Budiman, Daniel Prasetio (2026) DESIGN AND DEVELOPMENT OF A TOURISM BUS MANAGEMENT WEBSITE WITH PRICE PREDICTION USING THE XGBOOST ALGORITHM (CASE STUDY: PT. SETIO BUDI LUHUR TRANS). Undergraduate thesis, UPN Veteran Jawa Timur.
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Cover)
22081010304-cover.pdf Download (1MB) |
|
Text (Bab 1)
22081010304-bab1.pdf Download (189kB) |
|
Text (Bab 2)
22081010304-bab2.pdf Restricted to Repository staff only until 22 July 2028. Download (333kB) |
|
Text (Bab 3)
22081010304-bab3.pdf Restricted to Repository staff only until 22 July 2028. Download (5MB) |
|
Text (Bab 4)
22081010304-bab4.pdf Restricted to Repository staff only until 22 July 2028. Download (2MB) |
|
Text (Bab 5)
22081010304-bab5.pdf Download (181kB) |
|
Text (Daftar Pustaka)
22081010304-daftarpustaka.pdf Download (211kB) |
|
Text (Lampiran)
22081010304-lampiran.pdf Restricted to Repository staff only until 22 July 2028. Download (251kB) |
Abstract
PT. Setio Budi Luhur Trans is a tourism bus rental service company that still manages operational data and transactions manually, and determines rental prices based on subjective considerations without a consistent calculation standard. This condition leads to inefficiencies in data management and inaccuracies in determining bus rental prices. This study aims to design and develop a tourism bus management website integrated with a rental price prediction feature using the Extreme Gradient Boosting (XGBoost) algorithm. The system was developed as a web-based application using the Laravel framework with an Agile development method, encompassing features for managing bus data, customer data, and rental transactions in an integrated manner, as well as a price prediction feature that considers variables including travel distance, bus type, and season category (normal or holiday). The XGBoost model was trained using 196 historical bus rental price records, divided into 85% training data (166 records) and 15% testing data (30 records). Evaluation results demonstrated excellent model performance across four bus categories (Medium–Normal, Medium–Holiday, Big–Normal, and Big– Holiday), with an average MAPE of 0.74%, an average MAE of IDR 25.584, and an average coefficient of determination (R²) of 0.9934. The developed system successfully generates automatic and real-time bus rental price estimations, replacing the inconsistent manual pricing process, thereby providing convenience for both managers and prospective customers in obtaining accurate and transparent pricing information. Keywords: Management Website, Tourism Bus, Price Prediction, XGBoost, Laravel, Machine Learning
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science | ||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Informatics | ||||||||||||
| Depositing User: | Daniel Prasetio Budiman | ||||||||||||
| Date Deposited: | 22 Jul 2026 06:30 | ||||||||||||
| Last Modified: | 22 Jul 2026 07:46 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/57733 |
Actions (login required)
 |
View Item |