Automated television scheduling via evolving agents

Presents a novel method of incorporating agent methodology, genetic algorithms and game theory to analyze television scheduling. We designed the three major United States networks as autonomous agents that compete for viewership of a population of household agents. In a competitive environment, these network agents attempt to maximize their ratings by evaluating other network agents' behaviors and schedules. They then evolve their schedule via a genetic algorithm to better compete for viewers. Viewer agents choose activities that maximize their satisfaction. The richness of this simulation provides insight into the seemingly impenetrable dynamic environment of television scheduling.