DOI: -

Keywords: single moving average, weighted moving average, exponential smoothing, data-driven, tracking signal

ABSTRACT

Uncertainty in the number of vehicles undergoing periodic testing at the Motor Vehicle Testing Unit of the Department of Transportation in Surakarta City often leads to queues. To provide optimal service, forecasting the number of vehicles is necessary. This study aims to predict the number of vehicles undergoing periodic testing with the lowest possible error rate. The forecasting methods used include Single Moving Average, Weighted Moving Average, and Exponential Smoothing, calculated using the POM-QM for Windows 5 software. The most accurate method is the Single Moving Average, with a prediction of 1,671 vehicles, resulting in a MAD of 177.746, an MSE of 49,887.25, and a MAPE of 9.56%. The validity of the forecasting results was evaluated using a data-driven approach with a tracking signal of -2.232. These findings support resource planning and the enhancement of vehicle testing services