In this talk, we present an industrial use case on “anomaly detection” in steel mills based on IoT sensor data. In large steel mills and manufacturing plants, the top reasons for unplanned downtime are:
• Failure of critical asset
• Quality spec of the end product in line not being met
• Operational limits outside the recommended range (e.g. process, human-safety, equipment-safety, etc.)

Unplanned downtime or line stoppage leads to loss of production or throughput and revenue loss.

Anomaly detection can serve as an early warning system, providing alerts on anomalous behavior that could be detrimental to the equipment health or affect process quality. In this work, we are performing multi-variate anomaly detection on time-series sensor data in a steel mill to help the maintenance engineers and process operators take proactive actions and help reduce plant downtime. Anomaly is presented to the customer in terms of:
• “time-intervals” – startTime: endTime chunks that exhibit deviant behavior
• “anomaly-state” – type association of anomaly to a specific pattern or cluster state
• “anomaly-contribution” – priority association to sensor signals that exhibited deviant behavior within the multi-variate list (more like signal importance)

We shall introduce the approach, where we reformulate the unsupervised modeling to a supervised formulation to incorporate SHAP, LIME, and other explainable tools. We shall illustrate the steps to provide the above-mentioned meta-data for an anomaly to make it explainable and consumable for the end-customer.

 
 

Outline/Structure of the Talk

  • Introduction (1 min)
  • Problem Statement (2 mins )
  • Approach and Framework (6 mins)
  • Model Interpretability for few real use cases (8 mins)
  • Conclusion (1 min)
  • Q&A (2 mins)

Learning Outcome

After attending this session, the attendees will ...

  • Understand how model interpretability can be applied in unsupervised anomaly detection applications
  • Get familiar with future research possibilities in the manufacturing sector specifically with respect to asset health.

Target Audience

Data Scientists, Machine Learning/Deep Learning Practitioners, Industry Professionals from Manufacturing Sector, Researchers, Students & Faculty Members from Engineering and Technology.

Prerequisites for Attendees

Familiarity with the fundamentals of machine learning.

schedule Submitted 7 months ago

Public Feedback

comment Suggest improvements to the Author
  • POOJA BALUSANI
    By POOJA BALUSANI  ~  6 months ago
    reply Reply

    Hello,

    The abstract and slide deck has been updated as per the recommendation by the program committee. 

    Thank you 

  • Kuldeep Jiwani
    By Kuldeep Jiwani  ~  6 months ago
    reply Reply

    Hi Pooja,

    This is an interesting approach that you have proposed for identifying outliers.

    Your abstract is more focused on explainable AI, although this is a good topic. But going through your proposal it seems you have done good work in converting time series data into discrete states and then modelling them as classes for a supervised model. So just a generic suggestion, consider highlighting the application part also. As this seems to be your original work and shows the uniqueness of your work. While concepts of explainable AI are valuable and can be obtained in general, but your own experience is exclusively available to you. Show focusing on sharing that seems more valuable.

    • POOJA BALUSANI
      By POOJA BALUSANI  ~  6 months ago
      reply Reply

      Hello Kuldeep,

      Thank you for your valuable suggestion. We shall spend more time on problem formulation to highlight the application part as well. 

  • Natasha Rodrigues
    By Natasha Rodrigues  ~  7 months ago
    reply Reply

    Hi Pooja,

    Thanks for your proposal! To help the program committee understand your presentation style, can you provide a link to your past recording or record a small 1-2 mins trailer of your talk and share the link to the same?

    Thanks,

    Natasha