Our team provides professional support for data analysis and processes or product optimization, also with respect to the requirements of Industry 4.0. We assist you concerning data acquisition and data integration, we develop customized systems according to your individual requirements and we help generate the interface to your internal IT and process environment. For these purposes we use the latest methods of data analytics, machine learning and global optimization relating to the latest results of research and development.
“As the leading expert in the application of evolutionary strategies and meta-models to hard industrial optimization problems, Prof. Bäck has always delivered highest quality results for BMW. We truly appreciate our collaboration and warmly recommend a collaboration with Prof. Bäck to any industrial partner who is in the need of solving tough optimization, modeling, and prediction problems.”
“Besides researching and developing the most suitable individual software components, successful multi-disciplinary optimization in an industrial context requires a system level analysis of the problem including issues of workflow, interoperability and resource sharing. Professor Bäck is world-wide one of the leading researchers in computational intelligence and his contributions to the development of new algorithms and methods cannot be overestimated. However, from our perspective, it is his unique analytical ability to amalgamate systems engineering with computational intelligence which makes him such a valuable partner for our institute.“
"Guaranteeing highest quality standards of Ford vehicles requires a continuous evaluation of quality data from a variety of data sources. With divis GmbH, we have access to a competent partner in the areas of data warehousing, data analysis, and data mining. We would like to express our gratitude to Prof. Baeck and his team."
"Our past experience limited our trials to include only familiar ingredients, and the lengthy process of attempting each combination had led us to declare the desired product infeasible after three months of research. However, with ClearVu Analytics, we satisfied all the technical requirements faster than with traditional processes, and now have a new product we will bring to the market."
„Through the support by the divis experts, we were able to develop a model which allows us to optimize the parameter settings of the coating process. Normally, this requires a large number of costly and time consuming laboratory experiments. The competent support provided by divis made it possible to reduce this effort significantly and make the whole process more efficient. The mathematical models derived by divis facilitate a better understanding of the impact of changes in the process control parameters on the resulting process behavior. This insight contributes significantly to process understanding as well as process robustness and quality at the client site.“
„We are still implementing and evaluating our new model, but clearly we can now predict micro results for new formulations using known components at this time. We will continue to expand our use of the model, but critical to this process is internal data collection. But even this early in the project, we have a better picture of how our materials interact and the effects to the wipe. And we have a set of working rules for our formulators.“
“Finally, we are now in possession of the tools and information needed to give us optimal solutions – this is exactly the kind of breakthrough we have been looking for to significantly change our business processes.”
“Former ThyssenKrupp Uhde GmbH in Dortmund, today called Business Unit Process Technologies, part of ThyssenKrupp Industrial Solutions AG (a world-wide operating engineering corporation for the planning and construction of chemical plants) is using divis’s data mining software by divis GmbH for data driven modeling tasks. The goal is for chemical processes to describe cause and effect patterns faster and better than before, and to evaluate the process parameter’s impact through sensitivity analysis. The results obtained so far clearly illustrate the capabilities of the special methods provided by divis, such that ThyssenKrupp Industrial Solutions is now considering additional projects.”
All applications in which products or processes are intended to be optimized are suitable for our methods. You can use data from experiments, simulations or directly from the process. Based on these data we optimize the product or the process.
divis provides individual support according to your personal requirements—from the first consultation meeting to the final optimization. Of course, our specialists are always at your disposal.
We will be happy to advise you concerning the details of a thorough process analysis including all aspects from sensor technology to the final product or process.
We offer a process related system analysis to abolish inflexible maintenance schedules (reactive or preventive maintenance) and replace them by dynamic and cost efficient ones (predictive maintenance).
Upon request, we will develop an individual version of our software that contains a user interface which is adjusted to your personal requirements.
News for you at a glance
At the International Joint Conference on Neural Networks (IJCNN) https://www.ijcnn.org/ we will present our contribution entitled “Automatic Configuration of Deep Neural Networks with Parallel Efficient Global Optimization”.
The combination of complex constraints, various target functions and time consuming simulations (e.g. CFD) requires the development of new algorithms.
divis intelligent solutions GmbH
44227 Dortmund, Germany
+49 231 97 00 340