Effective consistency management for large-scale product data

Abstract

Abstract The research presented is aimed to develop and investigate an effective method for management of consistency of large-scale product data. Such data are usually driven by formal models in EXPRESS language being part of the STEP standard (ISO 10303). To be consistent and unambiguously interpretable by computer programs the data must satisfy syntactic and semantic rules defined by the standards. Although ACID principles (Atomicity, Consistency, Isolation, and Durability) are widely recognized and recommended for any information system, it is hard to guarantee the consistency of such complex data. Complete semantic validation requires extremely high costs, often exceeding the processing time of individual transactions. Periodic validation or validation on user demand is possible, but at a high risk of losing data that become useless in case of rule violation. In the paper an effective incremental method for semantic validation of product data is presented. It is guaranteed that the final data revision is consistent if only the original revision was consistent and the spot rules were not violated. Static analysis of the model specifications is applied and a dependency graph is formed. The dependency graph enables to identify the spot rules and the data that should be inspected against the rules. Computational experiments prove the effectiveness of the method in conformity to complex large-scale product data managed under an innovative platform PDMhub.