Overcoming the bottlenecks on the path to building Predictive Maintenance capabilities

Amidst all the hype around Industry X.0, you probably already know that implementing predictive maintenance capability has several advantages which include reducing machine downtime and avoiding unnecessary maintenance costs while adding revenue streams for equipment vendors with aftermarket services.

However, there are hurdles you need to overcome around process and data when incorporating predictive maintenance technology into your companies’ operations.

In this article, I will walk through the most common implementation obstacles that you need to overcome in order to successfully develop this capability internally.

Lack of internal know how on how to setup the program

Developing a systematic approach to predictive maintenance allows you to successfully build a real-time system using a predictive algorithm. The high level five-step workflow described below can provide a good structure:

  • Build a robust pipeline to sensor data – Gather data from databases, spreadsheets, and web archives, and ensure the data is in the right format and organized.
  • Developed and automate data Preprocessing approaches– Clean the data by removing outliers, aligning time series, and filtering out noise.
  • Feature Extraction – Capture higher-level condition indicators, such as frequency domain or time-frequency domain features, instead of feeding raw sensor data into the model.
  • Model training– Build models that classify equipment as healthy or faulty, that can detect anomalies, or that can estimate remaining useful life for components
  • Model deployment– Generate code and deploy models as an application on hardware.

Lack of proper Data Pipelines

Remember that predictive maintenance is not just about machine learning algorithms.  The most important aspect is that good and enough data must exist to create an accurate model. This data typically gets generated by machine sensors. Robustness of any model developed depends on how data is logged: preferably, machines will include logging options that can be modified to record more data, or simulation tools can be used to combine simulated data with available sensor data to build and validate predictive maintenance algorithms.

While developing a model, Data Scientists should avoid a condition where their systems get “Data starved”, a situation where little or no data is collected until a fault occurs. To prevent this, companies can change the data logging options to record more data, perhaps on a test fleet if production data is not available. It is also possible to generate test data using simulation tools by creating models that cover the mechanical, electrical, or other physical systems to be monitored and then validating against measured data.

Lack of Historical failure data points

Algorithm needs to know what conditions typically represent failure or precede failures. Hence, Failure data is a fundamental element of predictive maintenance. Unfortunately, this data may not always exist if maintenance is performed so frequently that no failures occur.

Even without failure data, unsupervised machine learning techniques can be used to identify normal and faulty behavior. For example, data could be collected from several sensors on an aircraft engine. A dimensionality reduction technique such as principal component analysis (PCA) could then be used to reduce the sensor data into a low-dimensional representation for visualization and analysis. In this representation, healthy equipment data may be centered around a normal operating point, while unhealthy equipment may be seen as moving away from normal conditions.

Lack of  skills to leverage the right Predictive Analytics algorithm

There’s a big difference between identifying a failure source and knowing how to predict it. That’s why engineers need to clearly define their goals—such as longer cycles and decreased downtime—and think about how a predictive maintenance algorithm affects them. They then should build a framework to test algorithms and estimate their performance, so they can get immediate feedback during design iterations. They can then use this framework to test simple models and apply their knowledge of the data to try more complex model types. They should keep things small, validate against data, and iterate until they are confident with their results.

Obstacles aside, data scientists and engineers can take solace in realizing that predictive maintenance is an achievable goal if they can locate the best balance of tools and guidance. The onus though is on engineers and data scientists to determine the features, methods, and models that work best for them—and keep iterating until they fully master these techniques.


Views my Own.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s