Use Case

Generating Actionable Insights from Vehicle Data Pipelines

Many vehicles on the road today are connected to the cloud and they generate, store and transmit large amounts of data. Leveraging this data to anticipate component failures, create personalized vehicle ownership experiences or generate advanced fleet insights holds significant potential in financial value for vehicle manufacturers and their ecosystem partners. However, data complexity and inadequate tooling have hindered even the best efforts to capitalize on vehicle data at scale thus far.

The keys to successfully transforming raw connected vehicle data into actionable insights are scalable data orchestration tools and powerful application-specific machine learning methods. 

This use case highlights the power of combining Xapix’s data orchestration technology and Viaduct’s AI solution on a large vehicle operator.



Our customer, a large commercial vehicle fleet operator struggled with turning their fleet’s raw vehicle data into business intelligence that would impact their bottom line. In the previous year, the company invested millions of dollars into connecting their vehicles to their centralized-cloud by retro-fitting them with aftermarket telematics units.

However, the fleet operator realized data collection isn’t enough. By combining vehicle data with various other data sources and leveraging purpose-built data analytics and machine learning methods, raw vehicle data was transformed into actionable insights. In this case, the fleet operator was interested in unlocking two high-value applications:

(1) predictive maintenance to save on maintenance dollars and improve vehicle uptime

(2) early detection of driving anomalies to improve driver safety and lower the fleet-level cost of insurance

Orchestrating, Aggregating and Deriving Actionable Insights From Vehicle Data

After attempting to build a solution in-house and scanning the available solutions in the market, the fleet operator brought on Xapix and Viaduct to help them derive actionable insights from their vehicles’ data.

It was the power of this combined approach that unlocked success: 

Combining the power of vehicle AI with the development of data pipelines and workflows allowed Xapix and Viaduct to onboard the fleet operator and build an end-to-end solution in less than 3 months. The fleet operator saved money across the connected fleet and reported a considerable increase in vehicle uptime. Vehicles no longer experienced catastrophic failures on the road that may take up to weeks to address, and predictive analytics improved service scheduling for more minor repairs, further increasing fleet-wide uptime metrics. In addition, the fleet operator had the ability to conduct early detection of driving anomalies and vehicle incidents and could proactively lower fleet-wide risk by targeting risky driving behaviors and risky drivers.

Data Types

Xapix and Viaduct ingested, integrated, analyzed and orchestrated a multitude of the fleet operator’s proprietary and externally provided data streams at scale, including:

Xapix Platform Features

By integrating Xapix into his data ecosystem, the fleet operator was able to create efficient and robust data connections in a low-code environment. Through additional data standardization and harmonization features Viaduct’s machine learning algorithms was able to instantly process the data and identified key events using predictive analytics.


Viaduct Platform Features

Viaduct’s Vehicle Profiles solution leverages data analytics and machine learning methods to make temperable predictions on large-scale connected vehicle data and generate actionable insights.

Vehicle profiles allowed the fleet operator to use historical vehicle telematics data and repair records to identify vehicles at high risk of immediate on-the-road failures, as well as early identification of aggregated issues that may require a swift and fleet-wide response. 

In addition, Viaduct’s Driver Fingerprinting solution used machine learning on-top of connected vehicle data to power traffic incident risk analysis. With Viaduct’s Driver Fingerprinting solution, the fleet operator could use driving histories to identify at-risk drivers and reduce traffic incident risk by issuing corrective actions and recommendations. It also automated accident alerts for faster responses to on-the-road incidents.





About Viaduct 

Viaduct is a technology company located in Menlo Park, California, and provides an end-to-end machine learning platform to empower automakers and their ecosystem partners to make safe, reliable, and personalized vehicles. Viaduct’s cloud platform derives actionable insights from connected vehicle data to power applications such as predictive maintenance, smart campaigning, driver identification and safety scoring, and in-vehicle experience personalization.

Learn more about Viaduct


About Xapix

Xapix is a data integration platform that provides demanding industries like automotive, logistics & supply chain and IoT with ultimate flexibility, speed, and security around data interoperability. New data connections can be set up within minutes in a low-code environment. By supporting real-time data streams and data standardization, Xapix is perfectly suited for scenarios where large amounts of data need to be transformed into actionable and precise insights. The platform is already in use with some of the leading automotive companies across Europe and the U.S.


Other use cases that might be interesting for you

Telematics & Data Standardization
Easily connect to a variety of different types of telematics units, no matter how many. Distribute, standardize and harmonize your data from dozens or even hundreds of different services and applications.
View Use Case
IoT & Telemetry Data Platform
Combine high frequency, real-time IoT telemetry streaming data with contextual data around weather, location, and traffic as well as metadata. Use data normalization, harmonization, and orchestration capabilities to feed and integrate applications and AI models.
View Use Case
REST to SOAP Transformation
Combine SOAP web services, REST APIs, and event streams towards cloud architectures. Support several hundreds of existing partners through legacy services while providing newer services and offerings.
View Use Case