A REVIEW ON THE INFLUENCE OF AI-ENABLED FIRE DETECTION AND SUPPRESSION SYSTEMS IN ENHANCING BUILDING SAFETY

Abstract

This study addressed the problem that conventional fire detection and rule-based panels in buildings can generate nuisance alarms and delayed verification, weakening coordinated response and safety assurance. The purpose was to quantify how AI-enabled fire detection and suppression capabilities predict perceived Building Safety Enhancement (BSE) using a quantitative, cross-sectional, case-study based survey design. A purposive sample of 210 stakeholders from building cases (62.4% commercial or mixed-use and 37.6% institutional or industrial) reported AI detection dashboards in 71.0% of contexts and suppression decision support in 54.8% across routine monitoring, drills, and incident response. Key variables were AI-Enabled Detection Capability (ADC), AI-Enabled Suppression and Control Effectiveness (ASCE), Integration and Real-Time Monitoring (IRM), and Predictive Maintenance and Fault Diagnosis (PMFD), with BSE as the outcome. Analyses used descriptive statistics, Cronbach’s alpha, Pearson correlations, and multiple regression. Results reported as a worked example showed high construct means (ADC M = 4.02, ASCE M = 3.88, IRM M = 3.95, PMFD M = 3.76, BSE M = 3.97) and strong reliability (alpha = 0.84 to 0.90). All predictors correlated positively with BSE (r = 0.48 to 0.62, p < .01). The regression model was significant (F(4,205) = 65.20, p < .001) and explained 56% of BSE variance (R squared = 0.56), with ADC (beta = 0.31) and IRM (beta = 0.27) strongest, followed by ASCE (beta = 0.19) and PMFD (beta = 0.14). Implications suggest prioritizing detection credibility and real-time integration, then strengthening suppression coordination and predictive maintenance to sustain readiness and reduce alarm fatigue.