Fabrizio Ruggeri

A Bayesian Belief Network modelling of organisational factors in risk analysis : A case study in maritime transportation

The paper presents an innovative approach to integrate Human and Organisational Factors (HOF) into risk analysis. The approach has been developed and applied to a case study in the maritime industry, but it can also be utilised in other sectors. A Bayesian Belief Network (BBN) has been developed to model the Maritime Transport System (MTS), by taking into account its different actors (i.e., ship-owner, shipyard, port and regulator) and their mutual influences. The latter have been modelled by means of a set of dependent variables whose combinations express the relevant functions performed by each actor. The BBN model of the MTS has been used in a case study for the quantification of HOF in the risk analysis carried out at the preliminary design stage of High Speed Craft (HSC). The study has focused on a collision in open sea hazard carried out by means of an original method of integration of a Fault Tree Analysis (FTA) of technical elements with a BBN model of the influences of organisational functions and regulations, as suggested by the International Maritime Organisations (IMO) Guidelines for Formal Safety Assessment (FSA). The approach has allowed the identification of probabilistic correlations between the basic events of a collision accident and the BBN model of the operational and organisational conditions. The linkage can be exploited in different ways, especially to support identification and evaluation of risk control options also at the organisational level. Conditional probabilities for the BBN have been estimated by means of experts’ judgments, collected from an international panel of different European countries. Finally, a sensitivity analysis has been carried out over the model to identify configurations of the MTS leading to a significant reduction of accident probability during the operation of the HSC.

Robust Bayesian analysis

We provide an overview of robust Bayesian analysis with emphasis on foundational, decision oriented and computational approaches. Common types of robustness analyses are described, including global and local sensitivity analysis and loss and likelihood robustness.

Mixtures of Gamma Distributions With Applications

This article proposes a Bayesian density estimation method based upon mixtures of gamma distributions. It considers both the cases of known mixture size, using a Gibbs sampling scheme with a Metropolis step, and unknown mixture size, using a reversible jump technique that allows us to move from one mixture size to another. We illustrate our methods using a number of simulated datasets, generated from distributions covering a wide range of cases: single distributions, mixtures of distributions with equal means and different variances, mixtures of distributions with different means and small variances and, finally, a distribution contaminated by low-weighted distributions with different means and equal, small variances. An application to estimation of some quantities for a M/G/1 queue is given, using real E-mail data from CNR-IAMI.

Using AHP in determining the prior distributions on gas pipeline failures in a robust Bayesian approach

Abstract The paper proposes a robust Bayesian approach to support the replacement policy of low-pressure cast-iron pipelines used in metropolitan gas distribution networks by the assessment of their probability of failure. In this respect, after the identification of the factors leading to failure, the main problem is the historical data on failures, which is generally limited and incomplete, and often collected for other purposes. Consequently, effective evaluation of the probability of failure must be based on the integration of historical data and knowledge of company experts. The Analytic Hierarchy Process has been used as elicitation method of expert opinion to determine the a priori distribution of gas pipeline failures. A real world case study is presented in which the company expertise has been elicited by an ad hoc questionnaire and combined with the historical data by means of Bayesian inference. The robustness of the proposed methodology has also been tested.

e-Participation and Decision Analysis

Decision analytic methods are being increasingly used to help to articulate and structure debate and deliberations among citizens and stakeholders in societal decisions. Methods vary, but, essentially, a public authority or agency, when faced with a significant set of issues, may organise one or more workshops with stakeholders and citizens as participants. Such methods of public engagement and participation are, by and large, conducted face to face. However, the advent of the World Wide Web brings the possibility of conducting citizen and stakeholder interactions in a distributed, possibly asynchronous fashion. In this paper we discuss the challenges that have to be addressed and overcome if such e-participation is to be a valid tool within a modern democracy. The difficulties are many and varied, but the pressures towards e-government, and better regulation in general, mean that such methods will be used in the near future. Thus, we outline a program of research and debate in which we believe that the professional decision analysis community should engage.

Bayesian analysis of M/Er/1 and M/H_k/1 queues

This paper describes Bayesian inference and prediction for some M/G/1 queueing models. Cases when the service distribution is Erlang, hyperexponential and hyperexponential with a random number of components are considered. Monte Carlo and Markov chain Monte Carlo methods are used for estimation of quantities of interest assuming the queue is in equilibrium.

A Bayesian Approach to Improve Estimate at Completion in Earned Value Management

Forecasting represents a core project management process. Estimates at completion in terms of cost and schedule provide essential data and advice to the project team in order to lead and control the project and implement suitable corrective measures. In order to improve the forecasting process, a Bayesian model has been developed within the earned value management framework aiming to calculate a confidence interval for the estimates of both cost and schedule at the completion of the project. The model is based on the integration of data records and qualitative knowledge provided by experts. The model has been tested in an oil and gas project.

A Bayesian hidden Markov model for imperfect debugging

In this paper we present a new model to describe software failures from a debugging process. Our model allows for the imperfect debugging scenario by considering potential introduction of new bugs to the software during the development phase. Since the introduction of bugs is an unobservable process, latent variables are introduced to incorporate this property via a hidden Markov model. We develop a Bayesian analysis of the model and discuss its extensions. We also consider how to infer the unknown number of states of the hidden Markov model. The model and the Bayesian analysis are implemented to actual software failure data.

Bayesian Forecasting for Accident Proneness Evaluation

As part of their resource allocation processes, insurance companies have to undertake various evaluation tasks concerning the accident proneness of their insurants. Bayesian methods are specially fit for that task since they allow for the coherent incorporation of all sources of information, including expert opinions and data. We describe three increasingly complex and realistic models for that purpose. For predictive and inference purposes, we have to rely on simulation methods. We illustrate the models with a real case and describe their implementation in a forecasting system developed for an insurance company.

CONTROLLING THE USABILITY OF WEB SERVICES

Service Oriented Architectures (SOA) enable dynamic integration of Web Services (WS) to accomplish a users need. As such, they are sensitive to user errors. This article presents a framework for mitigating the risks of user errors due to changes in the service delivery context. The underlying methodology incorporates usability in the design, testing, deployment and operation of dynamic collaborative WS, so that the error-prone elements of the User Interface (UI) are identified and eliminated. The methodology incorporates Statistical Process Control (SPC) of Web Service Indices (WSI), obtained by a Decision Support system for User Interface Design (DSUID), in which the users are elements of the control loop.