Microscope

Preventive maintenance first needs a valid data history. ZEISS Microscopy analyses the operating and condition data of its connected microscope systems with machine learning so that service technicians can act before a failure occurs. Because the devices rarely fail, there was hardly any training data at first. Data was therefore collected first and the truly relevant parameters identified, and only then were predictions introduced step by step.

IoT shortens the route to a repair because failure risks become visible early and information and recommended actions are available centrally before the technician sets off. At ZEISS Microscopy, service teams identify risks early, inform customers proactively and combine maintenance appointments. Fault diagnosis becomes simpler, and the service team can respond faster instead of being called only after a failure.

At ZEISS Microscopy, a generative service assistant helps. It draws on documented experience from maintenance sessions and manuals and makes it available as a chatbot. Technicians receive step sequences, tool and material lists and notes on critical points, so they work faster and more safely on site even without deep experience. According to doubleSlash, this reduces the effort needed to pass on knowledge within the service team.

The basis is a uniform connection for all device types with a shared architecture and a single security concept. ZEISS Microscopy and doubleSlash standardised heterogeneous data and connected it to a central platform. Central device and update management keeps software versions consistent. Because knowledge and data are available centrally, service processes become comparable worldwide, and new product lines can be connected efficiently.