Training – Advanced analytics for quality improvement and operational efficiency


Analytics for industrial operations

How to achieve lean operations using data analytics?

Target audience

Engineers, Operations Directors, Plant Managers, Energy Managers, Analytical teams.


  • Duration: One day workshop (8h):
    • December 2nd 2015
  • Location: European Data Innovation Hub @ AXA, Vorstlaan 23, 1170 Brussels
  • Price: 300€ per participant
  • Limited for 8-12 participants


Please register using Eventbrite following this link


Today the manufacturing industry produces huge amounts of data on a daily basis.

“The role of data in manufacturing has traditionally been understated. Manufacturing generates about a third of all data today, and this is certainly going to increase significantly in the future. Data forms the backbone of all Digital Manufacturing technologies, which will be the centerpiece of the strategy for advancing Manufacturing in the 21st century.” (Source: Predictive Analytics World).

Concepts such as Industry 4.0 and Industrial Internet of Things are only a few examples that can lead to real added value and significant savings. In many of today’s organizations highly advanced technology is working alongside assets that have been in use for many years, but all of them are supporting different data types, formats and frequencies. These large amounts of manufacturing data generated by machines and equipments therefore often stay unused. What if this cemetery of data was actually a real goldmine, which could conceal so many opportunities for your company? What if, from your flaws, you could learn lessons and forecast the future? And how can gained knowledge be easily shared with the rest of the organization? Analytics can support these industrial companies and turns this data into valuable pieces of information (actionable intelligence). This training provides a guideline how industrial organizations can use advanced analytics to increase performance and lower their operational costs. Besides this the course will also increase the understanding and collaboration of data analysts and business users to adopt and deploy seamlessly predictive analytics and achieve lean operations.


The management of production processes can get complicated because of the subtle interaction between the various manufacturing steps and the influence of many parameters. Maintaining the process quality and stability is a real challenge for all the industries. Analytics can bring an added value and help organizations to drive new efficiencies.

  • explore variability of past operations
  • search for anomalous events and correlations (detect abnormal patterns)
  • diagnose root causes of drifts
  • identify patterns
  • predict and optimize performance
  • optimize energy use
  • lower costs

In this training we will show you how advanced analytics can accelerate diagnostics and improve operations performance.

Learning objectives

After the training participants will be able to define and understand an analytics project:

  • Have a more detailed understanding of analytics in an industrial context
  • Understand and select the correct methods and techniques
  • Explain what a model is
  • Define an analytics project workflow


  • Introduction to data analytics/data mining in industry
  • Analytics and continuous improvement (BI, Six Sigma, TQM, …)
  • Data analytics process and applications
  • Data analytics techniques: Clustering, Classification, Dependency modelling, Summarization, Regression, Case-based learning, Time-series analysis, …
  • What is a model?
  • Method selection
  • Challenges in importing and pre-processing data
  • DATAmaestro cloud analytics (navigating, upload and merge data via DATAserver, managing your data, performing data exploration)
  • Hybrid approach
    • Explorative methods (histograms, dendrograms, scatter plots, curves, …)
    • Learning methods (decision trees, ensemble trees, multi-linear regression, neural networks, K nearest neighbors, …)
  • Project workflow using a practical use case from a manufacturing plant (loading the data, creating attributes and object sets, creating graphs, segmentation, predictive analytics, modeling, implementing the model, creating functions, reporting, …)
  • Conclusion


During the course we will use a practical project. You are also free to define your own project from your organization and bring relevant data. This could be for instance a project related to product quality, energy, defect rates, optimizing the process, … It is also recommended to bring your own laptop.

About the trainer

Dominique ArchambeauDominique Archambeau is an electrical engineer and has been a research engineer and assistant at the applied acoustics department of the University of Liège, where he developed expertise in digital signal processing field (audio and video signals), statistics and data mining area. He is a founding partner of PEPITe (, a Liège based player in industrial applications for data mining, machine learning and advanced Big Data analytics, and one of PEPITe’s senior analysts. Throughout the years Dominique has worked with many of our clients on their data analytics challenges and also contributes both to client opportunities, long-term R&D projects, proof of concepts and training and support activities. During the analytics courses his lessons and instructions are easy to understand and can be implemented quickly. His approach to working with clients moves beyond simple teaching though, as he gets very “hands-on” to improve your data analytics skills. Besides that Dominique is a big believer in sharing ideas with colleagues and clients, seeking their insights and feedback.

Leave a Reply

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

You are commenting using your 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

This site uses Akismet to reduce spam. Learn how your comment data is processed.