Deep ensemble–based probabilistic modeling of torsional capacity in reinforced concrete beams with uncertainty decomposition

Key Points

• Torsional capacity is critical for assessing structural integrity under load, especially in reinforced concrete beams.
• Findings show improvements in predicting torsional capacity variations with uncertainty decomposition techniques.
• Analysis using deep ensemble–based probabilistic modeling enhances prediction accuracy for structural engineers.
• These methods can inform future designs, underscoring the need for reliable assessments in construction projects.