Industrial Data Science combines expertise in a variety of areas, including data analysis and digitization, with a focus on industrial applications and best practices to drive business growth, productivity, and long-term learning.
Digitization, the Internet of Things (IoT), and the industrial Internet and Industry 4.0 solutions are transforming entire industries and enabling them to collect vast amounts of data at different aggregation levels and types. These include big data and streaming data, structured and unstructured data, text, images, audio or sensor data. Data Science, Data Mining, Process Mining, Machine Learning, and Predictive Analytics promise to generate enormous efficiency and competitive advantages.
Typical use cases are demand forecasting, predictive maintenance, prediction of machine failures and their prevention, the prediction of critical events, quality forecasts for process optimization and shopping basket analyzes. But also the prediction of assembly plans for new product designs in sectors such as automotive, aerospace, energy, mechanical and plant engineering, metal, etc. belong to these use cases. An essential criterion for the successful implementation of industrial data science projects is, in addition to the development of promising business cases, the efficient organization and execution of the analysis work. In doing so, the IPS Engineers specialize in mediating between the two worlds to be unified, the application world, with the irreplaceable domain experts, and the data analysis world, with the equally indispensable data scientists.
The Simulation-based data analysis and IoT integration are elementary steps towards the holistic implementation of data science in production; only by using data across all process steps can these be used profitably.
Modern production systems with widely implemented IoT architecture and widespread use of cyber-physical systems usually provide sufficient opportunities to derive data from the process and provide it for analysis. The main difficulty, however, is not in making the data available but in their meaningful processing and lucrative use. The problem here is, in particular, the use of statistical data analysis with a focus on past data without these anticipatory use.
Better integration of the (already existing) IoT techniques into the process analysis and the more holistic use of simulation as a tool of process optimization can reveal many potentials. These key technologies are critical success factors for many areas of industry and essential for maintaining and expanding competitiveness. Above all, the IPS Engineers here focus their competences in data analysis for systems of predictive maintenance. An important and necessary step for the modernization of maintenance processes.