems.
Prescriptive analytics on the edge is an emerging frontier that has progressed from post-event analysis of historical data, real-time event analysis, and predictive analytics, to the forward-looking automation of actions to optimize operations. Industry leading organizations, such as GE with Predix, are already powering prescriptive maintenance programs. Pilot programs to collect data from smart things has already been completed. Now we are seeing artificial intelligence being embedded along with prescriptive analytics.
By making connected things smarter and moving intelligence to the edge of the network or within the smart things, automated edge prescriptive analytics can drive big wins for enterprises. As companies consider how best to apply this new wave of edge analytics technologies into their business, they are embracing design thinking methodology to reimagine digital processes.
Jen Underwood is Founder and Principle of Impact Analytix, LLC. Impact Analytix is a boutique integrated product research, consulting, technical marketing and creative digital media agency led by experienced hands-on practitioners. Jen can be tweeted at @idigdata