MULTI-HIERARCHICAL AGENT BASED MODELLING SIMULATION FRAMEWORK

• Type: Article in International Conference Proceedings Book
• Date: 2011

Authors

Carlos Bragança de Oliveira (Author) FEUP; António Carvalho Brito (Author) FEUP

Conference proceedings International

Title: EUROPEAN SIMULATION AND MODELLING CONFERENCE 2011

Pages: 33-37

25th European Simulation and Modelling Conference (ESM 2011)

Guimaraes, PORTUGAL, OCT 24-26, 2011

Indexing

ISI Web of Knowledge - 0 Citations

Scopus - 3 Citations

Other information

Authenticus ID: P-009-AEF

Abstract (EN): The performance of a company depends on its strategic decisions but also on the efficiency of how those decisions are implemented by all of its sectors. In any company there are different sectors or groups, probably physically distributed, with different levels of autonomy, each one controlling local resources and interacting with the others. Many companies use simulation software to analyze critical sectors based in a number of scenarios, in order to estimate the efficiency of the different configurations. The process of simulate alternative solutions can be time consuming. Depending on the software used it may be necessary to generate new simulation models for each solution, develop new applications or change existing ones. This paper describes a simulation framework where each element is modelled using the discrete-event approach and interoperability between elements is achieved by the combination of web services and Agent-based modelling and simulation (ABMS). The use of ABMS in this framework provides an easier way to evaluate alternative solutions. Each sector or group is an agent that can be added or removed from the model to build different configurations. A multi-hierarchical simulation executive is proposed in order to enable the use of different time intervals in each level. This increases the autonomy of the simulation agents and promotes the use of parallel or distributed solutions.

Language: English

Type (Professor's evaluation): Scientific

No. of pages: 5