Automated model generation for complex systems

It is critical to use automated generators for synthetic models and data, given the sparsity of benchmark models for empirical analysis and the cost of generating models by hand. We describe an automated generator for benchmark models that is based on using a compositional modeling framework and employs graphical models for the system topology. We propose two novel topological models, and demonstrate their advantages, over existing graphical models, in better capturing the topological and functional properties of a class of real system, discrete circuits. We compare generated models to real systems (drawn from the ISCAS benchmark suite) according to two criteria: topological fidelity and diagnostics efficiency. Based on this comparison we identify parameters necessary for the autogenerated models to generate benchmark diagnosis circuit models with realistic properties.