Skip to Content
Product Information

New Weibull-Based PoF Curves in SAP Predictive Asset Insights

Introductory note: just in case you are wondering about the product mentioned in the title; the product previously known as SAP Predictive Maintenance and Service (PdMS) as of the 2011 release now includes the simulation capabilities of SAP Predictive Engineering Insights enabled by Ansys, and was renamed to SAP Predictive Asset Insights (PAI).

 

A brief recap of machine learning in Predictive Asset Insights

Over the years we have assembled a rich set of machine learning capabilities in PdMS and now PAI. This includes:

  • The set of out-of-the-box algorithms for supervised and unsupervised training and scoring, like Logistic Regression or Principal Component Analysis
  • The “super-algorithms” which on their own churn through several of the individual algorithms, for Automatic Anomaly Detection and Automatic Failure Prediction
  • Failure Mode Analytics, which allows you to do a text analysis of the long text of a notification, and the description of a failure mode, and have the system propose which failure mode should be assigned to which notification, and display occurences
  • Leading Indicator Analytics, which allows you to determine which of the sensor data streams (indicators), and for which specific conditions (e.g. “Inlet temperature > 125 degrees Celsius”), may have contributed to a breakdown notification. As a bonus such conditions can be turned into a new rule in the PAI rules engine.

 

New in PAI 2011: Failure Curve Analytics

As a new machine learning feature Failure Curve Analytics (FCA) rounds out the ensemble of existing algorithms. It is based on a traditional Weibull model. This is very interesting to customers who have no, or incomplete, or inconsistent sensor data, as FCA only needs breakdown notifications to calculate a Probability of Failure (PoF) curve, and can achieve good results even with few notifications.

Using FCA works in in four steps.

 

Step 1: create a new FCA model configuration and set general parameters

In PAI 2011 you will find in the Machine Learning Engine group a new application for “Failure Curve Analytics Model Configuration”

Open it to see a list of existing model configurations, and create a new one. Here you can enter, apart from its name and description, a few key parameters:

  • Age for Conditional Probability of Failure – FCA will always calculate a PoF curve for the fleet of selected equipment, but entering a number here will also calculate what the PoF is for the equipment having survived thus far, and surviving until that target age
  • First Installation Data is Maintained and Reliable – sometimes customers have incorrect or no information when an equipment was installed, i.e. when its initial age was zero. If this is the case FCA will allow to use the end of the first breakdown as the “birth” of the new equipment.
  • Equipment is Repairable – FCA assumes that after a repair an equipment is in an as-new state; if that is not the case the equipment is assumed to be scrapped.

Selecting these parameters influences what FCA calculates, and how it calculates it.

 

Step 2: selecting input business data

FCA works with three sets of input data:

  • A fleet of equipment – FCA allows the user to hand-pick which equipment which have been exposed to the same operating conditions shall be analyzed together. For this FCA offers a filter dialog with a few standard equipment attributes, and any custom attributes the customer may have added to the equipment.A fleet will be stored in a PAI “Fleet Group” (there is a separate application to view and edit them), and existing fleet groups can be reused for multiple FCA model configurations.
  • A set of failure modes – failure curves are always calculated per failure mode. The system will collect any failure modes assigned to the fleet of equipment, and the user can select all or some of them.
  • A date range of breakdown notifications – FCA will collect all breakdown notifications for the selected failure modes, for one or multiple date ranges

Step 3: train and score the model

In FCA we have introduced a handy new feature for machine learning applications: a single button which will train and score the FCA model configuration in one fall swoop!

 

 

If you want to or need to you can now watch the information / warning / error logs of the training and scoring run. What you will hope to see is a log without errors:

 

 

The training will calculate the parameters for a Weibull model, for the whole fleet of equipment, and the scoring will compute all the other output parameters, such as the age of the equipment, the time-to-failure, etc.

 

Step 4: view the output data in an equipment chart

When the training and scoring is done the user can select one of the equipment to navigate to its “object page” in the equipment application. There we have added a new entry in the Analytics section:

 

 

In that section the user will see a PoF curve. It is specific to (a) a model configuration (an equipment could have been used in multiple models, e.g. one for “All pumps older than 3 years” and one for “All pumps in location ABC from manufacturer XYZ”), and (b) a failure mode. The user can select from either.

 

The chart shows the current age of the equipment and the PoF curve, including a confidence interval.

 

 

The use can select a point on the chart for its specific age/PoF combination.

 

 

Selecting the point on the curve for the current equipment age also shows the calculated time-to-failure and that date.

 

 

Selecting a different model configuration or failure mode will show those results.

 

 

Using the information

Using the chart data they user can estimate how quickly the equipment may be in a critical situation.

 

More information

SAP Help is a good source for more information:

 

What else?

We have a solid set of additional features in mind (e.g. to write the results of the model into an indicator), but as this is fresh new functionality, we also plan to let customers play with it for a release, to see how well it works against their data. For some customers using Weibull analyses is common practice, but for other customers this is new ground.

 

What do you desire?

How about your company? Have you used Weibull models before, and what worked for you, and what not? What features are you missing? In which business processes did you use it? Would you want to use Weibull analyses in SAP applications?

I’d love to hear from you!

4 Comments
You must be Logged on to comment or reply to a post.
  • Thanks Oliver. This is them much needed functionality from our customers.

    I have couple of questions .

    Does FCA allow to calculate the time to failure in terms of “No of operations” apart from “no of days”? The reason behind this sometime equipment operate in one or two shifts and wear and tear of component depends upon no of operations and equipment loading .

    While customer may have target age for retirement of the assets but timely repair, maintenance also extends the life of the asset . How does failure curve analysis work in this scenario ?

    Equipment has multiple parts /components . Will FCA allow customers to roll up the Part level time to failure details for particular failure mode to parent equipment level ?

    Best Regards,
    Amar

    • Hi Amar,

      thanks for the reply!

      As for your first question, in this first 2011 release FCA only allows to measure days as the age of the equipment, but we know that this is not enough, that we need to extend it to other things you can measure on the equipment, like “kilometers driven” for a truck in a mine, or “count of products produced” for a machine on a shop floor. We want to open up FCA to make use of any numeric indicator which is enabled for the equipment. But first we wanted to hear back from more customers if they have any specific requirements for this. Let me know if you have details on what your customers need.

      As for your second question, FCA uses the malfunction start and end dates of breakdown notifications. Anything that is not a breakdown is seen as an extension tot he life of the equipment, as we do not regard the equipment to be “as-new” after a repair which was not for a breakdown. Of course that may need to change, to better take into account the subtleties of “refreshing” the equipment. Also here: if you have specific customer requirements, please let me know.

      And as for your third question, FCA currently does not do any roll-ups. We did discuss other kinds of data aggregations with customers, such as calculating the worst PoF across all failure modes of an FCA model configuration, or even the worst PoF value per failure mode across the equipment in a fleet which is used in an FCA model configuration, but not yet roll-ups across equipment hierarchies. As of 2011 PAI also has functional locations, so you’d want to see roll-ups there too. But this is also a larger PAI topic and not just related to FCA, as you’d want to roll up any kind of indicator, and not just specifically the PoF data.

      Best, Oliver

  • Dear Sir,

    Thank you for the great walkthrough about new feature to analyze failure data…I would think, this will really be a great breakthrough in IAM centric solution…

    I have few questions on the approach that currently in place for failure data analysis…

    First Question:

    When we talk about typical maintenance program, we do have various types of failure analysis depending upon severity of the failure, Viz.,

    1. Bad actor

    2. Pareto Analysis

    3. Root Cause Analysis

    4. Reliability Growth Analysis

    5. Monte Carlo Analysis

    6. Weibull Analysis

    7. Markov Analysis

    Each analysis has it`s own criteria to analyze the failure…Is there any significant criteria to chose only Weibull Analysis for ASPM?… because when I look at certain maturity criteria of current maintenance program, different clients may be using different analysis strategy to evaluate failures…Now irrespective of their current strategy, do they have to follow the “Weibull Analysis”?…when I look at our SAP IAM road map, we do have a “Root Cause Analysis” on the way…Is there any recommendation for maintenance team to chose either of them or both of them ?

    Second Question :

    I agree on a fact that, Weibull analysis is widely used in reliability centered maintenance and mainly used for Non-Repairable Technical Objects/Components…Is IAM with Weibull analysis addresses the MTTF(Mean Time to Failure) for each “Model” to evaluate premature failures/Age factor failures etc.,?

    I am curious to know that, with Weibull analysis, do SAP IAM provides any kind of option to set failure rate criteria for each of the model to draw a plot with respect to warranty periods to understand  reliability of the model at different operating conditions?…

    Sincerely,

    Pardha Reddy.C

    • Hi Pardha Reddy,

      thanks for your kind review!

      As for your first question, what we have and will deliver in terms of methods for failure analyses has pretty much followed customer demand. Different customers use different methods, and for some customers many of these methods are still pretty new. So because of our customer and partner influence councils responses you see root cause analysis on the roadmap, and Weibull as being delivered. We are also just starting some customer research on bad actor analysis.

      And just to clarify: the new Failure Curve Analytics (FCA) function resides in SAP Predictive Asset Insights (PAI, fka PdMS) (because that is the product where we make use of the Machine Learning Engine), and not in SAP Asset Strategy and Performance Management (ASPM). We are still working on an eventual data integration so that you could make use of the FCA data when e.g. doing an FMEA in ASPM.

      As for your second question, the MTTF/TTF calculation in FCA is done on a “fleet”, not an equipment model. You can assemble any equipment into a fleet, and yes, the TTF numbers are recalculated with each scoring of the model configuration. The idea is that you put the model configuration on automatically recurring scoring, like every day.

      The other things you mentioned (compare calculated PoF curve against manufacturer-provided curves and warranty periods, and see failure rate charts) have actually been in our sights as well, and we did discuss them in influence councils, but we wanted to let customers “play” with the newly delivered functionality for a few months before we decide what next features to add.