Search

Industrial manufacturing systems are subject to random events which disturb their working process: resource unavailability, out-of-stock condition, changes in orders, etc. It is difficult to decide whether a schedule needs to be adjusted or not by manual when random events occur. This paper presents a task graph and simulation based method to analysis the effect of random events on scheduling. With this method, schedulers can decide whether a schedule needs to be rescheduled or not. A summary about random events in manufacturing systems is described at first, then the process that analysis scheduling systems with task graph and simulation method is proposed, a case to prove the feasible of this method is given at last.

To solve the problems of network bandwidth bottleneck and insufficient service ability of networked manufacturing application service provider platform, a grid-based more-for-more service model was put forward on the basis of the present one-for-more model which adopts the central server. And the application service of manufacturing resource offered by the platform was executed cooperatively among more than one grid service nodes. A service-oriented classification model of manufacturing resource based on ASP platform more-for-more model was established. And the RDF-based metadata specification language was realized to enhance the ability of interoperability and access. In the aspect of manufacturing resource organization and discovery, the incorporation of distributed and centralized mechanism was adopted. Different types of manufacturing resource were managed by the corresponding ASP nodes, and these self-governed ASP nodes had constructed the grid-based network topology. Furthermore, the algorithm as well as performance evaluating of resource discovery was described. Finally, the implementation technologies of the platform were introduced, and feasibility and practicability of the proposed model were verified by the development and preliminary application.