Design

The digitization of processes is complex and error-prone. That is why manufacturing processes are monitored using statistical process control methods. The aim of the presented project was to answer the questions how the data basis for the use of the quality control chart (QRC) can be extended from random samples to near real-time data and how the implementation of the solution should be done. The software solution was developed and tested in the Fischertechnik learning factory. It could be shown that the data from the learning factory is suitable to be displayed in a closely timed manner and to be evaluated by means of process indicators of the QRK. In this way, errors can be avoided and capacities saved. (Only in German)

Artificial intelligence has the potential to bring companies and entire industries to a completely new technological level. The prerequisite is data with a high degree of maturity, with which companies can automate complex processes, calculate forecasts or create analyses. With the right data strategy, structuring and achieving the necessary data quality are no longer dreams of the future.

Abrasive blasting is often used to clean work pieces. During the process an abrasive medium is propelled with compressed air toward a given surface. Common abrasives are sand, glass beads, steel or corundum. For safety reasons the blasting process is carried out in closed blast cabinets or rooms. Abrasives and cut off material are filling the air so that the visibility is limited. Quality assurance and safety monitoring of workers in blast rooms are therefore difficult which is essential e. g. in atomic power plant demolition. This article describes the development and test of a camera to improve this situation. Compressed air flows through the camera housing to keep particles away from the lens. The air flow was optimized by computational fluid dynamics. A prototype was made by 3D printing and tested in an blast cabinet.

Procurement has the task to supply an organization with required but not self-produced goods. The goods vs. payment exchange with suppliers is laid down in contracts. “Electronic contracts" or “Smart Contracts” represent the logic digitally and thus enhance transparency. This can still evolve. In the future, improved algorithms and artificial intelligence will not only be able to administer contracts, but also to design them. This article presents the status quo of "Smart Contracting", places it in the "Legal Tech" topic and shows how artificial intelligence could be used in procurement.

Optimisation is an essential task in many situations. The class of evolutionary algorithms is a population-based, heuristic technique for optimisation. They allow the optimisation of multi-modal problems even with distorted search spaces. They can propose several solutions instead of just one. An important aspect of evolutionary algorithms is encoding search space candidates. In the optimisation of processes, this is a non-trivial task. This article describes a successfully tested encoding.

The competitive pressure in the contract logistics industry is intense. Logistics providers must respond to tenders quickly and with convincing concepts. This article presents initial approaches to how logistics process planning in tender management can be supported using supervised learning methods. Under the premise that similar processes from past projects can be transferred and adapted to a project to be planned, an AI-based assistance system suggests appropriate process steps and MTM (Methods-Time Measurement) codes during planning. This procedure can accelerate process planning and lead to an increased quality of logistics processes to be planned. (Only in German)