An Analysis of Electric Vehicle Sales Forecasting in Indonesia: A Quantitative Approach to Operational Planning

Abstract

This study applies a quantitative time-series forecasting approach using three methods: moving average, exponential smoothing, and trend analysis. Secondary data on electric car sales in Indonesia were analyzed to assess the predictive accuracy of each method. Forecasting performance was evaluated using the Mean Absolute Percentage Error (MAPE), a widely recognized accuracy metric. The findings indicate that the moving average method yielded the lowest MAPE for BEV sales at 9.72%, categorized as “highly accurate” (<10%), and for HEV sales at 16.94%, categorized as “accurate” (10–20%). For PHEV sales, the trend analysis method achieved superior accuracy compared to other models. These results highlight that forecasting performance varies across EV categories, reinforcing the importance of model selection based on market characteristics and data patterns.