Prediksi Lokasi Akhir Dari Perjalanan Taksi Berdasarkan Lintasan Terpendek Menggunakan Metode Location-Aware Attentional Trajectory-Bidirectional LSTM

Kurniatama, Farid (2021) Prediksi Lokasi Akhir Dari Perjalanan Taksi Berdasarkan Lintasan Terpendek Menggunakan Metode Location-Aware Attentional Trajectory-Bidirectional LSTM. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
06111740000052-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2023.

Download (2MB) | Request a copy

Abstract

Dalam Intelligent Transportation Systems (ITS), pemberitahuan informasi lokasi akhir lebih awal sangat berguna dalam menunjang pelayanan optimal dan meminimalisir biaya operasional dari perusahaan bidang jasa transportasi baik yang dikelola pemerintah maupun swasta. Penelitian ini bertujuan untuk memprediksi lokasi akhir dari perjalanan suatu kendaraan yang didasarkan pada informasi lintasan terpendek di awal serta mengevaluasi hasil prediksinya. Selanjutnya, penelitian ini menggunakan metode yang diusulkan yaitu Location-Aware Attentional Trajectory-Bidirectional Long Short Term Memory (LAT-BiLSTM) agar mampu mengingat informasi spasial dan temporal dari perjalanan suatu kendaraan tanpa memerlukan informasi tambahan. Sebagai studi kasus, penelitian ini menggunakan data histori umum perjalanan taksi di Porto Portugal yang sangat besar. Hasil eksperimen menunjukkan bahwa metode LAT-BiLSTM berhasil memperoleh hasil prediksi yang lebih akurat dibandingkan Location-Aware Attentional Trajectory-Long Short Term Memory (LAT-LSTM), Bidirectional LSTM (BiLSTM), dan metode first winner yaitu Multi Layer Perceptron (MLP) dengan perbandingan 2,258734 : 2,291371 : 2,293022 : 2,609707 dalam kilometer. Dari hasil prediksi tersebut evaluasi hasil prediksi LAT-BiLSTM memperoleh nilai yang baik dengan MSE = 0,002275, MAE = 0,014939, dan R2Score = 0,214122. Selain itu didapatkan efisiensi informasi data dengan penggunaan GPS saja ===================================================================================================== In Intelligent Transportation Systems (ITS), early notification of final location information is very useful to support optimal services and minimalize operational costs of companies in the transportation service sector, both government-run nor private companies. This study aims to predict the final location of a vehicle’s trip based on the shortest path information at the beginning and evaluate the prediction results. Furthermore, this study used Location-Aware Attentional TrajectoryBidirectional Long Short Term Memory (LAT-BiLSTM) method to be able to remember the spatial and temporal information of a vehicle’s trip without requiring additional information. As study case, this study uses public taxi trip historical data in Porto Portugal which very large data. The experimental results show that the LAT-BiLSTM method succeeded in obtaining more accurate predictions than Location-Aware Attentional Trajectory-Long Short Term Memory (LAT-LSTM), Bidirectional LSTM (BiLSTM), and the first winner method namely Multi Layer Perceptron (MLP) with the ratio is 2,258734 : 2,291371 : 2,293022 : 2,609707 in kilometers. From the prediction results, the evaluation of the LAT-BiLSTM prediction results obtained a good value with MSE = 0,002275, MAE = 0,014939, and R2Score = 0,214122. In addition, data information efficiency is obtained by using GPS only.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Intelligent Transport Systems, Prediksi Lintasan Terpendek, titik GPS, Shortest Path Prediction, GPS Point, LAT - BiLSTM
Subjects: Q Science > Q Science (General) > Q325.5 Machine learning.
Q Science > Q Science (General) > Q325.78 Back propagation
Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression
Q Science > QA Mathematics > QA402.6 Transportation problems (Programming)
Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Q Science > QA Mathematics > QA76.9.D343 Data mining
Q Science > QA Mathematics > QA9.58 Algorithms
T Technology > T Technology (General) > T57.6 Operations research--Mathematics. Goal programming
T Technology > T Technology (General) > T58.8 Productivity. Efficiency
T Technology > TE Highway engineering. Roads and pavements > TE228.3 Intelligent transportation systems.
T Technology > TL Motor vehicles. Aeronautics. Astronautics > TL798.N3 Global Positioning System.
Divisions: Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User: Farid Kurniatama
Date Deposited: 31 Aug 2021 05:14
Last Modified: 31 Aug 2021 05:14
URI: https://repository.its.ac.id/id/eprint/90926

Actions (login required)

View Item View Item