Applied Industrial Alarm Analytics for Safer and Smarter Process Operations

This project aims to improve alarm systems in industrial facilities to help plant operators respond more effectively during abnormal situations by developing advanced, data-driven solutions. While alarm systems are designed to notify operators of abnormalities or equipment failures, in practice, operators often face an overwhelming number of alarms, many of which are unnecessary or repetitive. This alarm overload can delay critical responses and increase the risk of safety incidents and operational disruptions. As industrial organizations place greater emphasis on sustainable Environmental, Health, and Safety (EH&S) performance and corporate responsibility, improving alarm systems becomes essential. Achieving EH&S goals can lead to substantial economic benefits by preventing costly shutdowns, equipment damage, and process failures. Building on the progress made during the first phase of this research, the current phase will continue to analyze alarm and process data to: (1) improve alarm configurations, (2) identify the causes of alarm overload, and (3) develop intelligent tools and techniques to reduce nuisance alarms and support operator decision-making. The partner organization, Parkland Refinery, will benefit from enhanced safety, fewer unplanned interruptions, and improved environmental compliance. More broadly, the project supports Canadian industry by training skilled personnel and contributing to safer, more sustainable operations.

Faculty Supervisor:

Tongwen Chen

Student:

Partner:

Parkland Burnaby Refinery

Discipline:

Engineering

Sector:

Manufacturing

University:

University of Alberta

Program: