Industrial manufacturing is undergoing a transformation: Today, machine vision systems do far more than just perform traditional quality control at the end of a production line. Combined with digital worker assistance, they monitor processes in real time, support employees directly at their workstations, and provide valuable data for the continuous optimization of workflows. MKey Solution GmbH develops customized solutions for industrial image processing and metrology. The company combines cameras, sensors, and intelligent software into systems that automate production and logistics processes, visually guide employees through workflows, and detect errors in real time. A particular focus is on digital worker assistance systems that make assembly and manufacturing processes more efficient, safer, and more consistent in terms of quality. In this interview, Managing Director Michael Kunze explains how companies can use these systems to reduce scrap, improve quality, and make the transition from reactive error correction to proactive process control.

Industrial Production: Mr. Kunze, image processing has traditionally been used in industry primarily for quality control. What changes are currently taking place?

Michael Kunze: The role of image processing is undergoing a fundamental transformation. In the past, it was mainly used at the end of a production process to detect defects. Today, it is evolving into an integral part of ongoing manufacturing. Modern systems no longer just analyze end products; they monitor the entire process in real time. Combined with digital worker assistance, this creates a system that actively contributes to the optimization of workflows.

What does this mean specifically for companies?

Companies in sectors such as manufacturing and maintenance gain a completely new tool for improving quality, reducing scrap, and refining processes based on data. Cameras continuously monitor the status of components, assemblies, and work steps. This information is fed back directly to employees. As a result, errors can be identified and corrected immediately—before they affect subsequent production steps.

Can you give an example of that?

For example, a system can detect whether a component has been positioned correctly, whether assembly steps have been completed, or whether there are deviations from the defined target state. Instead of downstream error detection, feedback is provided directly during the process. This transforms image processing from a passive inspection tool into an active production aid.

What role does digital worker assistance play in this?

It serves as the central interface between technology and employees. Through visual instructions—for example, projected directly onto the workstation—work steps are presented in a structured and context-specific manner. The information appears exactly when it is needed. This reduces cognitive load and minimizes sources of error, especially in complex or highly variable processes.

What additional benefits does data collection provide?

A key advantage lies in the comprehensive documentation of all work steps. Each process step is analyzed and assigned to specific stations, employees, or shifts. This real-time transparency lays the foundation for AI-based process analyses. Companies gain detailed insights into operations at the line or workstation level and can identify variations in processing times, error rates, or process stability.

How can companies put these insights into practice?

Kunze: For example, it is possible to identify differences between various departments or systematically apply best practices from individual employees. This enables continuous optimization of production. Bottlenecks can be specifically addressed, workflows adjusted, and training measures implemented as needed.

What technical components work together in this process?

The combination of image processing, operator assistance, motion control, and process data creates a fully integrated system. Employees are guided by projected instructions, work steps are automatically verified using 3D sensors, and results are documented in real time—reliably, consistently, and regardless of lighting conditions. At the same time, an integrated user management system ensures that only qualified employees can perform specific tasks.

How does this affect efficiency and quality?

The efficiency gains are measurable. Errors are detected and corrected early on, training needs become apparent, and scrap and rework are significantly reduced. At the same time, quality and output increase. In addition, downtime is reduced because problems can be identified and resolved more quickly.

What are the implications of this approach for maintenance?

That is precisely where it offers additional benefits. Maintenance and service processes can be standardized, documented, and evaluated for efficiency. Recurring errors can be identified and prevented in the long term. The data generated enables systematic AI-based analysis and continuous improvement of individual work steps.

In your opinion, in what direction is industrial manufacturing heading?

We are witnessing a clear paradigm shift: away from reactive error correction and toward proactive process control. Quality is no longer checked exclusively at the end of the process, but is continuously ensured throughout the process. Digital worker assistance helps make human work more precise, safer, and more efficient. In my view, this is precisely where the future of modern production systems lies.