A Common Service Middleware for Intelligent Complex Software System

With the rapid development of the Internet of Things (IoT) and artificial intelligence (AI) technology, various intelligent complex software systems (i-CSS) are increasingly popular, becoming one of the most important software system development paradigms. Its inherent growth construction and adaptive evolution properties pose new challenges to existing software design and development methods. Especially, how to achieve growth construction by quickly reusing existing excellent software resources, and how to establish data flow across system boundaries around the business flow to achieve adaptive evolution based on data intelligence. Facing the above challenges, this paper proposes novel data-oriented analysis and design method (DOAD), microservice and container-based mashup development method (SCMD). On this basis, the paper implements i-CSS common service middleware to support the above methods in engineering. In a real cloud-based PHM system and the other three industry projects, the proposed methods and middleware are used for application verification, the results show that they can greatly reduce the complexity of i-CSS design and development, reduce the ability threshold of the i-CSS development team, improve the development efficiency of the development team, reduce the team development workload by 31.5% on average, and help the i-CSS team effectively cope with the challenges of growth construction and adaptive evolution.