Introduction

In complex decision-making scenarios, it’s often necessary to consider multiple conflicting objectives simultaneously. This challenge is addressed through multi-objective optimization, a powerful technique aimed at finding solutions that represent a trade-off between competing objectives. In this section, we delve into the implementation of a genetic algorithm for multi-criteria optimization using the nsga2() function from the mco package. Our goal is to define an objective function, specify input variables and output measures, and establish lower and upper bounds for the optimization problem. By leveraging this approach, we can explore the Pareto-optimal front and identify solutions that strike a balance between conflicting objectives. Through this exploration, we aim to uncover optimal strategies that maximize performance across multiple criteria, thereby facilitating informed decision-making and driving superior outcomes in complex decision landscapes.

 Result

In our final analysis, we examine the influence of initial value settings on project operational outcomes by aggregating various criteria, including duration, energy consumption, emissions, and cost. In the visualization provided, the red lines correspond to solutions located on the Pareto front, which denotes the set of optimal solutions. These solutions represent the best trade-offs between conflicting objectives, where improving one objective comes at the expense of another. Conversely, the blue lines represent solutions that are non-optimal, meaning they do not lie on the Pareto front and thus offer suboptimal trade-offs between objectives.

Pareto Optimal: Optimal & Non optimal solutions