Multi-Objective Optimization for Selecting and Scheduling Observations

This paper presents a biased random-key genetic algorithm for solving a multi-objective optimization problem concerning the management of agile Earth observing satellites. It addresses the selection and scheduling of a subset of photographs from a set of candidates in order to optimize two objectives: maximizing the total profit, and ensuring fairness among users by minimizing the maximum profit difference between users. Two methods, one based on dominance, the other based on indicator, are compared to select the preferred solutions. The methods are evaluated on realistic instances derived from the 2003 ROADEF challenge.