Utility-driven simulation modelling and multi-objective evolutionary optimisation for BIM-based construction emission reduction

This research tackles the dynamic optimisation of construction carbon emissions by proposing a novel integration of building information modelling, utility-driven simulation, and multi-objective evolutionary computation.A core methodological contribution is a formalised utility function that quantifies and embeds project-specific decision-maker preferences for time, cost, and emissions directly into the optimisation search process.This function guides a bespoke evolutionary algorithm to automatically generate efficient, low-carbon construction plans, which are accurately evaluated by a high-fidelity discrete-event simulation engine using enriched building information modelling data.In an experimental study based on a publicly available office building dataset, our framework demonstrates superior performance, outperforming state-of-the-art benchmarks with a 5.6%-15.3%gain in hypervolume and achieving an 18.7% reduction in on-site emissions.This work provides a rigorous and actionable decision-support system for advancing sustainable construction practices.