Zaili Yang

Fuzzy Rule-Based Bayesian Reasoning Approach for Prioritization of Failures in FMEA

This paper presents a novel, efficient fuzzy rule-based Bayesian reasoning (FuRBaR) approach for prioritizing failures in failure mode and effects analysis (FMEA). The technique is specifically intended to deal with some of the drawbacks concerning the use of conventional fuzzy logic (i.e. rule-based) methods in FMEA. In the proposed approach, subjective belief degrees are assigned to the consequent part of the rules to model the incompleteness encountered in establishing the knowledge base. A Bayesian reasoning mechanism is then used to aggregate all relevant rules for assessing and prioritizing potential failure modes. A series of case studies of collision risk between a floating, production, storage, and off loading (FPSO) system and a shuttle tanker caused by technical failure during tandem off loading operation is used to illustrate the application of the proposed model. The reliability of the new approach is tested by using a benchmarking technique (with a well-established fuzzy rule-based evidential reasoning method), and a sensitivity analysis of failure priority values.

The use of Bayesian network modelling for maintenance planning in a manufacturing industry

This paper has been written in order to apply Bayesian network modelling to a maintenance and inspection department. The primary aim of this paper is to establish and model the various parameters responsible for the failure rate of a system, using Bayesian network modelling, in order to apply it to a delay-time analysis study. The use of Bayesian network modelling allows certain influencing events to be considered which can affect parameters relating to the failure rate of a system. Bayesian network modelling also allows these influencing events to change and update depending on the influencing data available at any given time, thus changing the failure rate or probability of failure. A methodology has been developed and applied to a case study in order to demonstrate the process involved.

Incorporation of formal safety assessment and Bayesian network in navigational risk estimation of the Yangtze River

Formal safety assessment (FSA), as a structured and systematic risk evaluation methodology, has been increasingly and broadly used in the shipping industry around the world. Concerns have been raised as to navigational safety of the Yangtze River, Chinas largest and the worlds busiest inland waterway. Over the last few decades, the throughput of ships in the Yangtze River has increased rapidly due to the national development of the Middle and Western parts of China. Accidents such as collisions, groundings, contacts, oil-spills and fires occur repeatedly, often causing serious consequences. In order to improve the navigational safety in the Yangtze River, this paper estimates the navigational risk of the Yangtze River using the FSA concept and a Bayesian network (BN) technique. The navigational risk model is established by considering both probability and consequences of accidents with respect to a risk matrix method, followed by a scenario analysis to demonstrate the application of the proposed model.

Use of fuzzy evidential reasoning in maritime security assessment.

Over the last few years, there has been a growing international recognition that the security performance of the maritime industry needs to be reviewed on an urgent basis. A large number of optional maritime security control measures have been proposed through various regulations and publications in the post-9/11 era. There is a strong need for a sound and generic methodology, which is capable of taking into account multiple selection criteria such as the cost effectiveness of the measures based on reasonable security assessment. The use of traditional risk assessment and decision-making approaches to deal with potential terrorism threats in a maritime security area reveals two major challenges. They are lack of capability of analyzing security in situations of high-level uncertainty and lack of capability of processing diverse data in a utility form suitable as input to a risk inference mechanism. To deal with such difficulties, this article proposes a subjective security-based assessment and management framework using fuzzy evidential reasoning (ER) approaches. Consequently, the framework can be used to assemble and process subjective risk assessment information on different aspects of a maritime transport system from multiple experts in a systematic way. Outputs of this model can also provide decisionmakers with a transparent tool to evaluate maritime security policy options for a specific scenario in a cost-effective manner.

Bayesian network with quantitative input for maritime risk analysis

This article presents an innovative approach towards integrating logistic regression and Bayesian networks (BNs) into maritime risk assessment. The approach has been developed and applied to a case study in the maritime industry, but has the potential for being adapted to other industries. Various applications of BNs as a modelling tool in maritime risk analysis have been widely seen in relevant literature. However, a common criticism of the Bayesian approach is that it requires too much information in the form of prior probabilities, and that such information is often difficult, if not impossible, to obtain in risk assessment. The traditional and common way to estimate prior probability of an accident is to use expert estimation (inputs) as a measure of uncertainty in risk analysis. In order to address the inherited problems associated with subjective probability (expert estimation), this study develops a binary logistic regression method of providing input for a BN, making use of different maritime accident data resources. Relevant risk assessment results have been achieved by measuring the safety levels of different types of vessels in different situations.

Maritime safety analysis in retrospect

Maritime safety has undergone considerable change in the past decades, particularly inits improved approach to risk quantification analysis. This article reviews the challenges of maritime safety analysis and the different approaches used to quantify the risks in maritime transportation. Formal safety assessment (FSA) is examined with a focus on its deficiencies, and its recent developments are described at the International Maritime Organization (IMO) level. The possible applications of FSA in maritime security and piracy analysis are discussed given its growing impacts on safety at sea. Some new uncertainty and risk modelling techniques are also presented to demonstrate how risk quantification analysis facilitates the transformation of maritime safety culture from a reactive prescriptive scheme towards a proactive goal-setting regime.

