Development of IoT-enabled wearable sensors for real-time livestock health monitoring

Thesis Abstract

Livestock health management remains a critical challenge in modern animal husbandry, often hindered by limited real-time monitoring capabilities, delayed disease detection, and inadequate data integration, leading to economic losses and compromised animal welfare. This study aims to develop and evaluate an Internet of Things (IoT)-enabled wearable sensor system designed to facilitate continuous, real-time monitoring of livestock health parameters, thereby enhancing disease prevention, timely intervention, and overall farm productivity. The specific objectives include designing a low-power wearable sensor prototype capable of capturing vital signs such as body temperature, heart rate, and movement patterns; establishing wireless data transmission protocols suitable for farm environments; and assessing the system’s accuracy, reliability, and impact on herd health management. The research adopts a mixed-methods approach, integrating engineering design principles with quantitative validation and qualitative stakeholder feedback. The study population comprises 150 cattle across three commercial farms within a major agricultural region. A stratified random sampling technique selects 50 animals from each farm to participate in the sensor deployment phase. Data collection involves continuous sensor data logging over a six-month period, supplemented by farmer interviews and focus group discussions to gauge usability and operational impact. The sensor system prototypes are developed using embedded systems technology and tested for validity and reliability through laboratory calibration against standard veterinary diagnostic instruments, employing Bland-Altman analysis and intraclass correlation coefficients. Data analysis encompasses statistical techniques such as regression analysis to examine correlations between sensor measurements and traditional health indicators, and repeated-measures ANOVA to assess variations in health parameters over time. Qualitative data from stakeholder feedback are analyzed thematically to identify usability challenges and acceptance factors. The anticipated results include high concordance between sensor data and clinical assessments, demonstrating the sensors’ accuracy; statistically significant improvements in early disease detection compared to conventional monitoring; and positive stakeholder perceptions regarding the system’s practicality and value. This research contributes to the existing body of knowledge by bridging the gap between IoT technological advancements and livestock health monitoring, demonstrating the feasibility of low-cost, sustainable wearable sensors tailored for farm environments. It proposes an integrated framework for deploying IoT solutions in agriculture, supported by empirical evidence of improved health outcomes and operational efficiencies. The study also contextualizes the application of relevant theories such as the Technology Acceptance Model (TAM) and UTAUT (Unified Theory of Acceptance and Use of Technology) to interpret user acceptance and adoption factors. The main conclusion underscores that IoT-enabled wearable sensors can significantly enhance livestock health management through real-time data acquisition and remote monitoring, ultimately reducing mortality rates, veterinary costs, and productivity losses. Based on these findings, recommendations include scaling the system for broader farm deployment, integrating AI-driven analytics for predictive health alerts, and developing user training programs to foster farmer engagement. Future research should explore the system’s adaptability to different livestock species, environmental conditions, and integration with farm management software, thereby advancing precision agriculture and sustainable livestock practices.