A Benchmark Diagnostic Model Generation System

Abstract--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 a three-step process for synthetic model generation: (1) domain analysis; (2) topology generation; and (3) system-level functional model generation. To demonstrate our approach on two highly different domains, we generate models using this process for circuits drawn from the ISCAS benchmark suite and a processcontrol system. We then analyze the synthetic models according to two criteria: topological fidelity and diagnostics efficiency. Based on this comparison we identify parameters necessary for the auto-generated models to generate benchmark diagnosis circuit and process-control models with realistic properties.

• Type: Journal
• Year: 2010
• Where: TSMC
• Authors: Jun Wang, Gregory M. Provan
• Keywords: Artificial Intelligence, Benchmark Models, Model Generation, Synthetic Models, TSMC 2010