Modernizing fleet orchestration for 300,000+ devices using AWS with ZF Group

With SCALAR, ZF provides global logistics and transportation companies with fleet management solutions, including systems that track vehicles in transit and communicate with drivers. It manages over 300,000 connected devices as part of an intelligent data and Internet of Things platform. To route the right vehicle to the right place at the right time, the platform ingests location, resting time, diagnostics, and other data from trucks and trailers.

The company saw opportunities to expand from its main customer base in Europe to new markets. But ZF’s monolithic architecture couldn’t scale to accommodate this growth, so the company turned to AWS to rebuild its offering as a cloud-based SaaS solution. ZF has used AWS services for years and appreciated the innovation they offered in other parts of its business. This, coupled with the global infrastructure, security, and plethora of services that AWS provides, meant that AWS was a natural choice. “We want to evolve into a product-driven SaaS company, so we need the capability to react to our customers’ needs and to the changing market conditions,” says Jef Vandenbergen, head of cloud platform engineering at ZF.

ZF began its modernization journey by establishing the foundational building blocks and core principles to work from. It implemented an automated GitOps way of working, including infrastructure as code, and adjusted the operating model while following industry best practices. “Using AWS as the complete backend has been instrumental in reshaping and rewiring the company,” says Vandenbergen. “Now we’re using an operating model where the developers who create a service take ownership of their modules, including operations, monitoring, and costs. We’ve empowered our teams to be responsible for their domains.”

SCALAR, ZF’s next-generation fleet orchestration solution, uses advanced algorithms to automate workflows and help businesses achieve profitable margins, leading to safer, more reliable transportation. Fleet managers use SCALAR to improve productivity and sustainability by dynamically optimizing routes in response to live conditions. The solution uses near real-time vehicle data to recommend proactive maintenance, so fleet managers can also minimize unplanned downtime. Customers can use the solution to demonstrate compliance with regulations, such as trailer temperatures and driver rest time. By rearchitecting an elastic backend with dynamic scaling, ZF reduced its costs per device per month by as much as 45 percent for data ingestion.

ZF is also realizing significant cost savings by implementing disaster recovery backups on AWS, removing the need for keeping largely unused hardware or hosting. The company applies budget ownership and governance by using tagging functions within AWS, obtaining a more granular view of costs for each team and each module. ZF also gained agility. “Requests to implement infrastructure changes, which previously took weeks for our hosting provider to accommodate, can now be done in days or even hours on AWS,” says Vandenbergen. “We can ask for something in the morning, and by lunch, it’s already done.”

SCALAR is powered by an event-driven microservices architecture. “We no longer wanted to rely on large monolithic databases,” says Vandenbergen. “Instead, we wanted to design a scalable architecture from the ground up.” ZF built the solution by using Amazon Elastic Kubernetes Service (Amazon EKS), a fully managed Kubernetes service, and AWS Lambda, a service to run code without thinking about servers or clusters. ZF also uses AWS services for messaging, networking, logging, and data storage and takes advantage of built-in security features and AWS best practices for security. “Running on AWS has really improved our security posture,” says Vandenbergen. “It was one of the main decision criteria in choosing AWS.”

This modernization supports ZF’s goal of becoming a product-driven SaaS company. Development engineering teams can now build, innovate, and take full product ownership through access to managed services that off-load complexity. These services include Amazon Relational Database Service (Amazon RDS), easy-to-manage relational databases, and Amazon ElastiCache, a serverless, fully managed caching service. “Moving to AWS, we got the benefit of all these managed services that we didn’t have before,” says Gregory Vander Schueren, global product owner of cloud platform engineering at ZF.

By modernizing and optimizing its fleet orchestration platform on AWS, ZF has the flexibility and scalability to grow and adapt to new opportunities. The company has plans in place to launch in new geographies in as little as 1 month. “If we see an opportunity, we are now set up to build a new product quickly and reliably using AWS,” says Vandenbergen. “We can have a global strategy because AWS provides the reliability and the global scale that we need to expand to new regions.”

ZF continues to change its methodology and modernize on AWS. To develop internal generative AI assistants, the company is using Amazon Bedrock, an easy way to build and scale generative AI applications. The assistants access team knowledge bases and locate documentation quickly to conduct root-cause analyses, and they might be used later for automated code reviews. “We don’t know what the products of tomorrow will be, but on AWS, we can adapt quickly to capture opportunities,” says Vandenbergen.