Improving the Utilisation of Observations using Artificial Intelligence and Machine Learning

About this presentation

The process of collecting observations in high-risk industrial settings is currently time-consuming, complicated and highly subjective. This creates inconsistent data that can seldom be used to identify patterns and trends across a facility.

In response, Empirisys created Boost. Developed closely with academia and industrial bodies, Boost utilises artificial intelligence and machine learning to improve the observation submission and analysis process.

By analysing the observation description, Boost automatically identifies safety critical barriers and life-saving rules at risk, additionally providing suggested actions and potential consequences. This simplifies the process of submission, while ensuring that high quality data is collected. The increase in accuracy, consistency and completeness enables the utilisation of observations as a leading process safety performance indicator.

This presentation will introduce the research and development of Boost from a process safety and technical perspective, finishing with a demonstration of the tool in action.

Speaker Bio:

Morgan is a highly qualified professional with an MSc in Chemical Engineering from the University of Birmingham (2021). His passion for leveraging data for analysis, optimization, and transformation has driven him to pursue further studies in Applied Data Science and AI at the University of South Wales (MSc, 2024). At Empirisys, Morgan serves as a Senior Data Scientist and is the driving force behind Boost, a project he originated during his own research endeavors. He holds the position of inceptor and technical lead, showcasing his innovative approach and leadership in technical development.