With severe market illiquidity in global financial markets and possible overcapacity in shipping, evaluation of shipping financial performance becomes an important issue for the companies to ensure their sustainability. The shipping cycle inevitably experienced a down term in tanker and bulk shipping industries. As a result, among 27 tanker shipping firms included in the Bloomberg tanker index, seven of them have filed bankruptcy in recent years. Given that shipping markets are highly competitive in nature, and shipping business features risks and challenges in a constantly changing environment, it will be necessary and beneficial to investigate financial risk management in the shipping segment. It is however well-noted that although analysing financial performance and predicting bankruptcy risk have a long tradition in economics, studies focusing on tanker shipping failure prediction are rarely to be found in the literature. This study aims to develop an assessment methodology using a Bayesian network (BN) to predict the failure probability of oil tanker shipping firms. To provide a foundation for building the BN model, we focus on the development of a qualitative BN using the casual factors which are identified using correlation analysis in this study. Of all, 27 publicly traded tanker shipping companies in the Bloomberg Tanker Shipping Index are investigated for the period from 2000 to 2010. Shipping companies’ performance and the likelihood of filing bankruptcy are influenced by many factors, including the global business cycle, demand for crude oil transport, supply of oil tankers fleet, shift of the trade routes and cost of tanker transport service at a macro level and sales performance, ownership structure, risk management, bank financing and merger and acquisition (M&A) at a micro/firm level. The dependency among the key factors and the relationship between factors and financial performance can be simulated using qualitative diagram in BN, while the quantitative configuration of such dependency (i.e. conditional probabilities) can be obtained using statistical regression analysis based on historical data and through a correlation analysis. It proposes a new framework capable of incorporating quantitative statistical measurement with BN in shipping financial risk estimation and prediction.