Towards an Effective Risk Management Framework for Large-Scale Infrastructure Projects
Year: 2026
Type: article
Abstract: Large-scale infrastructure projects (LSIPs) frequently exceed budgets and schedules despite the accessibility or the adoption of risk management frameworks (RMFs). This study addresses or discusses two gaps: the lack of triangulated evidence on why structurally sound frameworks such as ISO 31000 and PMBOK remain procedural rather than adaptive during delivery, and the absence of a practical, governance-integrated mechanism for strengthening them. An exploratory mixed-methods design combined a structured narrative review of the literature, a survey of infrastructure practitioners in Australia, semi-structured interviews with selected senior practitioners, and analysis of five international megaprojects. Findings indicate that risk management is predominantly compliance-driven: risk identification weakens during delivery, risk allocation performs weakest among the six domains, and digital tool adoption remains limited on a five-point scale where higher values indicate stronger performance. Governance was strongly associated with risk allocation and monitoring with identification, while appetite for reform was high but uncorrelated with operational domains. Internal consistency was acceptable to excellent. The study contributes (1) triangulated empirical evidence of systemic RMF weaknesses in LSIPs and (2) the cadence–ownership–feedback (COF) spine with a record-why-changed (RWC) routine, a governance-integrated mechanism for making existing frameworks adaptive, together with a five-stage implementation roadmap. These propositions are conceptually derived from the triangulated evidence rather than empirically validated, and their validation through Delphi review and structured pilots is identified as future work.
Source: Buildings
Institutions: The University of Sydney, UNSW Sydney
Language: English
Field: Decision Sciences
Domain: Social Sciences
Open Access status: gold
Estimated APC paid: $3,137
OpenAlex ID: W7204909724