基于组合模型的海上交通事故预测研究

A Study of Maritime Traffic Accidents Prediction Based on Combination Model

海上交通安全问题一直是海上交通的重要研究课题,而海上交通事故预测能为海上交通安全提供理论依据,做出相应预警防范。运用弱化算子改进传统灰色GM(1,1)模型对福建辖区未来三年海域海上交通事故预测,并与传统灰色预测对比。结果表明:改进后的灰色模型不仅拟合度高,并且预测模型误差从原来的二级提升到一级,平均相对误差由10.82%降低为4.69%,能够很好地反应福建辖区海上交通事故未来发展趋势。

Maritime traffic safety has always been an important topic of maritime traffic research.Maritime traffic accident prediction can provide theoretical basis for making corresponding early warning and prevention and safeguarding maritime traffic safety.The weakening operator is used to improve the traditional grey GM(1,1)model to predict maritime traffic accidents in Fujian in the next three years,and compared with the traditional grey prediction.The results show that the improved grey model not only has a high fitting degree,but also improves the prediction model error rate from the original second level to the first level,and the average relative error rate decreases from 10.82% to 4.69%,which can well reflect the future development trend of marine traffic accidents in Fujian jurisdiction.