Chapter 1: Industrial Internet of Things Framework
Layered View of IIoT systems
Analytics Capabilities in IIoT Systems Can Increase Job Satisfaction
Examples of IIoT Business Models
Power Distribution Systems in the IIoT
IIoT 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
References
Chapter 2: Industrial Analytics
Machine Learning
Supervised Machine Learning
Decision Trees for Classification and Regression
Random Forest Classification and Regression
Neural Networks for Classification and Regression
Sentiment Analysis and Machine Learning
Support Vector Machines
Unsupervised Machine Learning
Association Rule Mining
K-Means Clustering
Anomaly Detection Machine Learning
Analytic 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 Models
Substation Prediction Models
Infrastructure Prediction Models
Feeder Prediction Models
Proactive Fault Analytics Helps Improving the Business Model and Employee Satisfaction
References
Chapter 4: Analyzing Events and Alarms in Control Systems
Problem Statement
Background
Anatomy of Alarms in IIoT Distributed Control Systems
Alarm Data
Alarm Management Analytics Models
Sequence Pattern Mining and Association Rule Mining
Alarm Baskets
Alarm De-chattering Analysis
Alarm Sequence Analysis
Measures of Significance or Metrics for Sequence Analysis
Enhancing Expert Knowledge of Plant Operations Through Advanced Analytics Alarm Management
References
Chapter 5: Condition Monitoring of Rotating Machines in Power Generation Plants
Problem Statement
Background
Turbine Telemetry Data
Analytics for Anomaly Detection of Rotating Machines
Statistical Analysis of Turbine Data
Clustering Analysis of Turbine Data
Anomaly Detection Using Connectivity-based Outlier Factor
Enhancing Domain Knowledge of Power Engineers Through Anomaly Detection System
References
Chapter 6: Machine Learning Recommender for New Products and Services
Problem Statement
Background
Historical Data
Product and Services Recommender Analytics
Customer Classification Analytics
Market Basket Analysis
Sentiment Analysis
Enhancing Domain Knowledge of Service Engineer Salespeople Through the Product and Services Recommender System
References
Chapter 7: Managing Analytic Projects in the IIoT Enterprise
Definition Phases of an Analytics Project in the IIoT Enterprise
Delivery Framework for IIoT Advanced Analytics Projects
Sustaining 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 IIoT
Lack of clear business value
Absence of Large User Base
Takes Too Long to Develop the Solution
Organization Focused on Short-term Gains
High-level Complexity
Organizational Readiness for Change
References
Conclusions
List of Abbreviations
Dr. Aldo Dagnino is an Industrial Engineer and received his M. A. Sc. and Ph. D degrees in the Department of Systems Design Engineering at the University of Waterloo in Canada. He has collaborated with various universities such as North Carolina State University and the University of Calgary where he held Adjunct Faculty appointments to conduct joint research, co-supervise graduate students, and create industrial internship programs to bridge academia with industry needs. Dr. Dagnino has 30 years’ experience developing advanced software solutions for industrial applications. The main focus of his work has been to bridge the technical fields of Computer Science, Software Engineering, and Industrial Systems Engineering for the development of new production systems that will enhance environmentally sustainable industrial processes and create new job opportunities. Dr. Dagnino has also utilized the fields of Artificial Intelligence, Machine Learning, Data Mining, Operations Research, Robotics, Software Engineering, Industrial Engineering, and Manufacturing Engineering in the development of new software products and services for electronics, telecommunications, electro-mechanics, oil and gas, power generation, manufacturing, and power transmission and distribution. Dr. Dagnino led the Advanced Industrial Analytics Group at ABB US Corporate Research and is currently leading the advanced analytics activities within the ABB Global Information Systems organization.
This book presents the characteristics and benefits industrial organizations can reap from the Industrial Internet of Things (IIoT). These characteristics and benefits include enhanced competitiveness, increased proactive decision-making, improved creativity and innovation, augmented job creation, heightened agility to respond to continuously changing challenges, and intensified data-driven decision making. In a straightforward fashion, the book also helps readers understand complex concepts that are core to IIoT enterprises, such as Big Data, analytic architecture platforms, machine learning (ML) and data science algorithms, and the power of visualization to enrich the domains experts’ decision making. The book also guides the reader on how to think about ways to define new business paradigms that the IIoT facilitates, as well how to increase the probability of success in managing analytic projects that are the core engine of decision making in the IIoT enterprise.
Useful for any industry professional interested in advanced industrial software applications, including business managers and professionals interested in how data analytics can help industries and to develop innovative business solutions, as well as data and computer scientists who wish to bridge the analytics and computer science fields with the industrial world, and project managers interested in managing advanced analytic projects.