An introduction to multi-objective optimization

Multi-objective optimization (MOO) deals with problems involving multiple conflicting goals, where improving one objective may negatively affect another. Instead of identifying a single optimal solution, MOO aims to generate a diverse set of Pareto-optimal solutions and/or select one or some that aligns with the decision maker’s subjective preferences. This talk introduces the fundamental ideas of MOO, including key concepts and widely used approaches such as a priori and a posteriori methods, while addressing challenges like non-convexity and computational complexity. We also discuss its practical relevance in the energy domain, where MOO could support power system optimization, renewable energy integration, supply-demand balancing, emission reduction, and energy-efficient structures.