Short-Term Environmental Trend Analysis in a Broiler House Using Sensor Data

Abstract

Environmental conditions within broiler houses can change rapidly and require continuous monitoring to support timely management actions, particularly in tropical climates. This paper presents a short-term environmental trend analysis of a commercial broiler house based on multisensor data collected in Kampung Padang Rumbia, Pekan, Pahang, Malaysia. An IoT-based monitoring system was deployed to measure key environmental parameters, including temperature, ammonia (NH₃), carbon dioxide (CO₂), particulate matter (PM2.5), and pH, throughout a broiler production cycle. This study focuses on short-term temporal behaviour of the sensor data rather than long-term environmental assessment or biological performance. The collected time-series data were analysed using descriptive statistics and simple predictive techniques to examine short-term trends and fluctuations over hourly and daily intervals. Moving average analysis and basic regression models were applied to identify patterns and short-term changes in environmental conditions. The results indicate that temperature and air quality parameters exhibit noticeable short-term variability, particularly during periods of increased broiler activity and environmental load. Several instances of rapid parameter changes were observed, highlighting the importance of continuous monitoring compared to manual spot measurements. The findings demonstrate that simple trend analysis using sensor data can provide early indications of changing environmental conditions within a broiler house. This work supports the use of IoT-based monitoring as a practical tool for short-term environmental awareness and serves as a foundation for future development of early warning and decision-support systems in smart poultry farming.