Probabilistic Risk Modeling for Performance Deviation in Multinational Construction Projects in Bali

Abstract

Multinational construction projects in tourism regions are exposed to complex uncertainty arising from cultural differences, fragmented organizational systems, differing technical standards, and cross-border stakeholder coordination. This research examines how performance deviations may occur in multinational construction projects in Bali, Indonesia, by applying a probabilistic risk-based approach. The model was built incrementally, beginning with the identification of risks from earlier studies and expert input. The risks were then assessed through questionnaire surveys and a risk matrix before being modeled using triangular probability distributions and Monte Carlo simulation. The final stage employed sensitivity analysis to identify the most influential risks and to support the development of mitigation strategies. This research was carried out in the context of accommodation and tourism-residential projects in the Sarbagita area of Bali, where foreign owners, consultants, designers, or specialists frequently interact with local contractors. The dominant risks were selected by considering both their frequency and impact, and were then expressed as minimum, most likely, and maximum impact values for the time, cost, and quality models. A 10,000-trial Monte Carlo simulation indicated that the simultaneous occurrence of dominant risks could generate average deviations of 8.37% in time, 15.90% in cost, and 7.88% in quality. Sensitivity analysis revealed that the effectiveness of multinational team communication was the most influential factor for time deviation, whereas technical and project information readiness was the most influential factor for cost and quality deviation. This study contributes a structured decision-support approach for translating qualitative multinational construction risks into probabilistic performance estimates and practical mitigation priorities.