Low-carbon decision-making for decoration schemes under preference uncertainty: a robust method integrating LCA and Monte Carlo simulation

Interior decoration is a major source of embodied carbon emissions, highlighting the need for reliable low-carbon decision-making tools. However, the influence of preference uncertainty on ranking stability is seldom addressed in existing low-carbon Multi-Criteria Decision-Making (MCDM) studies. To overcome this limitation, this study proposes an integrated probabilistic decision-making framework that combines cradle-to-site Life Cycle Assessment (LCA), TOPSIS-based multi-criteria evaluation, and Dirichlet-driven Monte Carlo simulation to explicitly quantify weight uncertainty and ranking robustness. The study evaluated a real university Academic Exchange Center interior decoration project (GFA ≈ 4,200 m²) in Chongqing with three alternative schemes: high-end, medium-end, and basic. Results show that material production accounts for over 90% of cradle-to-site embodied carbon emissions across all schemes, indicating material substitution as the most effective emission reduction pathway. Although deterministic TOPSIS initially identified the medium-end scheme as optimal, 1600 Monte Carlo iterations revealed that the high-end scheme achieves the highest median closeness degree (0.5372) and the greatest probability of ranking first (54.94%), demonstrating it as the most robust optimal solution under preference uncertainty. The proposed framework advances low-carbon design methodology by shifting from deterministic ranking to probability- and robustness-based decision analysis, providing more compelling guidance for multi-stakeholder, preference-sensitive interior design practice.