While the world of production continues to develop rapidly, it seems that for quality assurance time advances somewhat more slowly. Many of the widespread concepts (lean, six-sigma, etc.) were born in the 60s and 70s and have hardly changed since.
Stefan Debelt, Sales Manager at Industrie Informatik, became involved in quality management already early in career. He sees that powerful MES systems make an important contribution here. They support manufacturing enterprises in their quality assurance early in the production process. The key, he says, is in the transparency that the use of an MES provides. This and other interesting insights await you in the following interview.
MES systems can significantly contribute
What are the classic causes of quality problems in production?
Stefan Debelt: One of the causes is certainly a lack of transparency in manufacturing processes. Quality deficits are difficult or impossible to identify without the necessary information about the production processes. An example is the lack of early warning systems for workers on a shop floor. An MES enables targeted information retrieval and processing to ensure quality already during the production process instead of in retrospect.
Can’t defects and looming down time be detected in time?
Stefan Debelt: The condition of machines and tools often cannot be observed due to a lack of transparency in production. Wear, looming down time and damages cannot be detected early in this way, so down time is the logical effect. Maintenance intervals are based only on experience or rules of thumb, and thus they are detached from actual equipment use and wear. Thus collection of actual data via an MES is a prerequisite for detecting possible down time early and taking appropriate measures.
Quality assurance is increasingly being saddled on development or production departments. How do you see that?
Stefan Debelt: Errors induce costs. The later an error occurs or is detected, the higher the resulting costs. The well-known “rule of ten” demonstrates this problem: An error detected in development costs 1 €. In production planning it rises to 10 €, and a step farther in production it becomes 100 €. If the error is detected by the customer, then the cost escalates to 1000 €. Quality assurance should therefore reside very early in the product life cycle.
Is quality assurance today primarily a big-data problem (Industry 4.0)?
Stefan Debelt: It would be a wrong approach to narrow the discussion to one problem. The more data I can extract from my processes, the more transparent they become. This makes the processes easier to manage and control. Also, weaknesses and potential optimization are exposed, which positively affects quality. An important prerequisite is comprehensive data collection, data aggregation, and above all targeted and context-oriented processing of results.
They say that regarding quality assurance, the degree of utilization and linking of technical systems with CAQ, PPS and ERP systems is still too low. Why is that? Do we lack powerful software?
Stefan Debelt: Quality assurance is interdisciplinary; first you need to acquire the big picture to inform an enterprise about its own system landscape. From the start we need to be clear about which data are collected from which systems and how the data flow behind them is to be designed. This clarity forms the basis for a homogeneous and efficient system landscape in which all involved systems make their contributions to quality assurance.
Learn more about quality in production via an MES: