Supernova Award Category

Internet of Things

The Problem

For over two decades, Navistar used several traditional data warehouses to ingest and analyze data from its customers’ vehicles. With the launch of Navistar’s complex and growing OCC service offerings in 2013, Navistar needed a solution that could support data from both internal and external sources, in both structured and unstructured formats. With most of the external data coming from sensors on vehicles in addition to weather and geospatial IoT services, combined with data from more than 300,000 vehicles, Navistar needed a scalable solution to analyze data in real time, across multiple telematics sources, to drive predictive analytics.

The Solution

In 2014, Navistar chose Cloudera Enterprise to build a large, scalable platform to enable the company to ingest, store and process multiple data sources and formats. All of the streaming data from sensors can now be directly ingested and combined with data from internal and external sources to drive insights and analytics, at considerably lower cost per terabyte. With Cloudera, Navistar is able to perform analytics on an ad-hoc basis and evaluate billions of rows of data in hours -- not weeks/months. By tracking information from vehicles, Navistar is able to schedule preventative maintenance in advance of issues, enabling customers to reduce vehicle downtime and increase their revenue. This ability has helped Navistar become a data-driven company and given the organization a competitive edge. Today, Navistar is using Cloudera Enterprise to bring together data from 300,000+ vehicles in real time, across multiple telematics sources, to drive predictive analytics.

The results

Navistar’s use of Cloudera has allowed the company to apply data management and analytics practices that accelerate the company’s IoT and connected vehicle journey – driving critical business decisions that help Navistar further pave the way for the vehicle manufacturing industry. With Cloudera, Navistar can run analysis and predictive modeling using advanced machine learning algorithms, to detect potential maintenance issues on customers’ vehicles. Navistar is also able to offer analytics as a service to its customers, who can now monitor the performance of their vehicles in real time from their smartphones or tablets, and use the resulting insights to schedule planned maintenance, thereby reducing unplanned repairs and downtime. Now able to access all of its data in one place, Navistar can perform analytics on an ad-hoc basis and evaluate billions of rows of data in hours, which previously took weeks/months. By utilizing a big data analytics platform, Navistar is now able to explore its IoT data in real time, providing answers to its customers easily, quickly and efficiently.

Metrics

Since partnering with Cloudera in 2014, Navistar has reduced maintenance costs and vehicle downtime by almost 40% and implemented remote monitoring coverage to more than 300,000 vehicles on the road. By using vehicle sensor data to detect when vehicles need maintenance, Navistar enables its customers to proactively schedule maintenance, which keeps drivers safe and able to deliver on time. Early detection also minimizes money spent on towing, with a significant reduction in tows year over year. With an average cost for a down vehicle being up to $750 per day, for 300,000+ vehicles across the company's 2,200+ customers, this savings is significant.

Navistar also services thousands of school buses nationwide, transporting thousands of students every day, using IoT to access detailed information about vehicle performance in real time, ensuring transportation conditions are safe and the buses are able to get kids to school on time. The Saratoga Springs (N.Y.) City School District, which serves 7,000 students and runs 110 school buses that travel 1.5 million miles a year, uses OCC to monitor engines, in external temperatures as low as -25 degrees, and is notified of issues immediately when a vehicle fault occurs – ultimately keeping children safe while getting them to school on time. By maintaining buses preventatively, not reactively, the district saw a drastic reduction in tows year over year, while increasing uptime, reducing cost and helping ensure child safety.

The Technology

Big data platform: Cloudera Enterprise

• 20TB of legacy data ingested into Hadoop, augmented with 22TB of IoT data (Telematics, Geospatial, Weather, FMCSA etc.), to generate an additional 20TB+ of project specific data
• Data ingestion patterns use: Sqoop, Spark, Flume, IBM Infosphere Datastage and manual file loads
• Kafka cluster to consolidate the data feeds
• Navistar’s development patterns: Cloudera Data Science Workbench, Spark Scala, PySpark, Python, R, HiveQL and Impala

Disruptive Factor

By reducing costs by nearly 40%, Navistar is an excellent example of how organizations can utilize big data platforms for driving IoT use cases. Navistar’s use of the Cloudera platform greatly expanded its ability to explore and draw insight from customer vehicles by gathering all data available from IoT sensors. The company will continue to innovate with the massive volumes, variability and variety of data gathered from IoT-connected vehicles – using the Cloudera platform to gain valuable insights from its growing sources of information. This, and other emerging connected-vehicle services enabled by big data analytics, offer the potential to transform the now $800 billion trucking industry by reducing maintenance cost and downtime and delivering new dimensions of safety, fuel efficiency and performance.

Similarly, school districts can use Navistar and OCC to proactively monitor their buses remotely, keeping their students safer and on time for class. Connected cars, including trucks and buses, generate up to 25 GB of data per hour, an offshore oil rig can generate 1-2 TB of data every day, and a turbine could generate up to 1-5 TB of data per day – meaning a highly scalable solution is necessary. Industries on the verge of disruption by using IoT, such as manufacturing, healthcare and retail, have great potential to benefit by modeling their big data solutions after Navistar.

Shining Moment

CIO magazine announced Navistar as a recipient of the 2016 CIO 100 award for its innovation with OCC. In 2017, TDWI, the leading global provider of analytics and data management research and education, named Navistar’s deployment of Cloudera a winner of the Best Practices Award in the Big Data category. Navistar was chosen for its advanced use of IoT devices and real-time analytics to help manage and maintain its fleet of hundreds of thousands of vehicles through predictive analytics.