Marelli: Smarter manufacturing

Marelli has a transformation path ahead if they want to be prepared to implement the Industrial Internet of Things (IIoT). So, the company collaborated with Accenture to identify areas that will help it be more competitive.

Manufacturers are moving quickly to adopt what Accenture calls Industry X capabilities. With these new capabilities, they hope to gain competitive advantage, increase customer satisfaction and enhance brand reputation. Digital enablers are key; they collect data from systems, machines and workers to reveal valuable insights.

The automotive supplier Marelli wanted to understand what it needed to do to adopt Industry X capabilities that would improve operational efficiency through innovation. A priority was to improve their manufacturing systems and monitor asset efficiency.

Some assets, like the engines powering welding robots on production lines, are expensive, and strategically crucial. Engine maintenance, however, was happening on a fixed schedule instead of in response to equipment performance.

Marelli collaborated with Accenture to find a way to digitalize plant operations and processes in order to connect assets and unlock efficiencies.

• Production scheduling and execution using artificial intelligence and machine learning
• Quality management with digitalized processes and data
• Materials management to streamline production line feeding and material traceability
• Plant maintenance to improve asset utilization and workforce management

First, we identified where and how the current manufacturing application landscape could be enriched. Then we defined an integrated information technology and operational technology (IT/OT) application suite for manufacturing operations that would serve as a solid digital foundation for the future IIoT platform, leveraging capabilities from the Industry X Innovation Center in Modena, Italy.

We also helped Marelli understand how to improve efficiency monitoring with a proof of concept. It focused on manufacturing assets at two plants, one for lighting components and the other for suspension systems. The proof of concept looked at key production assets, such as the engines powering the welding robots and an injection molding machine. An IIoT laboratory platform collects and processes data from sensors and programmable logic controllers (PLC) on these assets.

We then developed predictive algorithms to identify the likelihood and potential timing of an asset’s failure, as well as critical conditions affecting production. To advance this work, we are currently developing an IoT Edge platform (based on Microsoft Azure) with a flexible and scalable architecture. The platform can source and process data from many different types of machines on the shop floor.

When the system detects an impending failure or critical status, it raises a notification to supervisors to act quickly before the machine breaks down.