UJI KORELASI DAN ANALISIS PREDIKSI KEPADATAN LALU LINTAS PADA DATA KENDARAAN BERMOTOR DI KABUPATEN MALANG MENGGUNAKAN METODE XGBOOST

Oktaviani, Sheny Eka and Salsabilah, Elina (2025) UJI KORELASI DAN ANALISIS PREDIKSI KEPADATAN LALU LINTAS PADA DATA KENDARAAN BERMOTOR DI KABUPATEN MALANG MENGGUNAKAN METODE XGBOOST. Project Report (Praktek Kerja Lapang). Universits Pembangunn Nasional Veteran Jawa Timur.

	Text (Cover) 21083010037_Cover.pdf Download (695kB)
	Text (Bab 1) 21083010037_Bab 1.pdf Download (193kB)
	Text (Bab 2) 21083010037_Bab 2.pdf Restricted to Registered users only until 19 June 2027. Download (278kB)
	Text (Bab 3) 21083010037_Bab 3.pdf Restricted to Registered users only until 19 June 2027. Download (1MB)
	Text (Bab 4) 21083010037_Bab 4.pdf Restricted to Registered users only until 19 June 2027. Download (175kB)
	Text (Daftar Pustaka) 21083010037_Daftar pustaka.pdf Download (143kB)
	Text (Lampiran) 21083010037_Lampiran.pdf Restricted to Registered users only until 19 June 2027. Download (306kB)

Abstract

The growth of motorized vehicles in Malang Regency has led to a significant increase in traffic congestion. This poses a challenge in managing congestion and ensuring smooth transportation, which is crucial for regional economic growth. This study aims to analyze the condition and predict traffic density in Malang Regency using the XGboost method, which is known to be effective for complex regression and non-linear data. The data used includes road length, area, and number of motor vehicles, which are taken from official sources such as the Bina Marga Public Works Office and the Transportation Office. The research stages include data preprocessing, correlation tests using Pearson, Spearman, and Kendall methods, and predictive modeling using XGboost. The correlation test results showed a weak to moderate positive correlation between road length and number of vehicles, with statistically significant correlation coefficient values. The constructed XGboost model showed good performance, with a Mean Absolute Error (MAE) value of 2752.98 and a coefficient of determination R-squared (R²) of 0.894. The prediction of traffic density for the next five years shows a steady upward trend, with significant variations between sub-districts. In conclusion, the XGboost method is effective for predicting traffic density in Malang District, providing valuable insights for better infrastructure planning and management. These predictions are expected to assist in designing appropriate policies to reduce traffic congestion and support sustainable development in Malang District. With a deeper understanding of the factors that influence traffic congestion, this research is expected to make a significant contribution to improving the quality of life of the people in Malang District. Keywords: Malang District One Data (KAMASUTA), Regional Apparatus Organization (OPD), Information and Documentation Management Officer (PPID)

Item Type:

Monograph (Project Report (Praktek Kerja Lapang))

Contributors: