Building Energy Management Systems (BEMS) generate significant operational data, but this data is often underutilized for basic monitoring and does not lead to meaningful improvements in operational efficiency. This passive use of data results in unnecessary energy consumption and reduced occupant comfort, with overcooling being a common HVAC inefficiency. This study introduces a three-phase analytical framework that utilizes BEMS data to systematically diagnose overcooling issues and provide actionable operational recommendations. The framework consists of: (1) Data Preparation, (2) Problem Diagnosis, and (3) Operation Guidance. The main contribution is the causal identification of temporal patterns, spatial distribution, and root causes of overcooling through statistical analysis and data mining techniques. These insights are translated into conditional execution rules that enable non-expert operators to implement solutions. An application of this framework to summer cooling data from a Zero Energy Building revealed that overcooling occurred during 70.7% of cooling operation hours. Root cause analysis indicated that overcooling is a systematic issue, with hourly patterns correlated to external environmental conditions. This research offers managers a practical methodology to improve energy efficiency and occupant comfort by transforming passive BEMS data into actionable control strategies.
Oh et al. (Fri,) studied this question.