Preprint Communication Version 1 This version is not peer-reviewed

Travel Time Prediction Based on Data Feature Selection and Data Clustering Methods

Version 1 : Received: 22 February 2018 / Approved: 7 March 2018 / Online: 7 March 2018 (13:30:06 CET)

How to cite: Chen, C. Travel Time Prediction Based on Data Feature Selection and Data Clustering Methods. Preprints 2018, 2018030054 (doi: 10.20944/preprints201803.0054.v1). Chen, C. Travel Time Prediction Based on Data Feature Selection and Data Clustering Methods. Preprints 2018, 2018030054 (doi: 10.20944/preprints201803.0054.v1).

Abstract

In recent years, governments applied intelligent transportation system (ITS) technique to provide several convenience services (e.g., garbage truck app) for residents. This study proposes a garbage truck fleet management system (GTFMS) and data feature selection and data clustering methods for travel time prediction. A GTFMS includes mobile devices (MD), on-board units, fleet management server, and data analysis server (DAS). When user uses MD to request the arrival time of garbage truck, DAS can perform the procedure of data feature selection and data clustering methods to analyses travel time of garbage truck. The proposed methods can cluster the records of travel time and reduce variation for the improvement of travel time prediction. After predicting travel time and arrival time, the predicted information can be sent to user’s MD. In experimental environment, the results showed that the accuracies of previous method and proposed method are 16.73% and 85.97%, respectively. Therefore, the proposed data feature selection and data clustering methods can be used to predict stop-to-stop travel time of garbage truck.

Subject Areas

data feature selection; data clustering; travel time prediction

Readers' Comments and Ratings (0)

Leave a public comment
Send a private comment to the author(s)
Rate this article
Views 0
Downloads 0
Comments 0
Metrics 0
Leave a public comment

×
Alerts
Notify me about updates to this article or when a peer-reviewed version is published.