USE OF AGENTS TO CONTROL THE EXECUTION OF SIMULATION COMPONENTS

To meet the urgent demand in Web-based simulation for high-level management of simulation analysis, two prospective and compelling technologies, components and agents, have been introduced into JSIM. Agents, in JSIM, not only free simulationists from the details of statistical analysis, but also provide high quality simulation output data. A scenario agent unifies the important properties from different components in a graphical user interface, and assists the user throughout the simulation process. A new design, in simulation model components, completely separates the component's communication portion from the portion, which represents a real physical system. Now it is possible to create the simplified model components by code generation. It also supports the extension of the communication portion without any effects on the simulation content portion. A virtual scheduler, supporting virtual time simulation in JSIM, provides another scheduling alternative for the simulation. The use of component and agent technology in the JSIM Web-based simulation environment allows simulation models to be treated as components that can be dynamically assembled to build model federations. It also allows simulation inputs and outputs to be dynamically linked to database systems, making storage of simulation results easy and flexible.

INDEX WORDS: output data analysis, agent-based simulation, JavaBeans, JSIM, component-based simulation, web-based simulation, scenario agent, Java, JDBC, ModelBean, model agent

• CHAPTER 1 INTRODUCTION
• CHAPTER 2 TECHNOLOGICAL INFRASTRUCTURE
• CHAPTER 3 SIMULATION OUTPUT ANALYSIS
• CHAPTER 4 OVERVIEW OF JSIM
• CHAPTER 5 MODEL BEANS
• CHAPTER 6 RUNAGENCY
• CHAPTER 7 DATABASE BEANS
• CHAPTER 8 AN EXAMPLE: MODEL FEDERATION
• CHAPTER 9 CONCLUSION AND FUTURE WORK
• REFERENCE
• APPENDICES