Journal of Engineering, Project, and Production Management, 2026, 16(1), 2025-155
Improved Multi-Objective Particle Swarm Optimization for Sustainable Building Design
Deputy Director, Infrastructure Construction Department, Jiangsu Maritime Institute, Nanjing, 211170, China, E-mail: dnfdusong@outlook.com
Project Management
Received August 10, 2025; revised September 25, 2025; accepted September 26, 2025
Available online December 24, 2025
Abstract: Traditional multi-objective optimization methods suffer from limitations such as sluggish convergence and high scenario sensitivity when it comes to multi-objective architectural design decision optimization for sustainable buildings. This study constructs a multi-objective architectural design decision optimization model that simultaneously considers comfort level and building energy consumption. The model is solved using a modified backbone multi-objective particle swarm optimization algorithm that optimizes the search strategy and introduces an adaptive perturbation mechanism. Results show the improved algorithm achieved higher hypervolume values than the rest of the mainstream algorithms in both building scenarios. Performance remained stable with minimal fluctuation between scenarios. In the single-room office scenario, the average hypervolume value of the improved algorithm was as high as 29,963. This substantially exceeded the non-dominated sorting genetic algorithm II (NSGA-II), which achieved 19,246. For the three residential scenarios, the improved algorithm reached an average hypervolume of 42,639, compared to 14,628 for standard multi-objective particle swarm optimization. Across scenarios, the improved algorithm’s average run times (1.38h and 3.38h, respectively) were lower than all other algorithms. In addition, the algorithm’s Pareto frontier solutions were concentrated in the low-energy, high-comfort region. In conclusion, the improved algorithm effectively achieves dual-objective decision optimization, balancing user comfort with building energy efficiency. Novelty lies in integrating a backbone guidance mechanism with an adaptive perturbation strategy. This addresses parameter redundancy and premature convergence in traditional multi-objective particle swarm optimization. The approach practically enables rapid generation of energy-efficient designs while maintaining high comfort levels. This provides architects with quantitative support for sustainable design across various building types, including offices and residences.
Keywords: Sustainable building, PSO, multi-objective, decision optimization, adaptive perturbation.
Copyright © Journal of Engineering, Project, and Production Management (EPPM-Journal).
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.
Requests for reprints and permissions at eppm.journal@gmail.com.
Citation: Du, S. (2026). Improved Multi-Objective Particle Swarm Optimization for Sustainable Building Design. Journal of Engineering, Project, and Production Management, 16(1), 2025-155.
DOI:
10.32738/JEPPM-2025-155
Full Text
