Research On Reconfigurable Manufacturing Enterprise Management Information System

The development of the enterprise management information system is confronted with the challenge of a dynamic business environment of manufacturing enterprises. Although there are some achievements in current reconfigurable information system development based on changes of business models, it has not succeeded in manufacturing enterprises. One of the key reasons is the lack of the fitting methods of transforming characteristic business models into correspondent software models. Therefore, the methods and key technologies of business model driven reconfigurable information system development are investigated in detail. The main efforts and contributions of this dissertation are as follows:By establishing business model ontology for clearly modeling characteristics of manufacturing enterprises in a standard way, a new business modeling framework with clear semantic meaning and object-oriented features is set up and therefore lays a theoretical basis for translating business models to component-based software system models. Moreover, the formal definitions of checking methods are proposed to examine both completeness and consistency of a business model with support from business model ontology.The mapping method for translating business model into component-based software system model is put forward by analyzing structure heterogeneity of the models. A new approach of business model driven reconstruction of management information system is developed based on the mapping method. By means of the interfaces of the business model, the specific methodology based on dynamic interpretation of the configuration rules is developed to introduce new extended functions that are not defined in business modeling framework into management information system. The methodology supports flexible dynamic function extensions which comprise user interface customization, the process reconfiguration and the data reconfiguration.To ensure the reconstruction system to meet the quality requirements of the corresponding business domain, the dissertation presents two testing and analysis models: tracking model and validation model. The tracking model applies imprecise probability theory and Bayes rules to software defects analysis and integrates experts'judgments with statistical data to obtain credible software defects assessment. The validation model can properly make quality assessments from several aspects of the reconstruction systems:functionality, efficiency, security, ease of use and reliability.The research achievements have been applied to customization and reconstruction platform for enterprise management information system. The practical application of the platform to PLM (Product Lifecycle Management) system for power plant boiler industry proves that it has the advancement and practicability in assisting enterprises to obtain reconfigurable management information system.