Eliciting information for product modeling using process modeling

Ghang Lee, Charles M. Eastman, Rafael Sacks

Research output: Contribution to journalArticlepeer-review

40 Citations (Scopus)


A product model is a formal and structured definition of product information. The most common procedure for defining a product data model is to first describe the business and/or engineering process in a formal process model, then to create a product data model based on the process model. However, there is a logical gap between process modeling and product modeling methods in the current ISO 10303 standard product modeling process. We propose a new formal approach, called Process to Product Modeling (PPM) in which process and product modeling can be logically linked. This paper focuses on the semantics and syntax for the Requirements Collection and Modeling (RCM) of PPM. The new RCM method aims to model heterogeneous business processes and their information flows. The major difference between the proposed method and traditional requirements collection methods is that it provides a theoretical linkage to integrate process models and a set of specific information items used in them. This theoretical linkage enables modelers to capture the contents, scope, granularity, and semantics of information used in activities, which are depicted in process models. A PPM method, called Georgia Tech Process to Product Modeling (GTPPM), has been developed and was deployed for the North American Precast Concrete Software Consortium; examples from this effort are presented. Experience to date indicates that the new RCM method and the GTPPM tool hold the potential to improve and expedite elicitation of information for product model development.

Original languageEnglish
Pages (from-to)292-307
Number of pages16
JournalData and Knowledge Engineering
Issue number2
Publication statusPublished - 2007 Aug

All Science Journal Classification (ASJC) codes

  • Information Systems and Management


Dive into the research topics of 'Eliciting information for product modeling using process modeling'. Together they form a unique fingerprint.

Cite this