A simulation-based multi-objective optimization framework for the production planning in energy supply chains

The work presents a simulation-based Multi-Objective Optimization (MOO) framework for efficient production planning in Energy Supply Chains (ESCs). An Agent-based Model (ABM) that is more comprehensive than others adopted in the literature is developed to simulate the agent’s uncertain behaviors and the transaction processes stochastically occurring in dynamically changing ESC structures. These are important realistic characteristics that are rarely considered. The simulation is embedded into a Non-dominated Sorting Genetic Algorithm (NSGA-II)-based optimization scheme to identify the Pareto solutions for which the ESC total profit is maximized and the disequilibrium among its agent’s profits is minimized, while uncertainty is accounted for by Monte Carlo (MC) sampling. An oil and gas ESC model with five layers is considered to show the proposed framework and its capability of enabling efficient management of the ESC sustained production while considering the agent’s uncertain interactions and the dynamically changing structure.