The synergy of disciplinary expertise
- IWR is actively bridging disciplines.
Developing systematic methods for mathematical and computational modeling and simulation in the sciences has a successful tradition in Heidelberg for more than 30 years. The IWR was built up in this highly productive research environment by Heidelberg University and the state of Baden-Württemberg in 1987. ( ... more)
From the beginning it was of great importance for the center to collaborate with research teams, institutes and faculties of Heidelberg University and abroad as the strictly interdisciplinary approach is a key idea of the centers concept.
Consequently IWRs network is built up by the intense research activities of more than 40 main research groups, and 21 independend and junior research groups which belong to the center today. The network consists of scientific and insitutional contacts, and is a driving force for the esablishment of academia exchange programs and intense industry collaborations. It radiates deeply into the international scientific community and attracts young researchers worldwide.
IWR fosters the interdisciplinary dialogue
The scientists of IWR constantly put their know-how in scientific computing to use in new research fields and don’t shy away from working in inter- and multidisciplinary projects with other scientific areas, as scientific computing as an instrument for gaining knowledge, is not limited.
Here are some examples for IWRs impetus on the sciences:
- The pioneering projects open new insight
To promote scientific computing in new application areas such as economics, finance and the social sciences, the humanities and cultural sciences, but also medicine and public health, the IWR has identified twelve so called Pioneering Projects. These projects cover important scientific areas in which the potential of modeling and simulation has yet to be fully exploited. (...more)
- IWR initiated the SCCH in Heidelberg, a conference on scientific computing and cultural heritage.
- IWR strengthens the computational humanities
The scientific analysis and preservation of cultural heritage presents numerous challenges that require state-of-the-art scientific computing methods to accurately and efficiently access culturally and scientifically invaluable material. With recent large-scale digitization projects, the humanities are gaining access to previously unknown quantities of visual data, which require automatic algorithms to make this data accessible that allow for quantitative analysis of the objects of interest and that assist scholars from the humanities in focusing on the relevant details. Under the umbrella of IWR Assyrologists and Mathematicians developed together new insights and jointly publish results.
>>> Visit also the IWR Pioneering Projects website for more information.
- Finite element simulation of tissue deformation due to external forces
- IWR's expertise in computational bio sciences and medicine
Complex processes in human physiology, systems biology, proteomics or in the cognitive sciences, are typically composed of hybrid models including detailed mechanistic models based on reaction-diffusion equations and the black-box type of statistical models. For both types parameters have to be fitted and to be learned. IWR's researchers develop methods that bridge the loosly connected area of model training and validation of statistical models with parameter estimation.
The interdisciplinary research projects are characterized by involving the excellent and innovative research infrastructures, that are located nearby. The unique environment in the field of bio sciences at Heidelberg promotes our projects outstandingly.
- When embedded in a suitable environment, excited or ionized atoms and molecules can hand over their excess energy to their neighbors extremely efficiently via the interatomic (intermolecular) Coulombic decay (ICD) mechanisms.
- IWR and the gain of knowledge in material and earth sciences, in computational chemistry and computational physics
All non-trivial systems such as live cells, organs or environmental systems at different scales and from different compartments, are characterized by and thrive on a high degree of spatial ordering and complex, often nonlinear, dynamics. It is impossible to understand or model these systems without taking their spatial ordering into account, and it is therefore imperative to deploy appropriate imaging techniques and to develop new methods for the analysis of the raw data they generate. Our physical envrionment, for example, comprises a network of highly nonlinear and strongly coupled processes, which are in a delicate equilibrium: External forces like the astronomical parameters that modulate solar energy and internal forcing from a multitude of processes that include volcano eruptions, earth quakes, anthropogenic emissions of greenhouse gases etc are forming our planet. At IWR scientists try to model and predict this large-scale complex system.
- IWR designs and analyses efficient and robust numerical methods for solving problems arising in sciences and engineering.
- Mathematics & Informatics
The application areas of modern mathematical and computational methods are ever growing. In this regard optimization-based methods play an ever increasing role as they provide intelligent algorithms for modeling and simulation, which are an integral part of modern sciences.
In addition the large amount of data produced in the sciences (f.ex. by image sizes, number of images) has to be computed. Computation of large data sets is often very expensive, if at all possible. Often the dataset is too large and the model is highly nonlinear and unsteady. 'Goal-oriented' adaptivity (mesh refining and coarsening) has to be applied in order to reduce the computational effort and to guarantee the accuracy of the results.
- At IWR models for prizing in commodities are developed. A detailed analysis of chosen multi-commodity markets is of great importance and global supply, demand, and trade of a wide product range need to be analyzed.
- Social and political sciences, the economics: scientific computing helps to understand global networks in detail
Commodity trading and management of associated risks have been gaining dramatically in volume, importance, and complexity over the past years. This is due to increased demand and to partial exhaustion of resources, as well as to a stronger influence of technological, political, social and environmental factors. The network of interacting factors have to be analyzed including supply and demand, the intrinsic dependence between different markets, e.g. between fuel and power markets, the technological processes, or transportation and storage for the better pricing of commodities and the management of revenue and risk.
- Motion parallax, facilitated by interaction and real-time rendering, is a form of error correction in visualization.
- Simulation software and scientific visualization
The implementation complexity of modern numerical algorithms involving, e.g., adaptivity, high-order discretization, and optimal-order solvers for challenging multi-physics and multi-scale application problems is clearly beyond the reach of a single doctoral project, especially when optimization and treatment of uncertainties make the use of parallel high-performence computers mandatory. Note, however, that the requirement of reusability adds yet another level of complexity to the software.
- Global distribution of tropospheric NO2. The observed patterns indicate the locations of ma jor sources.
- Model-based image processing
Image processing techniques have been adopted in numerous novel application domains, such as in the natural sciences and in industrial scenarios. Here, usually a high level of prior knowledge is available, which schould be fully exploited in order to extract the relevant parameters from the image data. In model-based image processing, such prior knowledge is formulated as systems of PDEs or as probabilistic graphical models.