THESIS ABSTRACT

Abstract

Hydropower Construction in Nepal is inherently high-risk due to tough terrain, remote sites, and poor safety infrastructure. Despite the existence of national occupational safety norms and contractual safety clauses, the majority of existing safety evaluations in Nepal's hydropower sector are qualitative, reactive, and rely primarily on compliance checklists and historical accident records. As a result, safety experts and site managers frequently rely on subjective judgement, personal intuition, and inconsistent hazard prioritisation, resulting in inefficient allocation of limited safety resources, recurring patterns of injuries and near-miss events, and a persistent inability to benchmark or systematically improve safety performance across projects. This research develops an integrated Fuzzy AHP and Fuzzy TOPSIS methodology to systematically assess and rank safety conditions across multiple hydropower construction worksites in Nepal. The approach is specifically developed to deal with the ambiguity, linguistic uncertainty, and expert subjectivity that are inherent in construction safety assessments, transforming qualitative expert judgements into measurable, reproducible weights and ranks. Five main safety criteria were identified: Management & Organizational Safety (C1), Worker Training & Behavioral Safety (C2), Worksite & Environmental Conditions (C3), Equipment & Material Safety (C4), and Emergency Preparedness & Safety Systems (C5). Fuzzy AHP was first used to calculate the relative weights of these criteria based on expert pairwise comparisons, resulting in the following weights: C1 = 0.4011 (highest), C2 = 0.1789, C3 = 0.1645, C4 = 0.1457, and C5 = 0.1095 (lowest), which indicate proactive measures are more valued than reactive measures. At the sub-criteria level, the three most influential factors globally are safety policy and management commitment (SC1) at 0.1632, safety supervision and monitoring (SC3) at 0.1259, and safety planning and risk assessment (SC2) at 0.1118, confirming that visible leadership commitment, competent field supervision, and systematic pre-task hazard analysis are the most powerful levers for improving workplace safety. Fuzzy TOPSIS was then used to calculate closeness coefficients (CCi) for four hydropower projects, resulting in the following ranking: Project 3 had the highest cumulative score (CC = 0.7123), 2 followed by Project 1 (CC = 0.5709), Project 2 (CC = 0.5467), and Project 4 (CC = 0.4565). Although all four projects maintain adequate overall safety conditions, the comparison analysis indicates a clear relative ranking, with Project‑3 having the highest safety performance and Project‑4 the lowest among the four. The results show that management and organisational safety have the greatest influence on worksite safety, while emergency preparedness has the least weight in present practices. The hybrid Fuzzy AHP-Fuzzy TOPSIS framework successfully manages uncertainty and subjectivity, providing a useful, reproducible tool for comparative safety assessment in Nepal's hydropower.

Keywords

Safety assessment, Criteria, Sub-criteria, Weightage, Fuzzy AHP, Fuzzy Topsis Closeness Coefficient