Predictive Maintenance Course

Lesson 1 • Maintenance Strategy Landscape

Contrasts reactive, preventive, and predictive maintenance philosophies using cost and reliability metrics. Provides the strategic context for all subsequent PdM decisions.

Lesson 2 • Business Case for PdM Programmes

Quantifies ROI, downtime reduction, and safety benefits to justify PdM adoption. Equips learners to present data-driven proposals to stakeholders.

Lesson 3 • Reliability Engineering Basics

Introduces MTBF, MTTR, availability, and reliability distributions as quantitative foundations. These metrics benchmark PdM programme effectiveness throughout the course.

Lesson 1 • Thermography and Thermal Monitoring

Explains infrared thermography principles and thermal anomaly detection for electrical and mechanical assets. Establishes temperature rise as a leading failure indicator.

Lesson 2 • Ultrasound and Acoustic Emission

Introduces airborne and structure-borne ultrasound for detecting leaks, bearing defects, and electrical discharge. Complements vibration and thermal methods for comprehensive coverage.

Lesson 3 • Vibration Analysis Principles

Covers frequency, amplitude, and waveform concepts essential for interpreting vibration signals. Connects vibration patterns to specific mechanical faults identified in Chapter 1.

Lesson 4 • Oil and Lubricant Analysis

Teaches particle counting, viscosity testing, and spectrometric analysis to assess lubricant and component health. Ties oil condition to wear progression models from Chapter 1.

Lesson 1 • Wireless and IoT Sensor Networks

Addresses wireless sensor deployment, power management, and network topology for remote or rotating assets. Introduces IIoT connectivity as an enabler of continuous monitoring.

Lesson 2 • Signal Conditioning and Filtering

Explains amplification, anti-aliasing filters, and noise reduction techniques that ensure signal integrity. Prepares learners to produce clean data for analysis in later chapters.

Lesson 3 • Data Acquisition Hardware and Protocols

Covers DAQ cards, PLCs, and industrial communication protocols for transferring sensor data reliably. Links hardware choices to system scalability and integration requirements.

Lesson 4 • Sensor Calibration and Maintenance

Establishes calibration schedules, drift correction, and sensor health verification to sustain data quality. Ensures measurement reliability as the foundation for accurate diagnostics.

Lesson 1 • Data Labelling and Annotation

Covers fault labelling strategies, expert annotation workflows, and handling class imbalance in maintenance datasets. Prepares supervised learning datasets used in Chapter 5.

Lesson 2 • Data Integration from Multiple Sources

Merges sensor streams with CMMS work orders, process historians, and environmental data for richer context. Enables multivariate analysis and cross-asset comparisons.

Lesson 3 • Industrial Data Storage Architectures

Compares time-series databases, data historians, and cloud storage for high-frequency sensor data. Aligns storage strategy with retrieval speed and scalability needs.

Lesson 4 • Data Cleaning and Outlier Handling

Teaches missing value imputation, outlier detection, and duplicate removal for sensor data streams. Directly improves model accuracy in the machine learning chapters that follow.

Lesson 1 • Unsupervised Anomaly Detection

Applies clustering, PCA, and autoencoders to identify deviations without labelled fault data. Addresses the common industrial challenge of scarce failure examples.

Lesson 2 • Supervised Fault Classification

Trains decision trees, random forests, and SVMs on labelled datasets to classify specific fault types. Builds on feature engineering and labelled data from Chapter 4.

Lesson 3 • Statistical Process Control for Condition Data

Uses control charts, threshold setting, and alarm management to flag abnormal equipment behaviour. Establishes rule-based detection as a baseline before advanced models.

Lesson 4 • Model Validation and Deployment

Covers train-test splitting, cross-validation, and production deployment of diagnostic models. Ensures models generalise beyond training data before operational use.

Lesson 1 • Uncertainty Quantification in Prognostics

Applies Bayesian methods, Monte Carlo simulation, and conformal prediction to bound RUL estimates. Converts point predictions into confidence intervals for risk-informed decisions.

Lesson 2 • Degradation Modelling Concepts

Introduces health index construction, degradation path modelling, and failure threshold definition. Bridges fault detection from Chapter 5 to forward-looking prognostic estimation.

Lesson 3 • Hybrid Physics-Informed Data Models

Combines physics constraints with neural network architectures to improve generalisation and interpretability. Represents the current state of the art in industrial prognostics.

Lesson 4 • Data-Driven RUL Prediction Models

Trains regression models, gradient boosting, and LSTM networks to predict RUL from sensor trajectories. Leverages labelled run-to-failure datasets prepared in Chapter 4.

Lesson 1 • CMMS Integration and Work Order Automation

Connects PdM alerts to CMMS platforms to auto-generate work orders and schedule technician tasks. Closes the loop between prediction and physical maintenance action.

Lesson 2 • System Scalability and Architecture

Addresses microservices, containerisation, and cloud-native deployment for scaling PdM across large asset fleets. Ensures the platform grows with organisational needs.

Lesson 3 • Digital Twin Integration

Synchronises physical asset data with digital twin models to simulate failure scenarios and optimise interventions. Extends PdM from monitoring to virtual experimentation.

Lesson 4 • Operational Dashboards and Visualisation

Designs real-time dashboards displaying asset health scores, alerts, and RUL trends for operators and managers. Translates model outputs into intuitive visual formats for non-technical users.

Lesson 1 • Regulatory Compliance and Audit Readiness

Aligns PdM records, inspection logs, and maintenance evidence with industry safety and quality standards. Prepares organisations for external audits and certification reviews.

Lesson 2 • Continuous Improvement and Model Retraining

Establishes feedback loops, model retraining schedules, and root cause review processes to sustain PdM accuracy. Prevents model degradation as equipment and operating conditions evolve.

Lesson 3 • KPIs and Performance Measurement

Defines leading and lagging KPIs—OEE, false alarm rate, cost avoidance—to track programme health. Provides the measurement framework for continuous improvement cycles.

Lesson 4 • Organisational Change Management

Addresses technician training, cultural resistance, and cross-functional collaboration needed for PdM adoption. Recognises that technology success depends on people and process alignment.