One of the requirements for an intelligent system is to construct a virtual reality in the computer. For materials development, the “reality” is a laboratory or a production line used to improve some aspect of a material by changing its composition, processing parameters, service conditions, etc. In the case of information in libraries, the reality is a search that uses a thesaurus and bibliographic (fact) databases. The greatest barrier faced by users of materials information is that of accessing necessary information through both a librarian's view and a scientific/technological expert's view. One of the objectives of a virtual production line is to reduce this barrier through user-friendly interfaces.

In designing materials, two typical approaches are combined to solve a given problem. The first is a top-down approach, in which a number of requirements are resolved to a set of possible and practical solutions for satisfying these requirements to a certain level. Almost all alloy development has followed such an approach. The second approach is a bottom-up approach, where different materials properties are described on the basis of underlying theories, preferably using first principles and fundamental data. But the vast number of possible materials makes the bottom-up approach unrealistic on its own. Instead, semi-empirical approaches are needed to bridge the gap between ad hoc data sets for practical applications and results produced by the bottom-up approach using fundamental data and first-principle calculations.