Skip to content

Predictive Maintenance (PdM) Data Analytics and Machine Learning 3-day interactive training course to taking pro-active measures based on advanced data analytics to predict and avoid machine failure


About the Course

The amount of data we have today, and the ease of access allows us to utilize this data in ways that were not previously possible. One of these ways is to optimize our maintenance activities through analyzing our predictive and preventive maintenance activity data.

Identifying failures through communicating with our industrial machines has now become easier than ever, this allows us the opportunity to determine failure patterns and their progression over time, finding root causes of failures in a more reliable and easier fashion and better defining failure rates of equipment and components allowing a more optimized range of activities.

It is well documented that the earlier we detect issues the better opportunity we have at not allowing them to reach terminal or catastrophic failure points and the better we can plan for outages to replace or repair these issues. This is becoming more accurate and simpler to do using analytics for predictive and preventive maintenance activities.

Key Take-Aways:

Expertise and knowledge are the basis when working within your industry organization, or as an industrial organization, stands on the principle of digitization.

  • As a maintenance professional, you are expected to do that you have sufficient knowledge and are able to use the latest monitoring techniques.
  • As an industrial company you want to work efficiently, tasks keep them internally where possible and ensure that assets function optimally on an operational and economic level.

Course Outline

  • Predictive maintenance techniques and which works for what
  • Understanding the P-F curve and the effectiveness of predictive maintenance techniques
  • The different maintenance strategies and relating them to specific failure types.
  • Failure characteristics as a determinant for planning maintenance activities
  • Weibull distribution and its utilization in identifying progression of failure modes
  • Using risk-cost profiles to determine optimal activities and their frequency
  • Data analytics for maintenance calculations on:
    • Weibull distributions
    • Poisson methods

    • Markov distributions

    • Monte Carlo simulations

    • Markov-Monte Carlo Series (MCMC)

  • Integrating digitized tools like CMMS, Historians, Condition monitoring software, DCS and other systems into one big data pool allowing for a rich data source for more accurate analysis.
  • Introduction to maintenance 4.0 and Big data concepts
  • Using reliability software tools to develop availability simulation models to accurately identify optimal activities over time.