A multi-agent system approach for performance monitoring of electronic channels in banks

Abstract

Modern banking is conducted within an environment of multiple electronic channels to meet the various needs of customers. These banking channels include Automated Teller Machines (ATMs), Point of Sale (POS) devices, Internet banking, and Mobile banking among others and banks are keen to monitor the performance of these channels. However, conventional techniques such as database queries, application monitoring, network links monitoring, data mining, and rule-based technologies used in monitoring the performance of electronic channels are mainly manual or semi-automated and lack an integrated intelligent model for this task. The literature that was reviewed did not show the application of multi-agent systems technology in monitoring the performance of banking electronic channels although this technology has been shown to be successful in several areas of systems performance monitoring. This study therefore set out to research on the use of a multi-agent system based approach to performance monitoring for banking electronic channels. The model used simulated data and it was evaluated based on key functionalities in terms of the ability to generate simulated transactions, monitoring channel activity and generate reports advising on e-channels performance and its realism. Four evaluators tested this model and confirmed it had the functionalities of a useful performance monitoring and reporting tool in banking and that it addressed key inadequacies current techniques face. The research found out that multi-agent system technology, through its use of intelligent and autonomous agents, could continuously monitor and report performance of the different e-channels with limited or no human intervention thus avoiding one of the main challenges conventional techniques face. This study is significant to the academic community interested in artificial intelligence or the specific area of using multi-agent systems in monitoring the performance of an electronic system. The findings will also prove useful to software developers and company procurement departments in identifying essential features of an intelligent performance monitoring and reporting solution. Keywords: Agent, Multi-Agent Technology, Multi-Agent System, Performance Monitoring, Electronic Channels, E-Channels, ECPM, JADE, POS, ATM, Artificial Intelligence, E-Banking