Breadcrumb
- Home
- Publications
- Proceedings
- 2025 Global Conference on Process Safety and Big Data
- General Program
- AI Applications in HAZOP
- "Leveraging Big Data from PHAs and API 556 to Enhance Risk Evaluation in Furnaces"
Introduction
The digitalization of HAZOP studies across our production facility has enabled a systematic evaluation of risk assessment criteria applied to various units, with a particular focus on more than 50 process furnaces. This volume of data supports the application of big data analytics to identify patterns and inconsistencies.
Our analysis focused on key safety aspects, including flameout scenario evaluations, treatment of “other composition” factors, instrumentation coverage, and the frequency of identified explosion scenarios.
Preliminary findings revealed significant variability in assessment approaches and a lack of standardization across units. To establish a benchmark, the results were compared against API 556 — Instrumentation, Control, and Protective Systems for Gas Fired Heaters — highlighting gaps in threat identification, hazard evaluation, and protective instrumentation.
This study underscores the potential of digital HAZOP data to drive standardization and improve furnace safety performance through data-driven insights.
Methodology
Use of existing digitalized HAZOPs as input for the study: Previously conducted HAZOP studies from various process units were collected and digitally structured. This digital database served as the starting point for the analysis, enabling a systematic and traceable evaluation of applied risk criteria.
Filtering of risk evaluations and scenarios applicable to process furnaces: The digital dataset was refined to isolate only those risk evaluations and scenarios relevant to process furnaces.
Statistical analysis of typical risk scenarios in process furnaces: A statistical analysis was conducted focusing on the most commonly evaluated risk events in process furnaces, such as flameout, liquid ingress, burner overpressure, and underpressure.
Comparison with risk scenarios suggested by API 556: The analysis results were compared with the risk scenarios outlined in API 556, identifying gaps and opportunities for improvement.
Development of a standardized template for HAZOP analysis of furnaces: A tool was developed that incorporates both regulatory guidelines and insights derived from the site's historical data.
Results As a result of the study, a standardized template was developed for the HAZOP analysis of furnaces. This tool enables the systematization of risk evaluations, improves traceability, and facilitates comparison across units. Additionally, it can be adopted as a standard in the company’s internal procedures for furnace risk assessments.
Key achievements of the work include:
Standardization of threats for deviations typically evaluated in process furnaces.
Clarification of the relationship between deviations and consequences.
Identification of typical safeguards.
Assignment of consequences with accurate orders of magnitude.
Reduction of criteria disparity across units.
Conclusions
The use of digitalized data from HAZOP studies has proven to be a powerful tool for improving the quality, consistency, and usefulness of risk analyses in process furnaces. The achieved standardization not only enhances operational safety but also facilitates data-driven decision-making and compliance with international standards such as API 556. The developed template represents a significant step toward a more robust and consistent safety culture within the organization.