Intelligent agent for serious game of real time gross settlement using forest fire model

The goal of the intelligent agent in this research is fulfill the settlement process in Real Time Gross Settlement (RTGS). The RTGS system is a payment settlement process for individual processing of payments in an electronic system for gross settlement in real time. Settlement process in this research is managed by clearing house. The mechanism at clearing house is a transmitter client sends a message of transaction through transmitter bank that having canal at clearing house then continues to receiver client through receiver bank by using forest fire model. This process is influenced by behavior of four agents (saving agent, reserves agent, loan agent and deposit agent). The result of this research indicates that the highest fluctuation of the agents that influence to increase the Net worth (NW) (4% to 96%) and fluctuation of the agents that influence to decrease the NW (2% to 94%) take place in the smallest probability (p = 0.1). While the most limited fluctuation of the agents that influence to increase the NW (35% to 60%) and fluctuation of the agents that influence to decrease the NW (40% to 60%) take place in the highest probability (p = 0.9). In the serious game of RTGS, this probability is used to determine optimal transaction and influence to stability in the RTGS system.