TOHIVA

Validation and analysis of 200 simulation runs in automotive engineering.

Analysis of energy balance groups.

The applied research project TOHIVA researches novel visual analysis technologies for high-dimensional data in automotive engineering, industrial manufacturing, and the energy sector.

The project TOHIVA is driven by application needs in automotive engineering, industrial manufacturing, and the energy sector regarding the effective analysis of high-dimensional data for tasks such as data quality assessment, statistical model building, and effective decision making. There is a significant gap between the potential of visual analysis technology to increase the efficiency and effectiveness of these tasks and the current adoption of the technology in daily routine. The goal of the project TOHIVA is to bridge this gap across the diverse markets of the project partners.

For AVL List GmbH, a key task is the optimization of engine designs by analyzing multiple simulations runs. For Plasmo Industrietechnik GmbH, the challenges include the online control of manufacturing processes (especially welding) as well as the offline analysis and optimization of production processes. For HAKOM Solutions GmbH, the focus is on the management and forecast of time series data in the energy sector. For Austrian Power Grid AG, the key value is the ability to identify potential for optimization in forecasts and power grid operation. As crosscutting aspects, all tasks involve complex and high-dimensional data such as technical indicators, process parameters, and time series of energy production and consumption. The software platform Visplore is a common basis for all research activities in TOHIVA and enables efficient technology transfer across the boundaries of the various markets.

In the scope of TOHIVA, VRVis will research new Visual Analytics methods to address specific tasks, including multi-criterion decision-making, data quality assessment, quality control of production processes, and the identification of high-quality forecast models. A key goal of TOHIVA is to enable the partners a tight integration of this new technology as part of their workflows. This includes the challenges of guiding both expert and novice users through powerful interactive visualization and of enabling a seamless analytical interplay of interactive visualization with standard third party software for data analysis like Matlab, R, and Python. As benefits, the project results are expected to significantly speed up the addressed tasks, improve the quality of concrete results like models and decisions, and increase the confidence in these results.