Specific MODELLINg systems

Special parts of the specific software development are modelling and simulation systems. ALL undertakes the elaboration of modelling and simulation systems for supporting the solution of various complex tasks. ALL seeks and works out the most appropriate modelling instruments and methods for the task during the elaboration of the system. Applied Logic Laboratory has developed special advanced mathematical, logical methods and informatics methods for modelling complex systems. Tools include algorithms and neural network-based self organising methods for modelling, prediction, identification and approximation of multivariate processes.

The company has developed innovative neuronets with active neurons specially arranged in multiple layers. Such neuronets are able to automatically generate the structure of the complex model and to predict the response of the system in "what if" scenarios. Self-organising modelling is supported by three methods: self-organising group method of data handling, bistable gradient neural networks and threshold networks. These techniques can be combined with special statistical and logic-based method if both quantitative and qualitative information should be encapsulated in the model.

The company has developed sound methodologies of combining different types of models, like static and dynamic, logical and numerical system descriptions. Such combination of sub-models is of primary importance when developing models of "in silico" medicine trying to synthetise what we know about structures and processes at genome, cell, organ and organism level.