<![CDATA[The rapid growth of motorized vehicles in Indonesia, unmatched by proportional expansion in road infrastructure, has intensified pressure on the national transportation system. This study examines the application of a Long Short-Term Memory (LSTM) model to analyze and forecast the national traffic load ratio, defined as the ratio of total motorized vehicles to total road length. Annual aggregate data from the Indonesian Central Bureau of Statistics (BPS) for the period 2016–2023 were used in the analysis. The results indicate that the model achieved a strong fit on the training data, with RMSE = 0.3652 and MAE = 0.3617, but performed substantially worse on the test data, with RMSE = 1.7585 and MAE = 1.7585. This discrepancy suggests overfitting, largely attributable to the extremely limited sample size. As such, the findings should be interpreted as exploratory rather than as evidence of reliable forecasting performance. Despite these limitations, the model projects a continued upward trend in national infrastructure pressure over the next five years. These findings provide an initial data-driven indication that transportation infrastructure demand in Indonesia is likely to intensify, while also underscoring the need for future research using larger datasets and baseline model comparisons before policy-level application can be justified.]]>
Copyrights © 2026
