| dc.description.abstract | The use of subsidized fuel oil (BBM) requires proper planning in all sectors, especially in the railway sector as one of the modes of transportation that is in great demand by the people of Indonesia. Planning for JBT fuel needs is one of the problems faced by PT. KAI. In the process of planning fuel requirements, PT. KAI does not have a definite method and it is found that there is a large enough difference between the proposed data on the realization of BBM on the Java Daop train every year, so the purpose of this research is to predict the method of forecasting the need for fuel oil (BBM) for trains. the appropriate Java Operational Area (Daop) for the January-March 2020 period and analyze the comparison of proposals, quotas, and realization of rail fuel in the Java Operational Area (Daop) in 2016-2019. The method used in this research is the regression method (linear, quadratic, exponential, cyclical, cyclical trend) and the single exponential smoothing method. The results obtained are that the kuadratis regression method is the best forecasting method based on the MAPE (Mean Absolute Percentage Error) value of 4.25 %, the Mean Absolute Deviation (MAD) error of 1.35 % and the Mean Squared Error (MSE) error of 2.81 %. with the determination of the fuel allocation of 41.497.999 Liters for the period January-March 2020 with an error difference of 3.67 % with the realization data of 38,640,712 Liters. strategic plan in an effort to increase the volume of passengers and freight transport. | en_US |
