Risk Assessment and Management Tool Use, Implementation Quality and Perceived Effectiveness During Shutdown and Turnaround Maintenance: Evidence from a Nigerian Upstream Petroleum Operator

Abstract

Risk assessment and management (RAM) tools are central to shutdown and turnaround maintenance (STAM), but frequent use does not necessarily demonstrate implementation quality or safety effectiveness. This cross-sectional study analysed respondent-level survey data from 148 STAM professionals in a Nigerian upstream petroleum organisation. Descriptive statistics, Spearman correlations, exploratory factor analysis (EFA), robust multiple regression and Mann-Whitney U incident-group comparisons were applied. JSA (M = 4.53) and toolbox talks (M = 4.35) were used more frequently than HAZOP (M = 2.30) and risk registers (M = 2.27). EFA showed only marginal sampling adequacy (KMO = .610; Bartlett's chi-square(10) = 54.70, p < .001) and a weak two-factor structure, confirming that C1-C5 should not be treated as a strongly unidimensional effectiveness scale. The regression model explained little variance in the composite score (R2 = .031; adjusted R2 = -.018; robust F = 0.62, p = .741), and no predictor was significant. Incident-experienced respondents reported modestly higher JSA use, incident-prevention perceptions and field-implementation perceptions, but the composite effectiveness difference was non-significant. The findings support a compliance-effectiveness paradox: high procedural activity can coexist with weak or inconsistent signals of substantive effectiveness. RAM assurance should therefore combine activity frequency with analytical quality, barrier health, control verification and action closure.