Research On Bulk Port Group Ship-Scheduling System Based On Mutli-Agent And Ontology

With the development of global economic and the port throughput increasing successively, the trade demands become more and more internationalization, which brings opportunities to the termimal and also presents the enormous challenge to the trade service. The traditional way of ship-scheduling and production resource allocation have been already unable to satisfy the demand of ship management and cargo handling. For the termimal,it has to pays attention both on its own benefit and the request of customers by promoting the quality of freight unceasingly. Therefore how to realize a system, which can adapt the ship-scheduling of the bulk port group, and improve the level of production scheduling for the whole group based on the increasing of ship operation efficiency, has become the key of infomationization and intelligentialize for bulk port ship-scheduling in China.This dissertation analyzes the manufacturing operations deeply, constructs the prototype of bulk port port ship-scheduling system by using the structure of multi-Agent and reasoning model of ontology. Main contents and results are as follows:(1) Proposed double-layer architecture of bulk port group ship-scheduling based on multi-Agent.Bulk port group ship-scheduling is a dynamic and open production process with multi-link interaction and interaction of multiple factors. The process consists of two phases:the group layer scheduling and the termimal layer scheduling. According to this characteristic, this paper plays the full advantages of Multi-Agent system on distributed computing and collaboration, and proposes the double tier architecture of bulk port group ship-scheduling based on Multi-Agent.The double-layer architecture is made up by the scheduling planning of terminal and the scheduling adjusting of group. Besides, this paper also constructed six kinds of agent to accomplish all the functionality, and these agents are separately responsible for perception of the external environment, ontology reasoning and algorithm model.(2) Constructed a coordination mechanism for Multi-Agent systemWhen dealing with the coordination problems among the Agents, this paper has introduced the coordination mechanism which is based on the port group organizational structure model as well as the consultation mechanism based on conflict resolution. In the conflict resolution mechanism, this paper proposes the dynamic compensation policy, which can provides a relatively level for all the Agents involved in consultation. In order to realize the balance between the global objectives and local targets in the port group ships scheduling.(3) Put forward the ontology construction method for bulk port ship-scheduling domainIt is an effective way to use ship-scheduling rules, accumulation of knowledge and experience to improve the quality of bulk port ship-scheduling. In order to solve the domain knowledge reasoning problem about the bulk port ship-scheduling, this paper put forward the ontology construction method in the bulk port ship-scheduling domain.The methodology, which based on the characteristics of multi-disciplinary cross in the domain of the bulk port ship-scheduling, has proposed the concept "Domain Correlation" in further. Besides, this paper also described multi-disciplinary ontology integration methods and models in detail, and at the same time constructed the ontology integration reasoning system for knowledge reasoning.(4) Establish the dynamic berth assignment model based on improved ant colony algorithm and the combination of ontology reasoning modelBerth assignment problem is the core of the port ship-scheduling, how to determine the order of the ship's operations, the match of the ship and the berth is the key of the berth assignment problem. In the study process, this paper makes the time of ship in port minimized as the optimization of key indicators, considering the different importance of different ships, this paper put forward a first-class model for the use of ship resources, through the improvement of the traditional ant colony algorithm, the ship-berth matching relation constraint matrix formed by ontology reasoning. Finally, the model based on improved ant colony algorithm and ontology reasoning was applied to the combination of the dynamic berth assignment problem solving process.(5) Construct the bulk port group ship-scheduling system prototype based on the Multi-agent and OntologyIt proposes the integration and technology architecture of the bulk port group ship-scheduling by using Multi-Agent and Ontology. The integration architecture mainly describes the integrated approach and the data exchange approach between the ship-scheduling system and PPMIS (Port Production Management Information System);Technology architecture is about technology combination of Ships scheduling system prototype. Finally, this paper describes how to use J2EE framework, JADE, Jess, Protege and other technology, to implement MAOPSS (Multi-Agent and Ontology Port-group Ship-scheduling System), and also shows some important features and interface.