Smart Manufacturing – Improving OEE via Predictive Maintenance and Anomaly Detection

Who Should Attend?


Duration: 3 Days
Training Date
  • 20 – 22 July 2020 (KL)
  • 10 – 12 August 2020 (Penang)
  • 19 – 21 October 2020 (KL)
  • 1 – 3 December 2020 (Penang)

Class Available

The main concept of the Industry 4.0 is to allow technologies and machines to communicate with human and business by exchanging data to make informed decisions. Technologies has allowed machines, devices, sensors and people to be interconnected and this results in enormous amount of data generated and exchanged. Such reform necessitates the systematic analytics on data to transform them into information could be used for decision making. Therefore, organizations must be able to adapt to big data phenomenon to meet the expectations of Smart Manufacturing. Incorporating a mixture of theories and hands-on, this course will guide you through the methodology to carry out an analytical project to improve machine availability. The essence of this course – the analytical methodologies to turn data into foresights will be the key to sustainable innovation in a smart manufacturing environment.

  • Address production challenges:
    • Improve Overall Equipment Efficiency (OEE)
    • Increase machine availability
  • Address analytical challenges:
    • Equipment and process complexity
    • Data quality
  • Articulate equipment anomaly and predictive maintenance scenarios
  • Identify relevant data sources and perform common data preparations
  • Build sophisticated prediction models
  • Evaluate model quality to relate back to business requirements
  • Deliver results to enhance availability of equipment
  1. Overview
    • What is Industry 4.0?
    • How big data analytics play a role in Industry 4.0 and Smart Manufacturing?
    • What could be done with big data analytics to solve business problems?
    • The concepts of data science
  2. Identifying and Getting Your Data Ready for Machine Availability Improvement
    • Where to obtain the relevant datasets?
    • How to cleanse and prepare your datasets?
    • The importance of incorporating multiple data sources
    • Transforming data to insights
  3. EDA: Exploratory Data Analysis
    • Understanding and exploring the information in your datasets
    • Descriptive statistics
    • Data visualization as part of data mining
  4. Predictive Analytics
    • Engineering and selecting the right features to model equipment breakdown characteristics
    • Apply anomaly detection techniques to identify abnormal conditions in the production line
    • Use machine learning to predict future equipment breakdowns
  5. Evaluation of Models and Results
    • Validation and prediction performance indicators
    • Mapping the results back to business objectives
  6. Delivering and Operationalizing Analysis Outcomes
    • Identifying and flagging abnormal equipment using an anomaly detection model
    • Predicting future equipment breakdowns using a trained model
    • Integration of the machine behavior prediction workflows into the smart factory ecosystem
    • Consuming equipment availability prediction results via dashboards for non-technical users
Case Study
  1. Predictive Maintenance
    • Predict failure of a machine from sensors’ reading
  2. Anomaly Detection
    • Identify rare events and detect abnormal behavior

Register Now

Getting interest to know more about this course? Pick the learning environment that you prefer.


Private Class Available

Contact us to find out more information if you are interesting in organizing private group class for your corporate.