Development of IoT-based sensors, big data processing, and prediction model for real-time monitoring system in manufacturing industry = 제조 산업의 실시간 모니터링 시스템을 위한 IoT 기반 센서, 빅데이터 프로세싱 및 예측 모델 개발

다국어 초록 (Multilingual Abstract)

With the increase in the amount of data captured during the manufacturing process, monitoring systems are becoming important factors in decision making for management. Current technologies such as Internet of Things (IoT)-based sensors can be considered a solution to provide efficient monitoring of the manufacturing process. In this study, a real-time monitoring system that utilizes IoT-based sensors, big data processing and a prediction model is proposed. Firstly, an IoT-based sensor that collects temperature, humidity, accelerometer and gyroscope data was developed. The characteristics of IoT-generated sensor data from the manufacturing process are as follows: real-time, large amounts and unstructured type. The proposed big data processing platform utilizes Apache Kafka as a message queue, Apache Storm as a real-time processing engine and MongoDB to store the sensor data from the manufacturing process. Secondly, for the proposed prediction model, Density-Based Spatial Clustering of Applications with Noise (DBSCAN)-based outlier detection and Random Forest classification were used to remove outlier sensor data and provide fault detection during the manufacturing process, respectively.The proposed model was evaluated and tested at an automotive manufacturing assembly line in Korea. The results showed that IoT-based sensors and the proposed big data processing system are sufficiently efficient to monitor the manufacturing process. Furthermore, the proposed prediction model has better fault prediction accuracy than other models given the sensor data as input. The proposed system is expected to support management by improving decision-making and will help prevent unexpected losses caused by faults during the manufacturing process. In addition, the overall proposed framework can be used as practical guidelines for the industrial practitioner in order to adopt IoT-based sensors, big data processing and machine learning model in manufacturing industry.

목차 (Table of Contents)

• Abstract i
• Table of Contents iii
• List of Figures vii
• List of Tables ix
• Chapter 1 Introduction 1
• 1.1 Research Background 1
• 1.2 Research Purpose and Contribution 4
• 1.3 Outline 5
• Chapter 2 Literature Review 7
• 2.1 IoT-based Sensor for Monitoring System 7
• 2.1.1 Monitoring system 7
• 2.1.2 IoT and monitoring system 8
• 2.1.3 IoT application in manufacturing 9
• 2.2 Big Data Processing 11
• 2.2.1 Big data in manufacturing 11
• 2.2.2 Big data technology 12
• 2.2.2.1 Apache Kafka 13
• 2.2.2.2 Apache Storm 14
• 2.2.2.3 NoSQL MongoDB 15
• 2.2.2.4 Integration of big data technologies 17
• 2.3 Machine Learning Methods in Manufacturing 18
• 2.3.1 Machine learning 18
• 2.3.2 Fault detection 20
• 2.3.3 Outlier data 21
• Chapter 3 Research Framework and Methodology 23
• 3.1 System Design 25
• 3.2 System Implementation 29
• 3.3 Data Collection 32
• 3.4 Experimental Environment 33
• Chapter 4 IoT-based Sensors 36
• 4.1 Proposed IoT-based Sensors 36
• 4.2 Result and Discussion 41
• 4.2.1 Computational cost 42
• 4.2.2 Network delay 43
• 4.2.3 Data format between JSON and XML 43
• 4.2.4 Trade-off between network delay and CPU usage 44
• 4.2.5 Comparison between Kafka and Mosquitto 45
• Chapter 5 Big Data Processing 47
• 5.1 Proposed Big Data Processing 47
• 5.2 Result and Discussion 49
• 5.2.1 Latency 50
• 5.2.2 Throughput 51
• 5.2.3 MongoDB vs CouchDB 52
• 5.2.4 MongoDB vs MySQL 53
• 5.2.5 Scalability of apache storm 54
• 5.2.6 Comparison with existing big data processing 55
• Chapter 6 Prediction Model 58
• 6.1 Proposed Prediction Model 58
• 6.2 Benchmark Dataset 64
• 6.3 Result and Discussion 65
• 6.3.1 Performance evaluation of the proposed model 66
• 6.3.1.1 Dataset I 66
• 6.3.1.2 Dataset II 67
• 6.3.1.3 Dataset III 67
• 6.3.1.4 Dataset IV 68
• 6.3.1.5 Dataset V 68
• 6.3.2 Impact of outlier data elimination on accuracy 69
• 6.3.2.1 Dataset I 70
• 6.3.2.2 Dataset II 71
• 6.3.2.3 Dataset III 72
• 6.3.2.4 Dataset IV 73
• 6.3.2.5 Dataset V 74
• 6.3.3 The implementation of proposed prediction model 75
• Chapter 7 Conclusions 76
• 7.1 Conclusion 76
• 7.2 Managerial Implications 77
• 7.3 Future Work 79
• References 80
• Appendix. Publication List 94
• Acknowledgements 97