Part I: Predictive Maintenance using Azure ML

Motivation

In this series of blog posts we will explore how to use Azure Machine Learning for Predictive Maintenance. It will be completely commercial free until the very last blog post, where we will sneak in a pointer to our Enterprise Asset Management solution. So whether you use our software or not, you should be able to get some valuable insight into Azure ML based Predictive Maintenance.

Predictive Maintenance

Within Maintenance Management we deal with different maintenance strategies / categories:

  • Corrective Maintenance
    Maintenance is performed as a result of a breakdown or some kind of reported anomality.
  • Preventive Maintenance
    Time, operation-count or calendar based maintenance.
  • Condition Based Maintenance
    Maintenance is perfomed based on the actual/current condition of the asset.
  • Predictive Maintenance
    Maintenance is scheduled as a result of statistical-process-control often combined with realtime asset monitoring.

Predictive Maintenance can be seen as an advanced form of Corrective Maintenance, where we perform maintenance because we statistically expect a breakdown or an anomality. The difference between Preventive Maintenance and Predictive maintenance is, that preventive maintenance is most often based on empirical evidence and linear projections on operation-counts or hours-in-use, as opposed to true statistical evidence. The obvious weakness of Preventive Maintenance is, that maintenance will often be either too soon or too late, as it is very seldom we see any kind of true linearity in observed machine breakdowns!

Statistical Analysis sounds pretty boring!

Knowing that statistical analysis is at the heart of Predictive Maintenance, we probably all feel a little discouraged, as it’s the very few of us who truely understood the Poisson and the Erlang Distributions back in high school or at the university. Stating statical problems is however quite fun, so let’s start with the fun part, and just assume we have some wizard at our disposal, who will take care of the analysis part.

Here some of the infinitely many statistical questions in the predictive maintenance category:

  • Do we have correlations between breakdowns?
    If we can determine statistically that a breakdown in one area of the production line will later cause a breakdown in another area, then we should issue a predictive work order immediately after the first breakdown, thereby avoiding a potentially costly ripple effect of breakdowns.
  • Is there a correlation between vibration levels and breakdowns?
    Or machine temperature, or RPM’s or…. If this can statistically be determined then predictive action should be taken if vibration reach a certain threshold.
  • Does the skill level of an electrician affect our mean time beween failure (MTBF)? 
    In this case you should probably start to require certain certificate for specific work order types.
  • Is there a correlation between certain items we manufacture and breakdowns? 
    Maybe the production scheduling engine should route these items to different production resources.

All these questions are from a Maintenance Management perspective pretty darn interesting!
And the answers are even more interesting!

Statistical Analysis sounds amazing!

If we just knew how to apply it….! Well, this is where Azure ML comes into rescue. The only catch is that in order to get all the Azure ML machinery up running we still need to understand some aspects of statistical tests, the algorithms supported by Azure ML and how to get them up running! So, No pain, No Gain however annoying that is!

So providing that understanding will be the topic of the upcoming part II in this series of blog posts. Stay tuned!

PS: Credit to Tomas Grubliauskas for providing the hardcore background material for these posts!

0 Comments

Write a Comment

Your email address will not be published. Required fields are marked *

More articles

Picture1_feature-960x300

Digital transformation through modern EAM

The need to be a good steward of assets is not a new concept...

image004

Dynaway EAM for Manufacturing on Microsoft® AppSource

Efficient preventive maintenance, advanced spare parts management, asset history, optimized uptime, and reduced maintenance...

Enterprise Asset Management

5 Trends of Enterprise Asset Management for 2018

 

Future of asset management first page

Future of Asset Management

As a capital-intensive business, you need to ensure your assets are performing optimally, and...

image003

Checklists in the Next Dynaway EAM Release

In the coming May 1 release of Dynaway EAM and Mobile Client for Microsoft...

EAM features

The features you find in Dynaway EAM

  With the EAM module from Dynaway we give you the power to focus...

close