ESR 3. Project Host: Hugin

Host institution: HUGIN              Duration 36 Months                   

Objectives:  This project would create data driven and knowledge enhanced Bayesian Network models that are able to predict the level of human operator performance and recommend adaptive automation. This model will take into account both human factors (level of stress, fatigue, work overload), based on biosensors on the operator, data streams from sensors from equipment sensors, along with assessment of task complexity and risk to predict operator performance and human error rates. The model will support real time response and support scenario-based ‘what-if’ sensitivity analysis. The creation of the Bayesian Networks will cover both structure and parameter learning, and online model refinement. The project will draw on both subject matter expertise and data, and will compare purely data driven BBN model induction, will purely subject matter expert knowledge driven BBN induction, and knowledge enhanced BBN models. The models will provide safety critical monitoring – triggering interventions in critical scenarios – and also can be used to develop guidelines through their use as simulations of systems. 

The candidate should preferably have the following skills:  

  • Master’s degree in computer science, artificial intelligence, mathematics or related field of study  
  • Knowledge of probabilistic graphical models will be considered an advantage   
  • Strong skills in software development (preferably Java or .Net), knowledge of python is considered an advantage  
  • Ambitious in research and problem solving 
  • Good communication skills in English (both in writing and orally) 
  • Strong motivation to work in a small team and individually 

Expected Results:

Bayesian Belief Network for real-time decision support and diagnosis in safety critical scenarios

Contributions for safety guidelines in modelled case-studies

Publish a number of high-impact articles communicating the results of the research project

Planned secondment(s): The PhD student is going to be seconded on M6 in TU DUBLIN for 12 months to work on the LIVE LAB 3. Then on M18 the ESRs is going to be seconded in POLITO for 6 months to work on process safety data monitoring for LIVE LAB 3.

to access the application form click on this link: Application form