Abstract lines and connections in blue

Industrial Data Science Workshop

Registration is now closed


Please contact the Sargent Centre (sargent.centre@imperial.ac.uk)  if you would like to be placed on the waiting list


This course will introduce participants to process data challenges and how to solve them with data science. The syllabus is geared towards general Machine Learning concepts for regression models (supervised learning) and anomaly detection (unsupervised learning).

Date: 19-21 September 2023

Venue:   The Sargent Centre for Process Systems Engineering
               Imperial College London
               Roderic Hill Building
               South Kensington Campus
               London SW7 2BB


This course will be delivered by: 

  • Dr. Francisco Navarro (Sr. Data Science Training Lead at IFF and Visiting Researcher at Imperial College London) [in] 

Senior Data Science Training Lead, Imperial College London Visiting Researcher and chemical engineer working at IFF, a global leader and manufacturer of food, beverage, health & biosciences, scent and pharma solutions

Franciso NAvarro is leading the application of Machine Learning in manufacturing, where prodution engineers use industrial data science to monitor, troubleshoot and optimize their chemical processes. His industrial and research experience in Solvay, P&G, and Bayer uniquely combined data-driven methods with manufacturing systems, advances process control and process systems engineering.

He holds a Ph.D. in modeling and simulation where he designed (and patented) multiphase-flow sonoreactors.  He also visited Prof. Jensen’s lab at MIT (USA) during his doctoral studies. In 2012, he co-created cacheme.org, an open-source ChemE organization based at the University of Alicante (Spain).

  • Dr. Mattia Vallerio (Manufacturing Excellence Site Manager at Solvay and Advanced Process Control Specialist) [in] 

As the Manufacturing Excellence Site Manager for Solvay's Spinetta Marengo production site, Mattia Vallero is leading a team in charge of delivering operational excellence leadership, guide the site digital transformation, implement advanced process control, industrial data analytics and operational technology solutions.

Mattia Vallerio has almost 10 years of experience in digital transformation and performance optimization of chemical production sites.  He has a strong experience in cross-functional project & team management, focusing on delivering value through operational excellence best practices, digitalization, advanced data analysis and advanced process control.

He holds a master degree in Chemical engineering from Politecnico di Milano and a Ph.D. in Engineering Science from KU Leuven.

Before his current position he worked @BASF Antwerp as Advanced Data Analytics lead and Advanced Process Control Engineer.  In this role he kick-started the industrial data science field within BASF and he was at the fore-front of the BASF Antwerp site digital transformation.

  • Dr. Carlos Perez (Industrial Data Scientist at Solvay and Optimization Specialist) [in] 

Carlos Perez Galvan is an Industrial Data Scientist at Solvay in Brussels, Belgium. Currently, he is the technical leader of an Advanced Analytics corporate team that collaborates with all of Solvay's businesses. 

In cooperation with the Advanced Process Control team, he focuses on leveraging data analytics, process control and process systems engineering methods to optimize plant performance. 

During his 7+ years career in P&G (modelling and simulation) and Solvay he has had the opportunity to develop practical expertise in the fields of modeling, simulation, optimization and machine learning. 

He holds a Ph.D. in Chemical Engineering from University College London. He graduated from Universidad Autonoma de Coahuila in Mexico as a Chemical Engineer in 2012.

The instructors combine more than 20+ years of industrial data science experience covering machine learning, first-principle modeling and simulation, optimization, process control, and chemical engineering applied to chemical and process industries.  

Their research is co-authored with Reinforcement Learning experts from Imperial College London and Manchester University: 

  • Industrial data science – a review of machine learning applications for chemical and processes industries [React. Chem. Eng., 2022, 7, 1471] 
  • Industrial Data Science for Batch Manufacturing Processes (arXiv:2209.09660v1 [cs.LG] 20 Sep 2022) 


Day 1 – Industrial data science 

  • Distillation tower (full example) 
  • Industrial databases (tags, historians, and automation pyramid) 
  • Contextual data (asset hierarchies, batch events) 
  • Quality and tabular data (LIMS, ERP) 
  • Data democratization and software alternatives 
  • Hands-on session (connect to databases with Excel, ODBC, and RestAPIs).  


Day 2 – Monitoring assets  

  • Batch dryer example   
  • Defining KPIs for continuous and batch processes (feature engineering)
  • Tracking variability (visual analytics, statistical process control, robust statistics)  
  • Batch data alignment (e.g., time warping)  
  • Machine learning for anomaly detection (KNN, PCA, Autoencoders) 
  • Identifying plant changes in the Tennessee Eastman Process 
  • Hands-on session (Bring your own data!) 


Day 3 – Troubleshooting processes  

  • Problem definition 
  • Screening process variables (bootstrap forest, decision trees, and boosted trees) 
  • Improving processes (sensitivity analysis, explainable AI with SHAP) 
  • Modeling processes (missing data, Lasso regression, and neural networks) 
  • Industrial applications (inferential sensors and digital twins)  
  • Hands-on session (Bring your own data!) 



Registration information

Registration Fee: 

 Please note that the maximum capacity is 30 participants for this event

  • Industrial price: £1050 
  • Academic price: £250
  • Sargent Centre Academic price: £200
  • Industrial Consortium Partners (One place per company only): Free  

To register, please follow this  link .


Full refunds, less 10% administration fee, will be given for cancellations that are received in writing on or before 31st August 2023. After this date, until 10th September 2023, participants who cancel will receive refunds of 50% of the registration fee paid. No refunds will be provided for cancellations received after 11th September 2023. 

Substitutions may be made at any time, whilst a valid place is held. The organizer cannot accept liability for costs incurred in the event of a course having to be cancelled as a result of circumstances beyond its reasonable control.  

Practical information

Venue:   The Sargent Centre for Process Systems Engineering
               Imperial College London
               Roderic Hill Building
               South Kensington Campus
               London SW7 2BB

For the campus website, use this link.

Closest Underground Stations are South Kensington or Gloucester Road.

Affordable summer accomodation, walking distance from the Roderic Hill building, can be enquired about via this link. The accomodation to enquire about is Prince's Garden or Beit Hall.