A Bio-Inspired Approach to Sustainable Building Design Optimization: Multi-Objective Flow Direction Algorithm with...

Area of Science:

• Building Science and Engineering
• Optimization Algorithms
• Sustainable Design

Background:

• Sustainable building design requires advanced optimization for economic and environmental goals, especially with mixed discrete-continuous variables.
• Existing metaheuristics have limitations in efficiently handling discrete choices in multi-objective building envelope optimization.

Purpose of the Study:

• To develop and apply a novel bio-inspired optimization approach for external wall insulation.
• To simultaneously minimize lifecycle costs and CO2 emissions for buildings across diverse climatic regions.
• To provide a climate-aware decision tool for sustainable building design.

Main Methods:

• Integration of the Multi-Purpose Flow Direction Algorithm (MOFDA) with One-Hot Encoding for mixed variable optimization.
• Application of the Complex Proportional Assessment (COPRAS) method, weighted by Shannon Entropy, to select Pareto-optimal solutions.
• Analysis of insulation performance across six updated climatic regions in Türkiye.

Main Results:

• Climate-dependent variations in optimal insulation thickness and material (e.g., 3.3 cm Rock Wool for cost in warm regions, up to 40 cm for emissions in cold regions).
• Shannon Entropy assigned high weights (88-92%) to CO2 emissions, indicating their significant impact on decision-making.
• Shift in preferred heating fuels from LPG in milder climates to Fuel Oil in harsher climates.

Conclusions:

• The MOFDA framework with One-Hot Encoding effectively handles mixed discrete-continuous optimization for sustainable building insulation.
• The bio-inspired approach provides a robust, climate-aware decision tool, demonstrating the translation of natural flow processes into engineering solutions.
• Environmental considerations, particularly CO2 emissions, play a dominant role in optimizing sustainable building envelopes across different climates.