A Spot-recommendation System for Taxi Drivers Using Monte Carlo Optimization

Conference proceedings article


ผู้เขียน/บรรณาธิการ


กลุ่มสาขาการวิจัยเชิงกลยุทธ์


รายละเอียดสำหรับงานพิมพ์

รายชื่อผู้แต่งPhanhong, Mullika; Likitpanjamanon, Pitchaporn; Chareonwai, Pitsanurak; Srisurapanon, Viroat; Phunchongharn, Phond;

ผู้เผยแพร่Hindawi

ปีที่เผยแพร่ (ค.ศ.)2020

ISBN9781728181066

นอก0146-9428

eISSN1745-4557

URLhttps://www.scopus.com/inward/record.uri?eid=2-s2.0-85097351949&doi=10.1109%2fIBDAP50342.2020.9245611&partnerID=40&md5=8db04b8a6c46983e14b2b940fef5ecbf

ภาษาEnglish-Great Britain (EN-GB)


ดูบนเว็บไซต์ของสำนักพิมพ์


บทคัดย่อ

Public transportation always plays an important role in human daily life. One of the most popular forms of public transportation is taxi. Even though there is a high demand for taxis, most of the time, the taxi drivers still have problems with seeking passengers. In this study, we aim to develop a taxi recommendation system in order to help the taxi drivers do their jobs better. Instead of driving aimlessly, using the recommendation system will suggest them of where they should stop and wait for passengers. The proposed system composes of three main parts, which are: i) defining mock-up taxi stops in the area using K-means ii) estimating the time that drivers have to wait for passengers at each stop through survival analysis and iii) recommending the optimal stops for the taxi drivers using Monte-Carlo Optimization. The objective is to find the optimal waiting spots that minimizing the unloading time and maximizing the loading distance. To evaluate the efficiency of the recommendation system, we used the back-testing process with the passenger simulation. The evaluation result pointed out that, with the recommendation system, the taxi drivers could drive with an average 17.85% less in the term of an unloading distances (distances without any passengers in the car) and average 52.45 kilometers of loading distances (distances with passengers in the car) more than usual daily. Therefore, using the spot-recommendation system would upgrade the taxi drivers' quality of work as well as help alleviate the traffic problems and improve the transportation system in Bangkok. © 2020 IEEE.


คำสำคัญ

Demand AnalysisIntelligent Transportation SystemMonte Carlo OptimizationOptimization model


อัพเดทล่าสุด 2023-23-09 ถึง 07:36