Digital Twins Bring Value to Big RFID and IoT Data

PACCAR is using a digital twin model to manage the maintenance and repair of some engines, by creating a virtual version of an engine based on sensor data from the real-world versions.
Published: April 2, 2018

Radio frequency identification (RFID)- and Internet of Things (IoT)-based systems (wireless or wired) have been collecting location and sensor data for years. However, an ongoing challenge has been how to manage the growing volume of accumulated data. In some cases, the solutions to manage that data have lagged far behind the hardware capabilities to collect it.

One way in which companies are now making use of the data from RFID, real-time location system (RTLS) and IoT solutions is to create replicas of real assets—known as digital twins—so that they can be measured against them, or put to the test, virtually. The twin concept has resulted from the wealth of data generated by the wireless transmission of sensor and location data regarding things and people.

Bsquare’s David McCarthy

By leveraging a digital representation of a physical asset, users can gain intelligence that can help them to test behavioral responses to conditions, or to predict failures. It’s the next step, some analysts say, to understanding and managing the vast volumes of data that come from sensors via RFID transmissions or other forms of wireless communication.

Typically, digital twin technology is most recently being employed for better predictive maintenance and repair, though it could cross multiple industries and market sectors, says David McCarthy, the senior director at IoT software company Bsquare, which offers a solution to customers in the industrial sector known as DataV. Bsquare has focused for several decades on bringing intelligence to physical assets, first as machine-to-machine data. Throughout the past year, the firm’s customers have used its DataV software stack to predict failures and capture data-driven diagnostics.

A digital twin can be used to set up baseline performance expectations and real-time comparisons against other devices, McCarthy explains. By creating this virtual device, based on sensor data from the real things, users can better understand how their equipment should be performing—and how it actually is performing. This enables the users to accurately predict when maintenance may be necessary, when a failure is imminent and what conditions are most favorable for a device’s operation.

PACCAR is using Bsquare’s solution to create digital versions of its equipment in order to create repair scenarios. The company can create a master twin against which real engines or parts can be compared. The data is being collected from sensors applied to engines. As information is received in the DataV system, it is correlated and compared against the conditions under which the engine may operate. A typical engine, under specific conditions, can then be created in the software.

PACCAR can then manipulate the information to learn how functionality could vary based on conditions, use or other factors. It could, for instance, vary the conditions for the twin virtually, such as increasing heat or pressure, and thereby better predict what might occur in the real world.

Traditionally, there has been less intelligence behind the capture of sensor data. The health of a vehicle’s engine was monitored via sensors that triggered alerts when conditions required servicing for the engine or other part. However, alerts could mean a repair or maintenance was crucial either in the immediate future, or many miles and days in the future. That could lead to unexpected engine failures, even if an alert was provided, simply because the alerts aren’t very specific. Because PACCAR also provides servicing of its products, it sought a system that could make those visits and repair work more efficient and accurate, based on good diagnostic information.

PACCAR launched the system with its MX-13 engines, for use by its servicing division. With the system in place, the company can compare real-time information about an engine’s use and performance, compare it against the specific model and predict any necessary services. This reduces unnecessary servicing, and ensures that predictive maintenance is performed on an effective schedule, in order to reduce the incidence of failures.

For companies like Bsquare, digital twins can serve as the natural maturation of the Internet of Things. Since the IoT was developed, McCarthy says, “there’s been a hierarchy of needs” that began with access to data. Once companies were able to access such data, the question has been how to use it.

Bsquare has typically been introducing the technology to companies by starting with early piloting. “When we bring our products, it’s a pilot at a limited scale,” McCarthy states. Once the businesses become comfortable with the technology, he adds, “then we just turn up the volume,” by making it available across a company’s entire enterprise as they scale up.

Following on the successful IoT system deployment across its MX-13 engines, PACCAR is looking for other areas in its business to leverage the technology.