<![CDATA[Estimasi waktu tempuh yang akurat sangat krusial bagi efisiensi sistem transportasi dan layanan logistik di perkotaan. Penelitian ini mengembangkan model estimasi waktu tempuh untuk wilayah Surabaya dengan mengintegrasikan metode interpolasi, regresi linear, dan teknik optimasi Newton-Raphson. Data yang digunakan bersumber dari rute OpenStreetMap serta variabel cuaca (curah hujan dan suhu) dari BMKG. Hasil analisis menunjukkan bahwa jarak tempuh, intensitas hujan, dan kondisi jam sibuk secara signifikan memengaruhi durasi perjalanan. Model ini memiliki tingkat akurasi yang tinggi dengan koefisien determinasi RSquare (???? 2 ) sebesar 0,92. Adapun tingkat kesalahan model diukur melalui Mean Absolute Error (MAE) sebesar 201,05 detik (sekitar 3,3 menit) dan Root Mean Squared Error (RMSE) sebesar 267,53 detik. Melalui simulasi optimasi rute, model ini mampu memberikan saran perjalanan yang 8–15% lebih cepat dibandingkan strategi pemilihan rute konvensional berbasis jarak terpendek. Dengan hasil tersebut, model ini dapat diimplementasikan pada sistem navigasi adaptif dan responsif terhadap perubahan kondisi lingkungan dan lalu lintas. Accurate travel time estimates are crucial for the efficiency of transportation systems and logistics services in urban areas. This study developed a travel time estimation model for the Surabaya area by integrating interpolation, linear regression, and Newton-Raphson optimization techniques. The data used was sourced from OpenStreetMap routes and weather variables (rainfall and temperature) from the BMKG. The results of the analysis show that travel distance, rainfall intensity, and rush hour conditions significantly affect travel duration. This model has a high level of accuracy with a coefficient of determination R Square (R2 ) of 0.92. The model's error rate is measured by the Mean Absolute Error (MAE) of 201.05 seconds (approximately 3.3 minutes) and the Root Mean Squared Error (RMSE) of 267.53 seconds. Through route optimization simulations, this model is able to provide travel suggestions that are 8–15% faster than conventional route selection strategies based on the shortest distance. With these results, this model can be implemented in adaptive navigation and responsive systems that respond to changes in environmental and traffic conditions.]]>
