OEE Management at Zentis - How Intelligent Analytics Can Help

Sophisticated analytics, mobile apps and easy-to-implement software help increase OEE (Overall Equipment Effectiveness) at Zentis.

The digitization of the industry

While a future project for the comprehensive digitization of industrial production had already emerged in 2011 under the term Industry 4.0, the Corona crisis has once again fueled the digitization trend. According to the study "Corona leads to a digitization push" by bitkom research, three quarters of companies with 100 or more employees have increased their investments in digital devices, technologies and applications as a result of their experiences during the Corona crisis. The comprehensive networking of processes in production is now a defining theme at virtually every major manufacturer in Europe, the USA and Asia. Industry 4.0 enables more efficient production methods and thus provides the framework for efficient use of capital and an improvement in machine productivity of up to 25%, according to the study "The Industry 4.0 transition quantified" by Roland Berger. In the process, the level of digitization is driving the corporate world far apart. This is shown by the study "Digital Maturity Is Paying Off" by the Boston Consulting Group (BCG). According to the study, around 23% of the companies surveyed are pioneers when it comes to digitization, while around a third (32%) are lagging well behind. Manufacturing is no exception. "Digital laggards must not allow themselves to be beaten off and must quickly create the necessary technological conditions," emphasizes Prof. Dr. Markus Focke from Aachen University of Applied Sciences. "Digitizing production forms the basis for using data analysis and artificial intelligence to strengthen competitiveness in the long term." The biggest inhibitors to digitization today, especially in medium-sized companies, are high investments as well as a lack of know-how. These challenges create potential for disruptive solutions. More and more companies are developing customized digital applications for specific challenges. For example, the start-up ifp Software GmbH from Aachen offers oee.ai, a plug-and-play system for analyzing and increasing effectiveness in manufacturing processes. "Our goal is to offer every company a fast and affordable way to collect production data, evaluate it automatically, analyze it and carry out targeted optimizations" explains Jörn Steinbeck, co-founder of oee.ai. Two data elements are in focus for the analysis of the OEE. The unit count vector describes the units produced over time. If the unit count expectation deviates from the actual units produced, the cause of the deviation is recorded for data analysis. The source of this data can be both the plant control system and human input from the plant operator. Using previously described data elements, the approach to effectiveness improvement is built following a holistic analytics approach that includes both business intelligence and AI-based advanced analytics. Here, the vision of OEE management extends to prescriptive analytics.

oee.ai in use at Zentis

Zentis Süßwaren GmbH & Co. KG has been using the oee.ai system holistically in the production of B2C confectionery products since the beginning of 2020. Thanks to a fast plug-and-play solution, a consistent system for recording productivity, as well as productivity losses and the key figure OEE, could be introduced within a few weeks, thus replacing an old tool for measuring performance. With the automated recording of OEE and the reporting of losses via a tablet, live key figures are now available for analysis. "At the start of the project, a timely and clear solution for displaying the OEE and thus deriving optimization measures was very important to us," says Zentis Managing Director Claus Ernst. In the daily shopfloor meeting, deviations are analyzed and short-term measures for optimizing the manufacturing processes are coordinated with the help of the visualization of the OEE. In this way, it is directly apparent which production lines have achieved the defined targets and where there is a need for action for optimization. In particular, the display of availability losses or downtimes via a Pareto chart provides very quick insights into which measures have the greatest influence on continuous improvement. In addition to the daily shopfloor meeting, Zentis uses the system to create cyclical evaluations on a cumulative basis. This illustrates which downtimes lead to performance losses over a longer period of time. On this basis, several optimization projects were initiated last year. In accordance with DIN 31051, the tasks of maintenance include not only the basic measures of maintenance, inspection and repair, but also plant improvement and the analysis of failure behavior. Here, the automated recording, analysis and evaluation of OEE can play a significant role. At Zentis, the maintenance team can also benefit greatly from the evaluations with oee.ai. These make it possible to react to weak points at an early stage and to implement preventive maintenance measures in an even more targeted manner. "By using oee.ai, we have gained deeper transparency, which helps us a lot in all sub-areas of OEE - availability, performance and quality - to identify the weak points in order to strengthen our competitiveness," explains Production Manager Caroline Pangritz. Thanks to cost-effective data collection and intelligent evaluation of the reasons for malfunctions, structured plant improvements and active avoidance of malfunctions can already be realized today. In addition, automated information processes can significantly reduce OEE losses. In this way, it is possible to act proactively and within a few minutes, where without the technological possibilities this was usually only conceivable at the next shopfloor meeting. These possibilities for increasing effectiveness have the potential to reduce costs in production, make efficient use of scarce capacities and improve delivery services. With first successes in the bag, Zentis now wants to use oee.ai for predictive and prescriptive analyses. Hereby, probable events are to be considered in order to reduce costs in production, to use scarce capacities efficiently and to improve the delivery service. It must always be kept in mind that evaluations and algorithms only give an indication of existing or expected plant behavior. Whether this indication can then actually be converted into an increase in productivity must be validated by the company.