PhD grant DC-05 -Chemical Engineering- PROSAFE Doctoral Network

Framework development for benchmark analysis and selection of QRA models.

The advertised subproject is fully funded by the Marie Skłodowska-Curie European Training Network “PROSAFE”. It will be carried out by one doctoral candidate at NOVOTEC and  enrolled at the Universitat Politècnica de Catalunya (UPC) Chemical Process Engineering doctoral program,  over a period of 36 months.

The candidate will work at the Major Accident and Process Safety department of NOVOTEC (Applus+ group). The department has more than 20 years of experience and a workforce of 14 specialists. NOVOTEC is specialized in the field of industrial safety and environmental protection, mainly for chemical, energy and oil refining companies, in Spain and abroad and is very active in research projects in this field.

The recruited researcher will  select the best models and methodologies (from both practical and scientific points of view) to ensure consistency in their practical application to assess major accidents effects in the frame of QRA, to reach the proposal of a convergent methodology able to be applied by the different teams working in the different State Members of the EU.

Training on advanced uncertainty and sensitivity analysis for QRA studies will take place at UPC and training on big data and artificial intelligence and their potential for improving QRA at Port de Barcelona (PdB, Spain).

The methodology features the following key aspects: (1) A systematic collection and benchmarking analysis will be made of the software commonly used in both general modelling of major accidents and, more specifically, in the QRA developed for the Safety Report required in the Seveso III Directive. (2) A critical and comparative study will be done, based on real representative cases (LNG unloading, regasification and storage, ammonia and methanol storage and distribution). (3) Validation of the results by comparison with liable data (even if rather scarce) available from real accidents will be made. (4) Sensitivity and uncertainty analysis of the results will be applied to ensure their reproducibility and consistency.