Multi-Objective Optimization and Decision-Making in Building Retrofits: A Systematic Review of Frameworks, Trends, and Research Gaps

Engineering notes

Engineering notes will be added by the Power for Solar editorial team.

Chinese explanation / 中文解读

中文解读待补充：本站会优先为光伏效率、钙钛矿太阳能电池、储能技术、太阳能热利用、BIPV、并网技术等高价值论文补充中文说明。

Original abstract

In this study, a systematic literature review is presented on optimization-based building retrofits, with an emphasis on how recent studies structure and use decision-support frameworks. The reviewed studies, published between 2020 and 2026, were classified according to the following criteria: retrofit strategies, objective functions, optimization algorithms, computational tools, climatic contexts, and the integration of Sensitivity Analysis (SA), Machine Learning (ML), Multi-Objective Optimization (MOO), and Multi-Criteria Decision-Making (MCDM). The findings show that envelope-related measures, including insulation, window, airtightness, and shading upgrades, remain the most frequently optimized strategies, followed by HVAC improvements and renewable energy integration. The dominant GA-based algorithm is NSGA-II due to its robustness in handling nonlinear, simulation-coupled, and conflicting objectives. However, the primary contribution of this review lies in identifying the fragmented but emerging integration of optimization, SA, ML, and decision-making frameworks. Recent studies have increasingly combined SA for parameter prioritization, ML for surrogate prediction and computational acceleration, MOO for Pareto-front generation, and MCDM for final solution ranking. Despite this progress, integration remains limited by insufficient surrogate-model validation, inconsistent reporting of computational costs, and inadequate assessment of uncertainty effects on Pareto-front accuracy and decision rankings. Economic and life-cycle carbon assessments also remain simplified through the use of static energy prices, fixed emission factors, and inconsistent LCA boundaries. Climatic analysis indicates a dominant focus on cold and multiple or mixed climatic contexts, whereas future climate scenarios remain insufficiently incorporated into the retrofit planning process. The most important gaps in optimization and decision-making frameworks for building retrofits include dynamic carbon and cost modeling, embodied carbon assessment, uncertainty quantification, future climate resilience, and stakeholder-informed decision-making. This review highlights the need for transparent, data-driven, and integrated assessment frameworks to support both resilient and implementable retrofit decisions.

Links and sources

• PDF from original source

Comments