Risk based decision making. Discussion on two methodological milestones

•The risk based design in process plants is introduced. The case study (allyl chloride production plant) is described.•The recursive operability analysis (ROA) is introduced and applied to the case study.•The fault trees are obtained from ROA and quantified. Numerical results are used to define design improvements.•The integrated dynamic decision analysis (IDDA) is introduced and applied to the case study.•Design improvements are discussed with respect to those proposed through the ROA and Fault Tree analysis.

Through the application to the risk based design of an Allyl Chloride production plant, the authors would like to illustrate the methodological approaches developed by two process safety pioneers, Norberto Piccinini and Remo Galvagni, both scientists and gentlemen, who devoted their professional lives to develop novel approaches to enhance accident prediction and prevention and to train young scientists.In particular, Norberto Piccinini has to be mentioned to have developed the Recursive operability analysis, a step ahead in the HazOp techniques, able to directly obtain the Logic Trees from the tables of the operability analysis, thus guaranteeing the coherence between the hazard identification step and the quantification step in the quantitative risk assessment.Remo Galvagni instead conceived and developed the integrated dynamic decision analysis (IDDA), an Event Tree empowered with conditionings, both logic and probabilistic. The tool, aimed to a correct and coherent application of probability theory according to the De Finetti's principles, allows the logical-probabilistic model to run integrated with a deterministic model of the plant to have guarantees of consistency and completeness in a risk assessment used as a basis for a proper plant design.Comparing the qualitative and quantitative results of the two methods applied to the same case study allows discussing their effectiveness in supporting the risk based decision-making.