This paper explores the ability of a team of autonomous software agents to be effective in unknown and changing optimization environments by evolving to use the most successful algorithms at the points in the optimization process where they will be the most effective. We present the core framework and methodology which has potential applications in layout, scheduling, manufacturing, and other engineering design areas. The communal agent team organizational structure employed allows cooperation of agents through the products of their work and creates an ever changing set of individual solutions for the agents to work on. In addition, the organizational structure allows the framework to be adaptive to changes in the design space that occur during the optimization process — making our approach extremely flexible to the kinds of dynamic environments encountered in engineering design problems. An evolutionary approach is used, but evolution occurs at the strategic, rather than solution level — where the strategies of agents in the team (the decisions for picking, altering, and inserting a solution) evolve over time. As an application of this approach, individual solutions are tours in the familiar combinatorial optimization problem of the traveling salesman. With a constantly changing set of these tours, the team, each agent running a different solution strategy, must evolve to apply the solution strategies which are most useful given the set at any point in the process. As a team, the evolutionary agents produce better solutions than any individual algorithm. We discuss the extensions to our preliminary work that will make our framework highly useful to the design and optimization community.

This content is only available via PDF.
You do not currently have access to this content.