A qualitative analysis of human-agent functions for collaborative multi-agent system

Abstract

There is no one method for agent modeling due to the uniqueness and complexity of application domains. Consequently, agent modeling faces several difficulties in development and implementation and has been continuously researched on the conceptualization of complex systems which are composed of several agents. Consequently, we introduced the concept of node (or the nodal approach) as a novel agent-based model that consists of a human, one or more agents and their corresponding functions that represent an intelligent entity. The instantiation of multiple nodes with diverse functions and workflow process could represent a domain in which humans interact with other humans via software agents to achieve some common goals. In this paper, we conduct a questionnaire survey to investigate on the generalized functions of humans and agents that are required for a node. A number of devices and media including Document, Email, Phone Call, SMS are selected to validate the proposed model. This paper presents the results of the survey that solicits information pertaining to the tasks performed by humans to determine generalized functions of humans and agents in a node. © 2014 IEEE.