A MDA APPROACH FOR PRODUCT DATA QUALITY THROUGHOUT VACCINE PRODUCT LIFECYCLE

Data quality is widely considered as a very serious problem for the majority of companies due to the specificities of each business context and the lack of adapted solutions. We present in this paper the benefits of a Model-Driven Architecture (MDA) in ensuring the interconnection of different business contexts' specifications by providing a linked structure of models. This enables to generate bridges that connect implementations in different platforms. In this way, the systems interoperability can be satisfied throughout product lifecycle. The MDA approach is widely considered as a methodology for software generation from models, with a focus on enterprise and business models. Deploying a MDA approach in the supply chain context of vaccine industry allows us to deal with product data quality. In fact, it helps to translate some business models at a computer independent model through the MDA framework to generate a newly data model as well as some business rules and recommendations helping to communicate data models. To ensure the quality of product data in the Enterprise Resource Planning (ERP), the new generated data model is compared with the one of the ERP and proposed mapping rules are structured through a data reference model. Finally, three levels of reference frames are proposed to ensure the share and the traceability of generated metadata in order to ensure the evolution of defined models and preserve product data quality throughout its lifecycle.