Smart Safety: Safety for modular production plants at run time

SmartFactoryKL presents a “Smart Safety” concept at the Hannover Messe
AI minimizes safety risk at modular units

The safety concept is designed to protect people from the risk associated with manufacturing jobs. Increasingly complex, modular Industrie 4.0 production plants demand a new safety concept. Thanks to Artificial Intelligence (AI), it is now possible to make modular work stations much safer. The SmartFactoryKL partner consortium shows how to use AI to minimize the safety risk for Industrie 4.0 production plants at the Hannover Messe on April 1-5, 2019. Learn more at the joint exhibit stand of SmartFactoryKL and German Research Center for Artificial Intelligence (DFKI) located in hall 8, booth D18.

The safety policies must adapt to the flexible processes of the modular plant. The production processes in Industrie 4.0 plants often vary because of the various lot sizes – often at lot size 1. The safety of the production staff is of utmost importance and new policies must be implemented. AI provides an ideal solution to the challenge of minimizing the safety risk at modular work stations. Risks can be detected and the appropriate risk minimization measures can be initiated by adaptive AI algorithms.

“Artificial intelligence helps reduce the safety risk associated with modular work units. In the more complex Industrie 4.0 production plants of the future, such a smart safety concept will be necessary as today’s strict safety policies will no longer be adequate for the flexible factories of the future,” said Prof. Dr. Martin Ruskowski, Department Head of Innovative Factory Systems research at DFKI and expert in the use of AI for manufacturing.

Automatic certification of modular work stations

Every equipment manufacturer must provide a declaration of conformity confirming that the unit can manage all hazard scenarios and that the minimum safety requirements are met. In the future, confirming this conformity quickly after each conversion of a modular plant, will require a modular certification concept that checks the entire network of linked systems.

AI algorithms can automatically perform the certification process for the interfaces between two production stations. Using smart software, so called safety agents, will reduce the complexity of the interfaces between linked units. The safety agent checks various scenarios to determine, for example, whether there is a mechanical or a chemical risk. If the risk is negligible, a safety certificate is prepared and the production can continue – avoiding long downtimes in the truest sense of Plug & Produce.

Use Case: Integrated Flexible Transport System for Module Safety

A flexible transport system (FTS) is included in the safety concept of the SmartFactoryKL Industrie 4.0 production plant. In the event that the FTS moves into the zone of a production line, it is automatically assigned to that unit and, consequently, also to the associated emergency stop circuit. Similarly, instead of the transport system, another module component could also be flexibly exchanged and automatically certified by AI.

“The advantage of this solution is the significant reduction in downtime after a plant conversion, because the most expensive and time consuming decisions about plant safety are now performed by artificial intelligence. We demonstrate automatic certification at our exhibit stand,” said Prof. Dr. Detlef Zühlke, Chairman of the Board of SmartFactoryKL.

Use Case: Smart Vision Systems for Safety Checks

When a signal comes from the FTS that it has arrived at the docking station, the workpiece carrier moves forward on the conveyor belt of the Industrie 4.0 production plant at the SmartFactoryKL However, at this point there is no official confirmation from the safety system that the signal actually came from the FTS and there may be a risk if the workpiece carrier moves. Eventually, the workpiece carrier could fall off the belt and disrupt the process in the long term.

A smart vision system uses a camera to check the interface between the docking station and the FTS, for example, looking for the correct orientation of the conveyor belts. The optical data is evaluated by agent systems. If the FTS is correctly positioned, the safety certificate is created and the process is released – the conveyor belt with the workpiece carrier starts to advance.

This use case is visualized using colored light installations. If the vision system has not performed a final risk check, a flashing light on the FTS is activated. After checking for safety risk and confirming the certificate, the FTS becomes part of the safety circuit of the respective production line and the flashing light changes to reflect the corresponding color of the line. Only then the workpiece carrier can approach the FTS.

The Industrie 4.0 demonstrator partners at SmartFactoryKL are: B&R Automation, Bosch Rexroth, EPLAN Software & Service, Festo, HARTING, Huawei, IBM, KIST Europe, METTLER TOLEDO, MiniTec, PHOENIX CONTACT, Pilz, proALPHA, TÜV SÜD, and Weidmüller