Modelling port choice in an uncertain environment

Port choice is an important issue to be investigated to ensure the effective integration of container supply chains and the sustainable development of regional economy. The selection of appropriate ports to facilitate shipping activities and international trade is crucial for many stakeholders, including shipping lines, port administrators, cargo shippers and national governments. The task is essentially a process of multiple criterion decision-making (MCDM) under uncertainty, requiring analysts to derive rational decisions from uncertain and incomplete data related to different quantitative and qualitative determinants. This paper aims at proposing a new conceptual port choice method by explaining the role fuzzy logic in evidential reasoning in a complementary way, in which various forms of raw data (either objective or subjective) collected to evaluate port performance can first be converted into and presented as fuzzy grades defined using linguistics terms with degrees of belief (DoBs) and second be combined using evidential reasoning to produce a port choice preference score. The method is applied to analyse the selection of major Northeast Asian (NEA) container ports from a shipping line’s perspective. The outcome, a port choice preference score, is calculated using evidential reasoning to directly synthesize the true estimation of the port with respect to each criterion and therefore, unlike a relative ranking index, keeps the ‘goodness’ of port evaluation, capable of benchmarking a specific port’s performance and monitoring the increase of its competitiveness in a longitude study with respect to an individual criterion or all the criteria as a whole.

Approximate TOPSIS for vessel selection under uncertain environment

The selection of appropriate vessels to carry out shipping activities is crucial for many maritime stakeholders including charterers, shipowners, brokers, surveyors and safety engineers. The task is essentially a process of multiple criteria decision making (MCDM) under uncertainty requiring analysts to derive rational decisions from ambiguous and incomplete data contained in different quantitative and qualitative forms. Fuzzy Techniques for Order Preference by Similarity to an Ideal Solution (TOPSIS) have been well documented in the literature and commonly used in the process of group decision-making under fuzzy environment. While showing the attractiveness in dealing with ambiguous estimates, they have been criticised to be incapable of modelling incompleteness encountered in decision analysis. This paper therefore uses the concept of degrees of belief to develop a novel approximate interval TOPSIS approach for overcoming some of the drawbacks of classical fuzzy TOPSIS methods and facilitating the development of reliable vessel selection models under uncertain environment.

A novel technique for evaluating and selecting logistics service providers based on the logistics resource view

Modelling logistics outsourcing process using logistics resources under uncertainty.Using FDEMATEL to analyse the logistics resources impact relationships and weights.Identifying independent logistics resources and evaluate LSPs using FTOPSIS.Testing the model effectiveness through a developing-economy real case study.Confirming the model robustness through a two-phase sensitivity analysis. The increasing importance of logistics outsourcing and availability of logistics services providers (LSPs) highlights the significance and complexity of the LSP evaluation and selection process. Most existing LSP evaluation and selection studies use historical performance data and assume independence among decision criteria. This paper proposes an integrated logistics outsourcing approach to evaluate and select LSPs based on their logistics resources and capabilities. This novel approach combines a fuzzy decision making trial, evaluation laboratory (FDEMATEL) and fuzzy techniques to order preferences by similarity to ideal solution (FTOPSIS) methods. The new multi-criteria decision making (MCDM) model addresses the impact relationships between decision criteria and ranks LSP alternatives against weighted resources and capabilities. The effectiveness of this approach is demonstrated through a real case study and a two-phase sensitivity analysis confirms its robustness.

Facilitating uncertainty treatment in the risk assessment of container supply chains

Over the last few years there has been a growing international recognition that Container Supply Chains (CSCs) contribute to economic prosperity. But they are uniquely vulnerable to many risks caused by both the traditional hazards, such as physical breaches in the integrity of shipments and the newly rising threats associated with pirate and terrorist attacks. To allow better understanding and control of the risks, it is necessary for the stakeholders to proactively assess the chains’ security and safety in advance, or reactively discover risks after a detrimental event occurs. This paper explores the various CSC risks, identifies common themes, and deals with the corresponding uncertainties by developing two novel risk modelling methods. One is to develop a fuzzy evidential reasoning approach for carrying out the security estimation of a vulnerable port system against terrorism attacks and the other is to produce a Bayesian network decision support tool for identifying vulnerable assets in a port security protection scenario. Consequently, the methods can be used to assemble and process subjective risk information on different aspects of a container transport system from multiple experts in a systematic way. Outcomes of the models can also provide decision makers with a transparent tool to evaluate CSC safety and security policy options for a specific scenario in a cost-effective manner.