Multi-objective near-optimal necessary conditions for multi-sectoral planning.

In the energy transition context, restructuring energy systems and making informed decisions on the optimal energy mix and technologies is crucial. Energy system optimisation models (ESOMs) are commonly used for this purpose. However, their focus on cost minimisation limits their usefulness in addressing other factors like environmental sustainability and social equity. Moreover, by searching for only one global optimum, they overlook diverse alternative solutions. This paper aims to overcome these limitations by exploring near-optimal spaces in multi-objective optimisation problems, providing valuable insights for decision-makers. The authors extend the concepts of epsilon-optimality and necessary conditions to multi-objective problems. They apply this methodology to a case study of the Belgian energy transition in 2035 while considering both cost and energy invested as objectives. The results reveal opportunities to reduce dependence on endogenous resources while requiring substantial reliance on exogenous resources. They demonstrate the versatility of potential exogenous resources and provide insights into objective trade-offs. This paper represents a pioneering application of the proposed methodology to a real-world problem, highlighting the added value of near-optimal solutions in multi-objective optimisation. Future work could address limitations, such as approximating the epsilon-optimal space, investigating parametric uncertainty, and extending the approach to other case studies and objectives, enhancing its applicability in energy system planning and decision-making. [Display omitted] • We present an approach to exploring near-optimal spaces in multi-objective problems. • Using necessary conditions, better insights can be provided to decision-makers. • The methodology is applied to a case study of the Belgian energy transition. • We reveal minimum resource needs to achieve a reasonable cost and EROI. [ABSTRACT FROM AUTHOR]