ウェザールーティングのための東西指数の予測手法の研究

摘要

ウェザールーティング（WeatherRouting、以下WR と記す）とは、航海中の気象・海象状態をできる限 り正確に予測し、当該船舶の航海性能を考慮して、 ある評価基準に対する最適な航路を選定することで ある。WRにおいて、気象・海象情報は不可欠な要 素であるが、現在のところ気象庁が予測する気象・ 海象データは、最長で192時間（8日）先の予報値 までしか発表されておらず、例えば北太平洋におい て日米間を横断する場合、データが不足するという 問題がある。%Effectiveness of the result of weather routine depends upon the accuracy of forecasting sea conditions. However, the latest weather the forecasting is restricted to 8 days ahead prediction in Pacific Ocean. On the other hand, Dr.Suda emphasized on the importance of the concept of zonal index. To solve this problem, the authors propose the application of seasonal adjustment model and its estimation method by Kalman filter for predicting the evolution of the zonal index. The accuracy of the predicted zonal index is estimated by comparing the predicted data with the data obtained from the weather analysis issued by the Japanese Meteorological Agency (JMA). The best model selected in various model's parameters gives accuracy of 76%. The correlation between the zonal index and wave distribution is also investigated by using weather analysis data of the JMA, and the authors make wave distribution patterns from the predicted zonal index. Finally, the authors describe the efficiency of the proposed predicting method on ship's weather routing, by voyages simulation for a container ship.