Multi-objective optimization

So far in this chapter, we have taken the examples of the problem with one objective (finding a food source for an ant colony). However, in real-world scenarios, often there is more than one objective that needs to be met by the individual agents as well as the swarm. For example, in the case of honey bees they need to look for the food source, gather the food, and find a safe and viable place for the beehive. One objective is fulfilled at the cost of another objective. The agent should be programmed to consider the trade-off in the larger interest of the swarm.

As far as possible, the optimization function for the agent should bring optimum solution for more than one objective, but it is not feasible to mutual exclusivity. In such cases, the agent should be able to operate without a central control and decide the objective weightage based on the environmental context and should favor the objective that will fulfill the swarm's overall objective for an elongated...