19th EMSS, 2007

19th European Modeling and Simulation Symposium

(Simulation in Industry)

4-6 October 2007

Bergeggi (Sv), Italy

Session CO-I: High Level, distribued & Continous Simulation

 

HIERARCHICAL INTELLIGENT HIGH-LEVEL SIMULATION, Tudor Niculiu, Anton Manolescu, Sorin Cotofana

Intelligence = (consciousness, adaptability, intention) and Faith = (inspiration, intuition, imagination) are linked by Conscience = (consciousness, inspiration).

 

Faith and intelligence form the of our life

(Way, Truth, Life).

            

To simulate the conscience we need transcending from    discrete  computability to continuous simulability. A way to begin is hierarchical simulation. Applying Divide et Impera   et Intellige to hierarchy types reveals their importance for intelligent simulation. A comprehensive construction is        enabled by object-oriented representation, symbolic meaning, and modular   approach. We create a theoretical kernel for self-organizing systems based  on formal hierarchical description. Different forms of mixed simulation work    together to make the simulated system aware of its knowledge and its activity.

 

 

ANALYZING SIMULTANEOUS EVENTS IN DISTRIBUTED SIMULATION, Patrick Peschlow, Peter Martini

 

Simultaneous events are a well-known challenge in discrete-event simulation: Different execution orders may lead to different simulation results. Commonly, simultaneous events are handled by tie-breaking rules using event priorities. A second approach, however, is analyzing the effects of simultaneous events with a branching mechanism. By examining different event orderings, and computing

multiple simulation branches, confidence in simulation results may be increased. In distributed discrete-event simulation, the realization of a branching mechanism is a complex task: Simultaneous events may be spread among different logical processes, there is no global view on event lists and the system state, and possible zero-delay events sent between logical processes have to be respected. In this paper, we address these challenges and present a branching mechanism for analyzing simultaneous events in distributed simulation.

 

 

ADAPTING AN EJECUTABLE ENTERPRISE MODEL ENVIRONMENT FOR HLA, Gutiérrez, Ma. De los Milagros, Leone Horacio P.

 

Executable Enterprise Models (E2M) are one of the most important sources of an organization’s information requirements where the business expert has not only an appropriate representation of the organization in terms of processes, information flows, user roles, but also a simulation capability for the interpretation of the dynamic behavior. In this context, we present an environment that allows building and analyzing E2M, made of Conceptual model and Simulation model. The distributed characteristic is needed for build in a modular way this model, as well as to participate as part of a supply chain simulation model. We know the contribution to employing HLA to support interoperability and reuse of existing simulations into the different member of the supply chain. In this paper we present the changes made in an executable enterprise model environment that uses Devs formalism in the execution of its models. The new simulations support interoperate with other simulators through HLA standard.

 

 

SIMULATION OF TELECOMMUNICATION FLOWS USING FLUID DYNAMIC MODELS, Alfredo Cutolo, Raffaella Frattaruolo, Rosanna Manzo

 

A software application, that simulates the information flow in a telecommunication network, is proposed. The simulator is based on a continuous model obtained imposing the conservation of the mean number of packets among two consecutive nodes. For dynamics at nodes, we consider two “routing algorithms”, RA1 and RA2. The main difference among the two algorithms is the possibility of redirecting packets by the RA2 algorithm. We  show the different behavior of the information flow in a simple network, using these two different algorithms.