Dynamic data driven re-engineering of digital models evaluated with an agent-based architecture

• Type: Article in International Conference Proceedings Book
• Date: 2005

Authors

Lebres, PJF (Author) Other; Gil Manuel Gonçalves (Author) FEUP; Worner, J (Author) Other

Conference proceedings International

Title: Modelling and Simulation 2005

Pages: 108-113

European Simulation and Modelling Conference (ESM 2005)

Oporto, PORTUGAL, OCT 24-26, 2005

Indexing

ISI Web of Knowledge - 0 Citations

Scopus - 0 Citations

Other information

Authenticus ID: P-000-69F

Abstract (EN): Nowadays, the execution of simulation projects in the field of production planning and logistics constitutes an integral part of the overall planning process, allowing for example the test of different system configurations and the choice of the most adequate model even before its implementation. The necessity of the manufacturing organisations to have internal systems that support their operations is bringing the definition of Digital Factory to reality. Today's available simulation software is based in static data input at start up, what can lead to the divergence of the real system and digital model results. Changes in the real system parameters are the major cause of that divergence, because the simulation stays with the initial set of values. This divergence does not allow a correct planning and results forecast, so it is necessary to create some kind of synchronisation between the digital model and the real system. Dynamic Data Driven Application Systems (DDDAS), a promising new concept that is starting to arise, consists in the ability of the simulation application to receive and respond online to physical data and measurements and/or control the measurements. This feedback keeps the digital model with the actualised real system parameters. In this work this concept is used to integrate data from a real production system into the digital model and to study the stability conditions of this feedback. A short discussion about the future benefits of using a dynamic data driven application system together with the concept of Grid Computing in an enterprise cooperative system will be also discussed. An assembly line where an automobile roof or sun-roof is attached to the car body is used as case Study. A virtual model of this system will be developed, using different tools (Petri Nets, State Flow, and object oriented simulation software) and the results will be compared. After the model's development, the study of the system performance using eM-Plant will be done and the system parameters optimised. The implementation of a real system feedback into the digital model, developed in the used simulation software, will be followed by a discussion of what improvements this method can bring in the future and how it can be implemented in practical real world scenarios.

Language: English

Type (Professor's evaluation): Scientific

No. of pages: 6