How Digital Twins are revolutionizing fleet management

Today, Digital Twins—even though the technology is still relatively new—are already helping them do that. The real-world applications of Intangles’ proprietary Digital Twin and AI platform provide useful illustrations of the technology’s potential for fleet management. Incorporating telematics, advanced analytics and powerful algorithms, the company focuses on predictive maintenance, driver behavior analysis and operational efficiency, providing insights, real-time alerts and even repair strategies.

Vehicle health

With a Digital Twin of a truck, Intangles’ platform can identify potential failures and vehicle maintenance issues days or even weeks before a Diagnostic Trouble Code, or DTC, is generated. Leveraging historical and real-time data, the platform also gives insight into the severity of the potential problem (minor, major or critical), enabling more effective decision-making for addressing maintenance or repair needs. For Intangles customers, the result has consistently been a significant reduction in on-road vehicle breakdowns, creating savings in maintenance and repair costs, as well as unplanned towing. It also means that fleet operators must make restitution to their customers or pay for alternative delivery methods less often. Predictive algorithms generated with the Digital Twin enable preventive and proactive maintenance rather than waiting for a DTC or a vehicle breakdown—saving time, money and vital customer relationships.

A good example is the monitoring and modeling of air intake in Intangles’ Digital Twin platform. The platform tracks engine turbocharger boost pressure in real time and is continuously “learning” from other similar twins to develop a mean boost pressure for the vehicle. Then, it compares the real-time data to the calculated mean, potentially identifying issues well before the truck’s onboard ECU generates an alert.

Since launching in 2016, the Intangles Digital Twin solution has detected millions of critical faults before a DTC was triggered. For Intangles-equipped fleet operators, that has resulted in a 75% reduction in breakdowns, a 10%-30% increase in asset availability, and a 5%-10% decrease in maintenance costs.

Case study: Municipal Fleet

A fleet of 1,400 municipal vehicles, including 90 refuse trucks from three different OEMs, were equipped with the Intangles AI-powered fleet maintenance system, with a view to improving turnaround times, fuel efficiency and preventative maintenance. During the pilot phase, the Intangles platform detected faults in 30% of the trucks—all before DTCs were triggered. The predictive alerts saved the municipality thousands of dollars in maintenance costs and helped reduce downtime, while increasing the fleet’s fuel efficiency by an estimated 5.6%. In total, the municipality realized cost efficiencies amounting to US$500 per vehicle per month.

Fuel monitoring

In the trucking industry, fuel is a precious and expensive commodity. So, too, is DEF, the catalytic converter additive that helps reduce exhaust emissions and is often in short supply.

As a result, fuel and DEF theft has become a major concern for fleet managers and vehicle owners, and several fuel-tracking devices have come on market to address that concern. However, most of these products have been severely limited in their ability to provide detailed information about the location of potential theft, the amount of fuel/DEF missing, or the time of the suspected pilferage.

Intangles’ Digital Twins can provide fleet operators with far more detailed and useful information. By leveraging sensor-derived and historical data into machine-learning algorithms, the Intangles Digital Twin can detect the exact time, location and number of fillings and potential thefts. The modeling can also help fleet operators compute an accurate cost of fuel consumed per mile. Fleet managers using the Intangles platform report an average improvement in fuel consumption and fuel shrinkage of between 5% and 10%.

Driving behavior

It is estimated that 40% of vehicle performance depends on driver behavior and that 90% of all road accidents occur because of driver error. With a Digital Twin, fleet managers can gain a new level of oversight and insight into this vital element of safety and performance.

For example, the Intangles Digital Twin platform monitors more than 20 exceptions in driver behavior, including speeding, excessive idling, hard braking, fuel theft, over-acceleration and unscheduled driving, and compiles them into a peer-to-peer ranking model. The model provides real-time, accurate reports on gear utilization, then compares the data against best practices to identify areas for improvement and increase fuel efficiency. The Digital Twin performs similar analyses for free running (in neutral gear) and hard braking to help reduce the risk of accidents.

Through the platform, drivers get real-time feedback on their driving practices and suggested strategies for improving performance. Meanwhile, fleet managers can use the findings of the behavior scorecard to take corrective action and/or identify opportunities for improved driver training.

Range prediction

As EV trucks become more commonplace, fleet operators seeking to maximize efficiencies will have to overcome a major roadblock: understanding and predicting range. How far an electric vehicle can travel before charging depends not only on battery capacity and other physical factors but also driving conditions, driver behavior and other variables. In fact, the battery discharge rates of similar vehicles on similar routes can vary widely, and typical onboard information systems that estimate distance-to-empty are often unreliable. For commercial vehicles, this uncertainty can lead to proactive but unnecessary charging sessions, disrupting schedules and increasing asset downtime.

A Digital Twin platform can help resolve the uncertainty. The Intangles system, for example, provides comprehensive data on charging cycles and tracks the charge degradation data provided by the vehicle’s battery management system, then enhances charge tracking by considering ambient and driving conditions. Through predictive modeling on the Digital Twin, the system makes predictions based on motor torque, wheel speed, weather (present and forecast) and outdoor light conditions, which impact a truck’s climate control and lighting systems. The platform also monitors driver behavior, including idling and overuse of HVAC and lights.