A Contribution to Quality Assurance of Ship Product Model Data

摘要

An approach for automatically detecting faulty ship product model data defined in the structural design process is presented. Quality criteria are covering a wide range of design and production knowledge. The criteria are formulated in form of rules and are applied in automated error detection processes. A rich set of parameters serves to configure the knowledge represented by the rules to allow for a universal usage for different projects or even different ship yards. For rule execution a multipurpose rule engine is applied. Based on this rule engine the implemented system for the automated checking of quality criteria will be described. Independently from the C AD-system used for the definition of the ship structural steelwork, checks can be performed on the product model data at any time. The rules used for representation of the quality criteria can be formulated and selected by the engineers. By this quality control is feasible at any phase in the design process as well as at data exchange procedures between partners collaborating in the ship design. Applying the system being presented, errors can be detected at an early design stage, the time required for error elimination can be reduced substantially. The overall costs in design as well as in production will be decreased. Examples serve to show the potential of the system which could be called a "spell checker" for ship structure model data.