Residential buildings contribute significantly to energy use, health outcomes, and carbon emissions. In New Zealand, housing quality has historically been poor, with inadequate insulation and inefficient heating contributing to widespread energy hardship. Recent reforms, including the Warmer Kiwi Homes program, Healthy Homes Standards, and H1 Building Code upgrades, have delivered health and comfort improvements, yet challenges persist. Many retrofits remain partial, data on household performance are limited, and decision-making support for homeowners is fragmented. This study presents the design and evaluation of an AI-powered decision-support tool for residential energy efficiency in New Zealand. The prototype, developed using Python and Streamlit, integrates data ingestion, anomaly detection, baseline modeling, and scenario simulation (e.g., LED retrofits, insulation upgrades) into a modular dashboard. Fifteen domain experts, including building scientists, consultants, and policy practitioners, tested the tool through semi-structured interviews. Results show strong usability (M = 4.3), high value of scenario outputs (M = 4.5), and positive perceptions of its potential to complement subsidy programs and regulatory frameworks. The tool demonstrates how AI can translate national policies into personalized, household-level guidance, bridging the gap between funding, standards, and practical decision-making. Its significance lies in offering a replicable framework for reducing energy hardship, improving health outcomes, and supporting climate goals. Future development should focus on carbon metrics, tariff modeling, integration with national datasets, and longitudinal trials to assess real-world adoption.

翻译：住宅建筑对能源消耗、健康结果和碳排放具有显著影响。在新西兰，住房质量历来堪忧，保温不足和供暖效率低下导致广泛的能源困境。包括"更温暖的新西兰家庭"计划、健康住房标准及H1建筑规范升级在内的近期改革措施已带来健康与舒适度的改善，但挑战依然存在：改造往往不彻底，家庭用能数据有限，为房主提供的决策支持碎片化。本研究提出一款面向新西兰住宅能效的AI驱动决策支持工具的设计与评估。该原型采用Python和Streamlit开发，将数据采集、异常检测、基准建模和情景模拟（如LED改造、保温升级）集成于模块化仪表盘。15名领域专家（包括建筑科学家、顾问和政策实践者）通过半结构化访谈对工具进行测试。结果显示：可用性评分较高（M=4.3），情景输出价值显著（M=4.5），且专家普遍认为该工具有潜力辅助补贴计划与监管框架。该工具证明了AI如何将国家政策转化为个性化、家庭层面的指导，弥合资金、标准与实际决策之间的鸿沟。其重要意义在于提供了可复制的框架，以减少能源困难、改善健康结果并支持气候目标。未来开发应聚焦碳指标建模、电价模型研究、与国家级数据集整合，以及开展纵向试验评估实际应用成效。