The creation of high-quality data models is critically important to IT projects that require database design or redesign. There is no universally recognized set of rules for the design of data models, aside from such concepts as normalization. Matthew West (1996) provides one well-respected and concise set of rules for building a high-quality data model. David Hay (1996) expands on this concept of data model rules to propose that good data models are built from patterns.  Read the full paper here.