Digital twins can assist in decision-making to prevent problems before they occur

Federal agency leaders are often tasked with making important decisions on highly complex topics with the information available in short time periods. However, what if technology could aid in decision-making beforehand or even help prevent incidents from happening?

A software technology called “digital twins” can help address these unique and highly dynamic challenges. This technology combines the power of in-memory computing and distributed caching with an updated digital twin software model. This model employs digital twins as a virtual representation of any real-life device or data source, and each digital twin maintains contextual information about its corresponding data source to aid real-time analysis and management.

Hosting digital twins on an in-memory computing platform enables them to run at scale to simultaneously track millions of data sources and provide fast responses. This approach allows real-time analytics to filter large volumes of live data and present actionable results to federal agency personnel and emergency responders. It also boosts situational awareness by supplementing the “silos” of information stored in databases to make insights from analytics instantly accessible and actionable. Due to their ability to run analytics in real-time, digital twins can immediately signal abnormal events and send alerts to personnel.

Simulating real-world scenarios

Beyond just analyzing telemetry from physical devices and data sources in real time, digital twins also can model those data sources in simulation. This enables the creation of workload generators built using a collection of digital twins. By using digital twins to generate telemetry, simulations can assist in developing real-time analytics applications prior to their deployment.

For example, digital twins can model vehicles in a large fleet by generating telemetry that the vehicles will send to real-time streaming analytics, which also can be implemented using digital twins, as previously discussed. Digital twins offer the flexibility to both simulate a data source’s behavior and analyze its telemetry in real time.

Digital twins also can simplify the design of large simulations that model complex systems. Because digital twin models running on an in-memory computing platform can be configured to simulate faster than real-time, applications can model complex systems with digital twins and extract results in advance. This aids decision-making for operational managers by enabling them to create predictions of future events based on parameterizable scenarios.

In practice, here are a few practical applications for real time digital twins and simulations:

• Security Monitoring – In the event of a security breach, intel is critical to understanding the situation and mitigating its impact. Real-time digital twins can give city leaders, security departments and emergency responders the information they need by analyzing data from tracking cameras and motion sensors at key locations to detect intrusions. They can enhance their analysis with contextual information about the locations, such as the presence of key persons to be protected, time of day, event calendars and correlation to signals from nearby devices. For example, digital twins can aid the Capitol Police by monitoring the safety of 1,800+ VIPs across the country and providing instant notifications of unusual sensor patterns.

• Crime Detection – Detecting cloned license plates and convoys of stolen vehicles en route to shipping ports has become a major challenge for law enforcement agencies. Real-time digital twins can analyze license plate telemetry from CCTVs distributed across metropolitan areas and highway systems to detect possible cloned license plates based on contextual information, such as recent locations and details about each vehicle. In-memory computing can correlate analysis results from these digital twins to detect convoys, and this correlation can be enhanced with knowledge about suspected lead vehicles. This technology can alert law enforcement within seconds so that personnel can intercept convoys before they reach their destinations.

• Transportation Safety – The recent series of derailments in our freight rail system has highlighted the need for better monitoring of rail cars to detect issues, like overheated axles, before they become emergencies. Today’s monitoring systems make automated radio alerts to personnel when a potential emergency is detected. By sending telemetry from existing sensors to the cloud for continuous analysis by real-time digital twins, trends can be detected and personnel alerted well before a small issue becomes a major event. An in-memory computing platform hosting digital twins can simultaneously track all rolling railcar axles in the nation to provide enhanced alerting that the rail system lacks today.
• Disaster Recovery – If a major fire, earthquake, tornado or other natural disasters occurs, federal and city response leaders need to track emergency supplies to ensure timely delivery to locations where they are needed. Real-time digital twins can keep track of thousands of supplies across the country, the possible delays for shipping, emerging traffic delays and more. They can also detect emerging forest fires and predict their motion towards towns and structures, so that firefighting assets can be mobilized, and communities warned.

Even the best data and intelligence professionals in federal agencies can’t humanly process millions of fast-moving data sources in real time. Technology is the key to spotting abnormal events and trends in large streaming data sets in real time to make decisions that may impact our nation. The above are just a few examples of how digital twins can help federal leaders. There are endless possibilities for employing this technology to help improve decision making and lower response times so that we can mitigate security breaches, crimes, natural disasters, cybersecurity threats and more.

William Bain is CEO and Founder of ScaleOut Software.

Copyright © 2024 Federal News Network. All rights reserved. This website is not intended for users located within the European Economic Area.