Enhancing safety and risk management in biodiesel production: a machine learning approach using k-means clustering to address operational challenges and standardization issues

Abstract

Industrial-scale biodiesel production plays a critical role in advancing sustainable energy but involves significant operational risks due to complex processes and hazardous materials. Traditional risk assessments often fall short in addressing these specific challenges comprehensively. The study conducted a detailed risk assessment for industrial-scale biodiesel production using Bow Tie analysis to identify hazards, causal factors, and mitigation strategies. Bow Tie analysis effectively mapped top events such as equipment failure and fire incidents, tracing their root causes and potential consequences across key stages like feedstock handling and chemical processing. Integration of preventive and mitigated barriers provided a clear framework for risk containment. The priority weightage analysis identified uncertainty Level (0.5) as the most variable factor, followed by Risk Score (0.25), Incident Probability (0.16), and Safety Alert Frequency (0.09). Correlation analysis shows strong negative correlations between safety outcomes and factors like process automation ( r = -0.98658), regulatory compliance ( r = -0.95542), and safety policy implementation ( r = -0.97946). Dataset 38 shows optimal safety performance with the highest automation level (84), top compliance (0.93), lowest risk score (13), low uncertainty (5 %), and minimal incident probability (0.02), making it the most efficient and secure configuration. This integrated approach supports the sustainable growth and resilience of biodiesel production facilities.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 15 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.