Development of IoT-enabled predictive maintenance system for industrial machinery

Thesis Abstract

In the contemporary industrial landscape, the reliance on machinery for sustained production efficiency underscores the critical need for effective maintenance strategies to minimize downtime and operational costs. Traditional reactive and scheduled maintenance approaches often result in unanticipated failures and inefficient resource utilization. Addressing these challenges, this study aims to develop an Internet of Things (IoT)-enabled predictive maintenance system that leverages real-time sensor data and advanced analytics to forecast machinery failures accurately. The primary objectives include identifying key parameters influencing machinery health, designing a scalable IoT architecture for data acquisition and processing, and validating predictive models through empirical testing. The research adopts a mixed-methods approach, combining quantitative and qualitative techniques to ensure comprehensive analysis. The quantitative component involves a descriptive survey targeting maintenance engineers and operators within the manufacturing sector, with a sample size of 150 participants selected via stratified random sampling to ensure representativeness. Data collection instruments include structured questionnaires focusing on existing maintenance practices, technological adoption, and perceived challenges in IoT implementation. Additionally, quantitative data from industrial machinery sensors—such as vibration, temperature, and pressure sensors—are collected over a six-month period from 20 operational machines across multiple manufacturing plants. Qualitative data are obtained through semi-structured interviews with key stakeholders to gain insights into system integration and operational considerations. Data analysis employs statistical techniques such as regression analysis to determine the relationship between sensor parameters and machine failure modes, while time-series analysis is utilized to identify failure trends. Machine learning algorithms, notably Random Forest and Support Vector Machines (SVM), are trained and validated for failure prediction accuracy. The theoretical framework integrates the Reliability-Centered Maintenance (RCM) theory and the Technology Acceptance Model (TAM), to analyze factors influencing system adoption and effectiveness. The development of the predictive model aims to achieve at least 85% accuracy in failure prediction, with the system being scalable and adaptable to various machinery types. Expected findings of this study include establishing a significant correlation between sensor data patterns and impending machinery failures, demonstrating the efficacy of machine learning techniques in predictive maintenance, and highlighting critical success factors for IoT system integration in industrial environments. The results are anticipated to contribute to the body of knowledge by providing a practical, technology-driven framework for predictive maintenance that enhances operational efficiency, reduces downtime, and optimizes maintenance schedules. Moreover, the study seeks to offer insights into the technological, organizational, and human factors influencing the deployment of IoT-based maintenance solutions within manufacturing settings. The study concludes that IoT-enabled predictive maintenance systems significantly improve machinery reliability and operational productivity, provided there is aligned organizational commitment and effective change management strategies. Recommendations include adopting modular IoT architectures for flexibility, investing in staff training on IoT and data analytics, and fostering collaborative environments involving stakeholders in the development and deployment process. Further research should explore the integration of edge computing to enhance real-time decision-making and expand the predictive maintenance framework to include condition-based monitoring of complex and hazardous industrial equipment. This research is intended to serve as a foundational reference for industry practitioners, researchers, and policymakers aiming to harness IoT innovations for sustainable industrial maintenance practices.