ILNumerics^® Case Studies

ILNumerics is used by engineers and scientists around the world. We want to give you an impression about how ILNumerics changes the way they are working. Learn how our software is applied in science and industry:

A more general insight into common applications for ILNumerics is given below:

Quality Assurance

A semi-conductor company produces wafers and needs to improve quality assurance. It commissions a software application that visualizes deviations from the norm on the basis of points measured on the wafers’ surface.

Without ILNumerics, the prototypical algorithm is developed with numpy, an extension to the popular programming language Python. In order to incorporate this algorithm into the final software application, interfaces to Python help accessing numpy from .NET. To realize the visualizations, the developers use third party modules. The final application contains three technologies: Python, .NET, and the modules needed for the visualization feature. For all future changes, appropriate expertise and licences are needed.

With ILNumerics, both the algorithm and the visualization feature can be implemented into the .NET application code directly, which means the application becomes much more stable and maintainable. A connection to external modules is not necessary. Future updates can be made with low costs. The company saves 30 % on development costs.

Risk Management Software

In order to simulate credit losses, a bank wants to develop a new software application. This application is to be used in every branch of the bank.

Without ILNumerics, the research and development department uses mathematical prototyping software to create a prototype. In the second stage, this algorithm is implemented into the new software application. To achieve the performance that is needed, the developers decide to use C++ and the whole algorithm has to be re-written from scratch. All future changes to the algorithm must be applied to both domains – separately, for each platform involved.

With ILNumerics, the research and development department designs the algorithm directly in .NET. Because of the excellent performance of ILNumerics, the same code can be used for the new software application. All future changes take little effort as they can be made directly in the application code. A single application is created. It runs on all platforms without modifications. This cuts costs by 50 %.

Optimizing Performance

The research and development department of a large automotive company has been using .NET applications for mathematical tasks for years. These tools are used on both Windows and Linux systems. In order to improve certain operations, the execution speed of some of these tools needs to be optimized.

Without ILNumerics, performance-critical parts are implemented into native modules written in C and linked to the corresponding software tools. The tools lose their wide platform support and future versions for Windows and Linux must be developed separately.

With ILNumerics, no native modules are needed: the performance critical parts can be implemented directly in a high level .NET. language. The tools still support multiple platforms – they run on both Windows and Linux. The company saves 30 % on development and maintenance.