System Level Reliability Modelling and Analysis of a Container Terminal
E. Çoksayar (TU Delft - Mechanical Engineering)
A. Napoleone – Mentor (TU Delft - Transport Engineering and Logistics)
R. Leite Patrão – Mentor (TU Delft - Transport Engineering and Logistics)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
This thesis presents a system-level reliability modelling framework for analyzing how failures, maintenance activities, and operational disruptions spread through interconnected subsystems in container terminal operations. The research combines reliability analysis with probabilistic reasoning to provide a clear understanding of how technical, human, and environmental factors affect delays and system performance. This connection is established by translating reliability states—such as equipment health, availability, and maintenance effectiveness—into probabilistic delay outcomes, allowing the model to quantify how reliability losses lead to operational delays.
The framework integrates Fault Tree Analysis (FTA) and Bayesian Networks (BNs) to capture both causal and probabilistic relationships within the terminal system. FTA decomposes the top event— total operational delay—into its contributing factors, providing the structural foundation for the BN. The initial BN structure captures subsystem interactions through nodes representing equipment failure, availability, efficiency, and environmental conditions. Each node describes a key aspect of terminal performance. Equipment failure indicates the operational state of critical assets and is modelled with a Weibull distribution. Availability represents the percentage of time each subsystem could work. Efficiency reflects performance under different external or internal constraints. Environmental factors, yard storage fullness, and terminal busyness acted as external drivers that influenced subsystem efficiency and delay propagation.
The BN model was further expanded to include maintenance and operator availability as new influencing factors. Both preventive and corrective maintenance were considered within the model. The maintenance effectiveness was determined using Weibull-based reliability parameters. The scale and shape parameters were adjusted through maintenance effect multipliers to account for imperfect repair conditions. This method enabled the BN to show how preventive maintenance improves equipment availability. Operator availability was also considered to account for human-related variability, showing how workforce presence impacts subsystem operability and overall delays.
The analysis of the BN used forward inference and sensitivity testing. The results indicate that disruptions in one subsystem can spread through the terminal, causing cumulative delays and underscoring the strong interconnections between quay cranes, yard cranes, and horizontal transport. The maintenance analysis revealed that by implementing preventive maintenance strategies, equipment availability could be increased. Operator availability affected total delay mainly during severe disruptions like strikes, while daily variations had a limited impact.
This developed framework combines reliability modelling and operational performance analysis in one structure. It offers a clear way to study how maintenance, operator availability, and environmental conditions influence reliability and delay in terminal operations. While the framework was created for container terminals, the same approach could be adapted to other connected systems where equipment, people, and external conditions interact to affect overall performance.