Dealing with unforeseeable changing situations, often seen in exploratory and hazardous task domains, requires engineered complex systems that can feature flexibility to fulfill multiple resolutions over long lifespans, robustness to deal with environmental changes and resilience to sustain system damage. The current top-down engineering design approach has its limitations in cases where it is impossible to fully consider and predict the true operational uncertainties because the hidden interdependencies among the system components can lead to unforeseeable interactions at operation time. The challenge for engineering design researchers and practitioners is how to devise new ways to design such adaptive systems. Nature embodies certain qualities such as evolution, cellular organization, and self-organizing behavior, which seem to overcome the deficiencies of the traditional top-down engineering design process. Taking advantage of the flexibility of multi-agent systems, we proposed a self-organizing systems approach, in which mechanical cells or agents organize themselves as the environment and tasks change based on a set of predefined rules. This study is positioned at the interface between the science and engineering of complex systems by taking the “by emergence” approach to achieve desired functions of cellular self-organizing systems.
Through the case studies we investigated the impact of social rule based social structuring, measured by social rules adoption rate and the size of agent population, on the system performance in the face of increasing task complexity. The results have shed interesting insights including: the behavior of self-organizing systems becomes more chaotic when tasks are more complex; stronger social structuring is effective for a smaller number of agents and weaker social structuring is more effective for a larger number of agents; and there can be a “singular” number of agents where social structuring is neither effective nor efficient. In conclusion, self-organization has profound implications in dealing with task complexity and can be used intentionally as a tool in the design of adaptive complex systems. The balance of task complexity, the number of agents, and social structuring is the key. Understanding self-organization, minimizing harmful effects, and promoting positive effects will become essential in future engineering design of complex systems.
Through the case studies we investigated the impact of social rule based social structuring, measured by social rules adoption rate and the size of agent population, on the system performance in the face of increasing task complexity. The results have shed interesting insights including: the behavior of self-organizing systems becomes more chaotic when tasks are more complex; stronger social structuring is effective for a smaller number of agents and weaker social structuring is more effective for a larger number of agents; and there can be a “singular” number of agents where social structuring is neither effective nor efficient. In conclusion, self-organization has profound implications in dealing with task complexity and can be used intentionally as a tool in the design of adaptive complex systems. The balance of task complexity, the number of agents, and social structuring is the key. Understanding self-organization, minimizing harmful effects, and promoting positive effects will become essential in future engineering design of complex systems.
For the full paper see ASME's Digital Collection.
RSS Feed