Agent-Directed Simulation and Systems Engineering
Editors:
Dr. Levent Yilmaz, Assistant Professor, M&SNet: Auburn M&S Laboratory, Computer Science and Software Engineering Auburn University, Auburn, AL, USA yilmaz@auburn.edu
Dr. Tuncer I. Ören, M&SNet of SCS, Professor Emeritus, University of Ottawa, Ottawa, ON, Canada oren@site.uottawa.ca

System engineering (SE) involves the application of engineering and management practices to transform the user needs into a system specification and realization that most efficiently meets the need. SE entails the technical management functions that controls and coordinates the overall system development activities. Agent paradigm and its related theory and methodologies, on the other hand, opened new frontiers for advancing the physical, natural, social, military, and information sciences and engineering. This book will cover (1) the application of agent-directed simulation concepts and technologies to systems engineering and (2) the use of novel systems engineering principles in developing large complex agent-directed simulations.

The contributions are sought on the synergy of agent-directed simulation and systems engineering.

Agent-directed Simulation for Systems Engineering 

Agent simulation involves the simulation of agent systems. Agent systems possess high-level interaction mechanisms independent of the problem being solved. Communication protocols and mechanism for interaction via task allocation, coordination of actions, and conflict resolution at varying levels of sophistication are primary elements of agent simulations. Simulating agent systems requires understanding the basic principles, organizational mechanisms, and technologies underlying such systems. The principle aspects underlying such systems include the issues of action, cognitive aspects in decision making, interaction, and adaptation. Organizational mechanisms for agent systems include means for interaction. That is, communication, collaboration, and coordination of tasks within an agent system require flexible protocols to facilitate realization of cooperative or competitive behavior in agent societies.

Agent-based simulation is the use of agent technology to generate model behavior or to monitor generation of model behavior. This is similar to the use of AI techniques for the generation of model behavior; e.g., qualitative simulation and knowledge-based simulation. The perception feature of agents makes them pertinent for monitoring tasks. Agent-based simulation is useful for having complex experiments and deliberative knowledge processing such as planning, deciding, and reasoning.

Agent-supported simulation deals with the use of agents as a support facility to enable computer assistance by enhancing cognitive capabilities in problem specification and solving. Hence, agent-supported simulation involves the use of intelligent agents to improve simulation and gaming infrastructures or environments. Agent-supported simulation is used for three purposes:

(1)   to provide computer assistance for front-end and/or back-end interface functions;

(2)   to process elements of a simulation study symbolically (for example, for consistency checks and built-in reliability); and

(3)   to provide cognitive abilities to the elements of a simulation study, such as learning or understanding abilities.

Systems Engineering for Agent-directed Simulation

Simulation is becoming a dominant technology in many systems engineering applications. In defense-related training systems, simulations are being embedded to create virtual scenarios. Symbiotic simulation systems have been proposed as a way of solving this problem by having the simulation and the physical system interact in a mutually beneficial manner.

Modeling & Simulation (M&S) development costs are rising partly due to increased complexity. Craftsmanship approach to M&S in the small does not scale to M&S in the large. Consequently, such complex and extremely large simulation systems require technical system management and SE oversight. Unless such oversight is present, the following problems are likely to emerge.

·         Simulation system becomes unmanageable.

·         Costs are overrun and deadlines can be missed.

·         Greater risk exposure arises.

·         Requirements may not be met.

·         The simulation fails to satisfy its objectives.

·         Maintenance costs increase.

Agent systems are goal-directed and adaptive: In well-adapted systems, goals and constraints are often implicit and embedded in the process. In a dynamically changing environment, an effective realization depends on self-organizing and adaptive mechanisms that are in place to change properties of the process to meet the current needs.

Agent system processes are frequently modified to update the structure and mechanisms to keep some measure elated to the relevant performance objective near an optimum. Control of adaptation, however, is distributed across all components and subsystems. A useful and credible model for the analysis of the process and prediction of responses to changes in the circumstances must reflect the mechanisms underlying the evolution of the process.

Many agent systems (e.g., sociotechnical systems) are human-centered:  If our critical infrastructures are to continue to provide vital services safely and reliably, the linkages between people, organizations, and technology need to be fully understood and managed holistically. Human actors that manage and processes are adaptive and goal-directed agents. What people actually do, how they communicate and collaborate, how they solve problems, resolve conflicts, and learn behavior matters in the outcome of a process. Hence, representing activities requires modeling communication, collaboration, team work, conflict resolution, and tool and technology usage.

Given the above observations, a systems engineering perspective and methodology for agent systems should represent not only technical activities, policies, and procedures, but also the resources, preferences, and cognition of staff members, together with functional and social organization and strategic management, all in unified and coherent terms.

To assume a consolidated approach, we would like to proceed in two phases. During the first phase, the background chapters will be written; and appropriate chapters will be made available to the authors of the second phase.

 

Key Dates: