Optimization

Figure 4: Optimization

Game Strategy

• Pareto Game: uses the classical hierarchical multi-population Pareto optimality.
• Hybrid Game (Pareto & Nash): uses a combination of Pareto and Nash game strategies. It consists in one Pareto Player and many Nash Players and can produce a Nash-equilibrium and Pareto non-dominated solutions simultaneously

Optimization Method

• Genetic Algorithm (GA): uses a modified version of the Non-domitated Sorting Genetic Algorithm II (NSGA-II)
• Particle Swarm Optimization (PSO): this algorithm treats each individual in the population as a particle, then it moves each particle in the search space to a given position and with a certain velocity (thereby the name of swarm). The swarm moves to local minima and search for other minima if there are better ones.

Choice of Optimization Progress

• Live progress: does the same as Update to file and additionally it shows a pop-up window with a live progress of the optimization.
• Update to file: logs all the outputs from the analyzer and RMOP to a file named HPRMOP.log located in the project directory.