Data-Driven Maritime Processes Management Using Executable Models

Abstract

In this paper we describe a decision support system for maritime traffic and operations, based on formal models and driven by data from the environment. To handle the complexity of system description, we work with a decomposition of the system to set of abstraction levels. At each level, there are specific tools for system functionality specification, respecting particular domain point of view. From the business level point of view, the system consists of processes and vehicles and facilities over those the processes are performed. From the engineering point of view, each process consists of a set of devices, that should be controlled and maintained. Software engineering point of view operates on reading and converting bytes of data, storing them into variables, arrays, collections, databases, etc. For complex trading processes management purposes we need to cover all levels of abstraction by specific description, suitable to model and automate the operations on each particular level. As a case study we use salmon farming in Norway. The system implementation is based on Reference Petri nets and interpreted by the Petri Nets Operating System (PNOS) engine. This approach brings formal foundations to the system definition as well as dynamic reconfigurability to its runtime and operation.