Towards Climate-Responsive Office Architecture in NCR India: A Multi-Objective Optimization Study of Cooling Load, Energy Use Intensity, and Daylight Performance

This study presents a coupled building simulation framework that evaluates thermal and daylight performance concurrently within a unified multi-objective decision space. Unlike conventional sequential workflows, where daylight metrics are assessed after energy optimization or used primarily for compliance verification, the proposed approach embeds EnergyPlus and Radiance simulations directly within the same optimization loop. This structure enables a systematic exploration of non-linear interactions between Energy Use Intensity (EUI), cooling loads, Spatial Daylight Autonomy (SDA), and Annual Sunlight Exposure (ASE) during early-stage façade design. The framework is demonstrated through a medium-rise office building in India’s National Capital Region, a composite climate characterized by strong seasonal and directional variability. Parametric variation in façade orientation, window-to-wall ratio, and external shading configurations was explored using a multi-objective genetic algorithm to identify Pareto-optimal performance regimes. The results reveal distinct orientation-dependent trade-off structures between solar exposure, cooling demand, and daylight availability that are not evident in rule-based or sequential simulation approaches. In particular, a transitional East-facing façade regime emerges in which balanced shading and glazing proportions achieve near–North-facing cooling performance while maintaining high daylight autonomy under controlled sunlight exposure. Rather than proposing a single optimal solution, the study demonstrates how tightly coupled thermal–daylight simulation can function as a knowledge-discovery tool, enabling the extraction of transferable façade response patterns from simulation outputs. The findings highlight the limitations of prescriptive orientation hierarchies in composite climates and illustrate the value of integrated simulation workflows for performance-driven early-stage design across diverse climatic contexts. Although the study references thermal performance, the optimization objectives are limited to peak cooling load and annual Energy Use Intensity (EUI). Occupant comfort indices such as Predicted Mean Vote (PMV) and Predicted Percentage Dissatisfied (PPD) were not explicitly simulated. Therefore, results are interpreted as energy–daylight performance optimization rather than direct thermal comfort optimization.