Data Analytics in the Era of the Industrial Internet of Things

Chapter 1: Industrial Internet of Things FrameworkLayered View of IIoT systemsAnalytics Capabilities in IIoT Systems Can Increase Job Satisfaction Examples of IIoT Business Models Power Distribution Systems in the IIoTIIoT in Process Control Alarm Management Power Generation Turbines Anomaly Detection Increase Share of Wallet of Industrial Services and Products Power Transformers and Utility Equipment Analysis Demand Forecast of Products and Spare Parts ReferencesChapter 2: Industrial AnalyticsMachine Learning Supervised Machine Learning Decision Trees for Classification and RegressionRandom Forest Classification and Regression Neural Networks for Classification and Regression Sentiment Analysis and Machine Learning Support Vector Machines Unsupervised Machine Learning Association Rule MiningK-Means Clustering Anomaly Detection Machine LearningAnalytic Conduits References Chapter 3: Machine Learning to Predict Fault Events in Power Distribution Systems Problem Statement Background Data for Forecasting Fault Events in Power Distribution Grids Forecasting Fault Events Creation of Machine Learning Models Zone Prediction ModelsSubstation Prediction ModelsInfrastructure Prediction ModelsFeeder Prediction ModelsProactive Fault Analytics Helps Improving the Business Model and Employee SatisfactionReferencesChapter 4: Analyzing Events and Alarms in Control Systems Problem StatementBackgroundAnatomy of Alarms in IIoT Distributed Control SystemsAlarm Data Alarm Management Analytics Models Sequence Pattern Mining and Association Rule MiningAlarm Baskets Alarm De-chattering AnalysisAlarm Sequence AnalysisMeasures of Significance or Metrics for Sequence AnalysisEnhancing Expert Knowledge of Plant Operations Through Advanced Analytics Alarm ManagementReferencesChapter 5: Condition Monitoring of Rotating Machines in Power Generation Plants Problem Statement BackgroundTurbine Telemetry DataAnalytics for Anomaly Detection of Rotating MachinesStatistical Analysis of Turbine DataClustering Analysis of Turbine DataAnomaly Detection Using Connectivity-based Outlier FactorEnhancing Domain Knowledge of Power Engineers Through Anomaly Detection SystemReferencesChapter 6: Machine Learning Recommender for New Products and Services Problem StatementBackgroundHistorical DataProduct and Services Recommender AnalyticsCustomer Classification AnalyticsMarket Basket AnalysisSentiment Analysis Enhancing Domain Knowledge of Service Engineer Salespeople Through the Product and Services Recommender System ReferencesChapter 7: Managing Analytic Projects in the IIoT Enterprise Definition Phases of an Analytics Project in the IIoT Enterprise Delivery Framework for IIoT Advanced Analytics ProjectsSustaining Phase Requirements Engineering Project Management Process Data Preparation Phase Analytics and Implementation Phase Technical Solution Process Verification and Validation Processes Agile Kanban Development Lifecycle Barriers for the Implementation of Analytic Projects in the IIoTLack of clear business value Absence of Large User BaseTakes Too Long to Develop the Solution Organization Focused on Short-term GainsHigh-level Complexity Organizational Readiness for Change References Conclusions List of Abbreviations