Simulation with digital twins helps decision-making for large systems

What if a telematics application could track telemetry from every vehicle in a fleet and instantly analyze it to identify problems such as missing or unstable drivers or new mechanical issues? What if you could continuously track the progress of passengers onboard and proactively respond to delays and cancellations to reduce stress and ensure smooth operations? What if railway operators could detect

Such applications simultaneously track the dynamic behavior of many data sources, such as IoT devices and sensors, to identify problems (or opportunities) as quickly as possible and provide operations managers with the best possible It should provide situational awareness.of ScaleOut Digital Twin Streaming Service Enables building streaming analytics applications that address challenges. The new release of the service also adds the ability to run these applications in simulation, both for testing under synthetic workloads and for modeling complex interactions.

A software “digital twin” model simplifies application development for both streaming analytics and simulation. Digital twins also provide the building blocks needed to decouple application design from the orchestration of large-scale deployments with thousands of entities.

Simulate a streaming analytics workload

To simulate a large number of data sources sending periodic telemetry messages, developers build digital twin models of a single physical data source, such as vehicles in a fleet, and run thousands of digital twins. to generate telemetry for all data sources. By acting as a workload generator, you can test streaming analytics applications (such as telematics applications) that run in simulation. This can also be implemented with a digital twin. Once the analytics code is validated, developers can deploy it to track live systems.

Many vertical applications can benefit from simulation of streaming analytics. For example, digital twins simulate perimeter devices that detect security intrusions in large infrastructures to help assess how well streaming analytics can identify and classify threats. You can also model rail cars in a nationwide rail system and validate streaming analytics that track mechanical issues on each rail car and alert engineers before derailments occur.

Simulate large systems with many entities

To aid in operational planning and decision-making, digital twins can also model thousands of interacting entities within large systems. For example, you can implement an airline simulation that includes thousands of airline passengers, aircraft, airport gates, and the air traffic sector. These digital twins maintain state information about the physical entities they represent, execute code at each time step of a simulation run, and exchange messages that model interactions. Simulation updates the state of the digital twin over time to track the outcome of interactions and provide insight to operations managers.

For example, airline simulations can measure the impact of flight delays on gate congestion or changes in passenger itineraries. In practice, airlines can use such simulations to model weather delays and system outages (such as ground stops) and evaluate alternative scheduling decisions to respond to these situations. Simulations run faster than real time and can provide predictions that help live system administrators make decisions.

Easily scale your simulation

ScaleOut Digital Twin Streaming Service uses scalable in-memory computing technology to provide the speed and memory capacity needed to run large-scale simulations on many entities. It stores the digital twin in memory and automatically distributes it across the cluster of servers that host the simulation. At each time step, each server executes the simulation code for a subset of the digital twin and determines the next time step at which the simulation should run. A streaming service coordinates the progress of simulations on the cluster, advancing simulation time at a user-selected rate.

Building simulation models using digital twins provides a clean separation of application code from simulation orchestration. Streaming services can host large-scale simulations and utilize as many servers as needed to run at maximum throughput. New servers can be added while the simulation is running, and any server outages can be handled transparently. Developers can simply focus on building digital twin models and deploying them to streaming services.

new path mapping

Digital twins have historically been used as a tool for modeling the detailed behavior of single complex physical entities such as jet engines. ScaleOut Digital Twin Streaming Service takes digital twins in new directions. That is, simulation of large systems with many interacting entities. ScaleOut Software’s highly scalable in-memory computing architecture makes it easy to simulate thousands of entities and their interactions. This provides a powerful new tool for extracting insights on large systems with complex behavior and provides important new analytical and predictive capabilities for operations managers.

(photo courtesy Donnie Jean upon unsplash)

tag: digital twin, Non-standard