Travel time model using gps data and machine learning for bus information systems

dc.contributor.authorSamarasinghe, P
dc.contributor.authorKumarage, A
dc.contributor.authorPerera, A
dc.contributor.authorNanayakkara, S
dc.contributor.editorGunaruwan, TL
dc.date.accessioned2023-10-12T08:20:19Z
dc.date.available2023-10-12T08:20:19Z
dc.date.issued2023-08-26
dc.description.abstractPublic Transportation modes are prevalent and extensively utilised means of transportation for commuters. Road congestion, bus crew issues, malfunctions, and miscellaneous factors impede buses from adhering to schedules. As a result, it is becoming problematic for commuters to arrange their travel plans confidently. Intelligent transportation systems use Global Positioning System (GPS) technology and data analytics to accurately predict real-time travel information and improve traveller and operator experience. The research gap is the unavailability of standardised techniques for mass travel time predictions using standardised analytical methods. However, recent research has focused on developing accurate travel-time models employing machine learning algorithms. Predictive models rely on past data gathered through GPS systems. The study uses the GPS data of public buses in Central Province, Sri Lanka, one thousand buses have been fitted with GPS units since 2019 (7 million of data). Realtime and historical data that were gathered through GPS units can be used to develop machine learning-based models to predict bus or passenger transport information accurately. The study analysed available data using Microsoft Azure, Statistical, Time Series and Machine algorithms for performance accuracy with lower error rates on predictions used for comparison purposes.en_US
dc.identifier.citation**en_US
dc.identifier.conferenceResearch for Transport and Logistics Industry Proceedings of the 8th International Conferenceen_US
dc.identifier.departmentDepartment of Transport and Logistics Managementen_US
dc.identifier.emailsamarasinghepanchali@gmail.comen_US
dc.identifier.emailamal.kumarage58@gmail.comen_US
dc.identifier.emailasoka.uom@gmail.comen_US
dc.identifier.emailsamudaya@uom.lken_US
dc.identifier.facultyEngineeringen_US
dc.identifier.pgnospp. 177-179en_US
dc.identifier.placeMoratuwa, Sri Lankaen_US
dc.identifier.proceedingProceedings of the International Conference on Research for Transport and Logistics Industryen_US
dc.identifier.urihttp://dl.lib.uom.lk/handle/123/21563
dc.identifier.year2023en_US
dc.language.isoenen_US
dc.publisherSri Lanka Society of Transport and Logisticsen_US
dc.relation.urihttps://slstl.lk/r4tli-2023/en_US
dc.subjectTravel timeen_US
dc.subjectPredictionen_US
dc.subjectGPS Dataen_US
dc.subjectMachine learningen_US
dc.titleTravel time model using gps data and machine learning for bus information systemsen_US
dc.typeConference-Full-texten_US

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
TRAVEL TIME MODEL USING GPS DATA AND MACHINE.pdf
Size:
128.29 KB
Format:
Adobe Portable Document Format
Description:

License bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description:

Collections