Development of a Configurable IoT-Based Adaptive Thermal Comfort Monitoring Platform Using Scrum and DevOps Practices

Abstract

Effective indoor thermal comfort management in multi-room buildings requires the integration of environmental sensor data and occupant feedback while remaining adaptable to different building configurations. However, many existing occupant-centric thermal comfort systems are tightly coupled to specific building environments, which limits their adaptability to new buildings and different room configurations. To address this limitation, this study develops and evaluates an IoT-based adaptive thermal comfort monitoring platform designed for configurable multi-building deployment. The platform integrates environmental sensor data with occupant feedback collected through room-specific QR code-accessible web forms and supports two complementary recommendation mechanisms: sensor-driven threshold alerts and feedback-driven recommendations based on aggregated occupant responses. The platform was developed using the Scrum methodology over seven one-week sprint cycles, supported by a DevOps CI/CD pipeline that automated application building, testing, artifact management, and deployment. Evaluation included unit testing, integration testing, Selenium-based user interface testing, and Apache JMeter performance testing, with all automated test scenarios passing successfully. Performance testing yielded APDEX scores of 0.997 and 0.960 for the General User and Administrator workflows, respectively, with zero errors across all tested endpoints. Flexible building and room expansion was achieved through an entity-based architecture that models Building, Room, and Sensor as independent entities, enabling the incorporation of new spaces without structural modifications. These results demonstrate that the proposed platform provides a configurable and extensible software solution for adaptive thermal comfort monitoring in multi-room building environments.