Customs Organizations are responsible for two Opposing In g yet equally importantrnresponsibilities . These are t e provision of efficient services to traders for thernsmooth flow of shipments and the protection of the country from any kind of riskrnthreats that is associated with interactional trade.rnReform and monetization of custom services through automation and settingrntransparent working procedures has resulted in the provision of efficient services .rnHowever addressing the issue of implementing a proper risk management strategyrnremain s a challengernThe Ethiopian Customs Authority at present is handling the control of customs risksrnusing subjective methods. The subjective method of handling risks solely dependsrnon experts' judgment of selecting shipments for physical examination. ThernSubjective method is essential since the knowledge and experience of customsrnexperts and their observation of the behaviors of intervening agents like traders andrnclearing agents is very important. However depending only on subjective analysisrnfor strategic risk management ha s its shortcomings.rnThis study was aimed at supporting the current selective physical examinationrnsystem o f incoming shipments in the Ethiopian Customs Authority with objectivernmethods using data mining. The study was conducted through the annals is ofrncustoms fraud cases seized in the past. For this stud y one of the data miningrntechniques known as decision tree was employed. The dataset used in the stud yrnconsisted of 10 364 record s out of which 17 0 cases were fraud cases.rnThe distribution of the two classes was highly imbalanced. To deal with the classrnimbalance pro blend the over- sampling approach was use d. Five experiments we rernconducted by varying the rate of over-sampling. After over- sampling the fiverndatasets had a ratio of 90: 10, 80:20, 70:30, 60:40 and 50 :50 all non- fraud to fraud.rnrnUsing an independent dataset that also contain 2616 non-fraud and 39 fraud casesrnall the models generated by the five data sets were validated. The dataset with arnproportion of 70:30 has shown the best result in terms of correctly classifying thernfraud cases. The model correctly classified 22 fraud cases out of the 39 cases.rnCombining the subjective method with the subjective methods can improve thernefficiency of risk assessment and selective physical examination of shipments in thernEthiopian customs authority. Models developed by automatic analysis can also bernused across all the different customs offices in the country consistently.