1.1.1 Motivations for Ontology

The ability to fix a domain vocabulary and its meaning in the context of use in this manner is

critical for true concurrent engineering. A large engineering or manufacturing project involves

the resources of many different clusters of cooperative agents (human or otherwise) in the given

endeavor. Each cluster makes its own contributions, and the overall success of the project

depends in large measure on the degree of integration between those different clusters throughout

the development process. A key to effective integration is the accessibility of rich ontologies

characterizing each of the domains addressed by each cluster. For instance, access to a

manufacturing ontology that includes constraints on how a given part is manufactured can aid

designers in their design of a complex product by giving them insight into the manufacturing

implications of their design concepts. Similarly, access to an engineering ontology that includes

constraints on how a given part is to function given a particular shape or fit can aid process

planners in their development of the appropriate manufacturing processes. A commonly

accessible collection of relevant ontologies thus permits more efficient sharing of information

arising from various sources within the enterprise.

A related motivation for ontology capture is the standardization of terminology. An enormous

problem in the coordination of large projects is the diversity of backgrounds the various kinds of

engineers bring to their respective roles. As a consequence, many engineers use similar

terminology in many different ways with many different connotations. Because of such

differences, the information that one engineer intends to convey to another may in fact become

garbled; in the best case, such miscommunications can be responsible for a great deal of lost time

and resources; in the worst case, such miscommunications can result in the loss of life.

Consequently, it is often necessary in the course of a large project to standardize the relevant

vocabulary. The ontology capture method provides a principled method for carrying out this task

efficiently and effectively, and maintaining the results of the task in a robust, accessible form.

This suggests another, strong motivation for ontology: reusability. Among the most significant

problems in engineering and manufacturing in general is the redundant effort expended in

capturing or recreating information that has already been recorded elsewhere. For example, in

programming, the same kinds of routines (e.g., in the design of user interfaces) are often used

again and again in different programs by (in general) different programmers. Consequently,

enormous amounts of time and effort have gone into reinventing the wheel time and again.

Recognition of this problem has led to the development of vast libraries that contain often used

routines which programmers can simply call straight into their programs, rather than having to

duplicate the function of existing code. Engineering and manufacturing face the same type of

6

problem. Manufacturing domains, for example, share many common features that are

independent of the specific characteristics of a given domain; and the more similar the domains,

the more such features they share. Rather than encoding this information all over again in every

new setting, analogous of the concept of a programming library, one can imagine collecting this

common information into ontology libraries, (i.e., large revisable ontology databases of

structured, domain-specific information.) Information in these ontologies can then be reused and

modified to suit the needs of the moment. Moreover, because ontologies provide a standardized

terminology by their very nature, no special additional effort need be expended on fixing domain

terminology. It must be emphasized, however, that, despite the potential size of a given

ontology, the concept itself is highly scaleable; that is, ontologies are no less effective in smaller

contexts than on very large ones.