Researchers Develop Framework to Merge AI and Human Intelligence for Process Safety

Researchers have developed a framework that combines artificial intelligence (AI) and human intelligence to improve process safety.
Artificial intelligence has greatly improved efficiency in various industries, but its integration with human intelligence is crucial for ensuring safe operations. The researchers’ goal is to develop a comprehensive framework that integrates AI and Human Intelligence (HI) into process safety systems, enhancing both safety and efficiency.

The research highlights the importance of combining AI and HI rather than replacing humans. Dr. Faisal Khan, a professor at Texas A&M University, believes that AI can analyze real-time data, predict maintenance needs, and automatically detect faults, improving safety outcomes. However, human decision-making is also essential for reducing incident rates, lowering operational costs, and increasing reliability.

The framework identifies several risks associated with implementing AI and HI in process industries, including data quality issues, overreliance on AI, lack of contextual understanding, model misinterpretation, training and adaptation challenges, and human error in feedback. The researchers are interested in exploring the synergy between AI and HI to meet the evolving demands of industrial safety.

The collaboration between AI and HI is seen as essential for advancing process safety. Dr. Khan emphasizes that developing reliable, trustworthy, and safe AI systems tailored to industrial applications is crucial. The framework aims to provide a clear understanding of the potential and limitations of AI, propose strategies for its effective implementation, and minimize risks and improve safety outcomes.

The research was published in AIChE Journal (2024) under the title “Process safety 4.0: Artificial intelligence or intelligence augmentation for safer process operation?” DOI: 10.1002/aic.18475.

+Source: https://techxplore.com/news/2024-07-framework-merge-ai-human-intelligence.html