Exploring energy-related occupant behavior in office buildings: an interdisciplinary study in the context of building physics and social psychology

Abstract

This thesis presents a novel approach to occupant behavior (OB) modeling in office buildings by integrating psycho-social behavioral clusters with environmental parameters through fuzzy logic modeling. Traditional building performance simulations often rely on static and generalized assumptions, which fail to capture the dynamic and complex nature of human behavior in energy use, contributing to the energy performance gap. This research addresses this challenge through a three-step methodology: (1) online study, (2) field study, and (3) fuzzy logic modeling. Each step informs the next, culminating in a holistic framework for more accurate energy performance predictions. In Step 01, three distinct behavioral clusters were developed using the Theory of Planned Behavior, Self-Determination Theory, and habit and comfort constructs, capturing different energy-saving tendencies, motivation, and comfort needs. Step 02 mapped these clusters against environmental parameters—outdoor and indoor temperature, humidity, and CO₂ levels—to create a dynamic, context-responsive model of occupant actions. Finally, Step 03 implemented a fuzzy logic system in MATLAB, designed to integrate with EnergyPlus™ via its Energy Management System (EMS). This integration enables real-time simulation of occupant responses to environmental shifts, incorporating psycho-social influences and seasonal variations, thereby improving the precision and adaptability of energy consumption forecasts.