DIGITAL, DATA-DRIVEN BUSINESS MODELS WITH ADDED VALUE FOR END CUSTOMERS

The Manufacturing industry in Germany has a long tradition. Many of the often family-run companies can look back on stable, organic, healthy growth. 

However, traditional IT structures often cannot keep up with the pace of innovation. The market demands services that are scalable, immediately available, and “always-on”. This ever-increasing demand is challenging IT departments to constantly find new ways to position themselves and initiate meaningful digital change.

HEIDELBERG offers its customers – including well-known publishing houses and printers – digital subscription models that include equipment, consumables, software, training, and consulting as well as service.

HEIDELBERG needed a powerful IoT platform to ensure the operation and success of these data-driven business models – and thus also sustainable added value for end customers – and to build a digital ecosystem. 

The platform should not only collect and visualize the data but also link it together in a meaningful and seamless way to enable predictive maintenance and remote maintenance, for example. At the same time, the HEIDELBERG data platform was to provide helpful data analyses from the day-to-day operation of the numerous printing presses to continuously improve them.

Since HEIDELBERG had been collecting data based on machine logs since 2005, the company already had extensive unstructured data at its disposal, which service technicians could use to support end customers remotely. 

However, the existing solutions were a mixture of in-house developed and purchased products, which meant that the data was in silos, maintenance was time-consuming, data consistency was not guaranteed and costs were high. 

Above all, HEIDELBERG wanted the possibility of seamless integration and real-time data processing. The new IoT platform was to fulfill these wishes.

A SCALABLE IOT PLATFORM IN THE AWS CLOUD

To begin with, the printing presses underwent an analysis of their data structures and communication channels to harmonize the data and facilitate migration to the cloud. Based on the qualitative and functional requirements, the team used a lightweight protocol based on MQTT, which opened up bidirectional communication with the cloud and enabled the machines to send a large number of messages and process them with low latency.

The end-to-end solution was validated with a proof of concept (PoC). This helped to test the feasibility of the solution and showed that data consistency and integrity were maintained even with large volumes of data and varying intervals. 

Downstream, the processes were supplemented via a provisioning function so that machines that were already in the field could dock securely to the cloud-based IoT platform. Fluid data processing for large volumes of data was implemented based on streaming services from the cloud provider AWS. 

The creation of a data lake helped to store the telemetry data in a structured manner so that HEIDELBERG could develop complex analyses – such as those required for predictive maintenance. This provided transparency and control, regardless of location and in real-time.

EVENT SOURCING AS THE BASIS FOR ERROR AND DATA ANALYSIS

Zoi began the technical implementation of the IoT platform with the development of an event-driven architecture. Event sourcing takes place through orchestrated state changes in events as a sequence (stream). This enables the use of serverless services and guarantees high scalability. The architecture was designed to be flexible so that future adaptations can be made easily. As all events are recorded in the event memory, this enables reliable traceability and deterministic error and data analysis. 

At the heart of the platform was an event bus, which represents the communication path between the individual services and the integration of consumer systems. We developed the following services for the IoT platform based on AWS:

• Provisioning service: Provisioning (authorization and authentication) of the machines via API gateway and X.509 certificates so that the machines can communicate securely with the platform. 
  
• Telemetry service: Receiving, validating, and processing telemetry data in real-time with IoT Core, Kinesis Data Stream, and Lambda. 
  
• Analysis service: Transfer telemetry data, events, and files such as logs, configuration files, and updates to an S3 Data Lake for data analysis with integration of S3, Lambda, Athena, and Glue. 
  
• Remote service: Possibility of remote maintenance on the machines by using AWS IoT Secure Tunneling.
  

ABOUT HEIDELBERG

Heidelberger Druckmaschinen AG (HEIDELBERG) is a global technology corporation and a leader in the printing industry. With a strong focus on digitalization, HEIDELBERG offers integrated and automated system solutions for machines, software, consumables, and services. In the 2021/2022 financial year, turnover amounted to around € 2.2 billion. The company employs around 9,800 people at 250 locations in 170 countries worldwide.

Every day, the platform records millions of events that provide information about the status of the machine, printing progress, and the number of correct or faulty prints, among other things, and thus serve as the basis for creating further data-driven applications.

Thousands of machines are already feeding in their data – and more end customers are continuously being connected to the platform. 

The IoT platform has the potential to become the global hotspot for remote maintenance and evaluation of printing presses. It also forms the basis for further digital business models in a cloud-based ecosystem that will continue to grow with planned ML and AI use cases as well as existing HEIDELBERG services that still run on-premise.

By developing the platform independently with Zoi, HEIDELBERG retains control over its data, can act at any time, and can also flexibly integrate future services. With a higher degree of transparency for operational processes, HEIDELBERG is now able to offer end customers an even higher level of service.

The underlying architecture of the IoT platform is based on AWS Managed Services.