Deep Learning for Traffic Prediction and Trend Deviation Identification: A Case Study in Hong Kong

Deep Learning for Traffic Prediction and Trend Deviation Identification: A Case Study in Hong Kong | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Deep Learning for Traffic Prediction and Trend Deviation Identification: A Case Study in Hong Kong Xiexin Zou, Edward Chung, Hongbo Ye, Haolin Zhang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4553585/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 04 Oct, 2024 Read the published version in Data Science for Transportation → Version 1 posted 9 You are reading this latest preprint version Abstract This paper presents a robust method for predicting traffic volume and speed on major strategy routes in Hong Kong, leveraging data from data.gov.hk and employing deep learning models. The proposed approach provides future predictions ranging from 6 minutes to one hour, considering the reliability of traffic detectors. The method involves extracting hidden deep features from the historical data of each detector to derive the detector profile. Subsequently, detectors are grouped into clusters based on the similarity of their profiles. A CNN-LSTM prediction model is provided for each cluster. The proposed approach demonstrates its resilience to detector failures. Various tests evaluating prediction performance across detector failure rates ranging from 1% to 20% validate its capability to deliver accurate predictions despite the presence of failed detectors. Without failed detectors, MAE, RMSE, and MAPE for traffic volume prediction for the next 6 minutes are 5.17 vehicles/6 minutes, 7.64 vehicles/6 minutes, and 14.07%, respectively. For traffic speed prediction are 3.70 km/h, 6.32 km/h, and 6.33%. Considering a failure rate of approximately 6% in the Hong Kong dataset, when 6% of the detectors failed, the MAE, RMSE, and MAPE for traffic volume prediction are 5.24 vehicles/6 minutes, 7.81 vehicles/6 minutes, and 14.21%, respectively. The corresponding values for traffic speed prediction are 3.87 km/h, 6.55 km/h, and 6.68%. However, the limitation of the proposed method is its potential to underperform when predicting rare or unseen scenarios, indicating the need for future research to incorporate additional data sources and methods to enhance predictive performance. traffic volume prediction traffic speed prediction deep learning clustering method detector profile Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 04 Oct, 2024 Read the published version in Data Science for Transportation → Version 1 posted Editorial decision: Revision requested 08 Jul, 2024 Reviews received at journal 08 Jul, 2024 Reviews received at journal 01 Jul, 2024 Reviewers agreed at journal 19 Jun, 2024 Reviewers agreed at journal 19 Jun, 2024 Reviewers invited by journal 19 Jun, 2024 Editor assigned by journal 18 Jun, 2024 Submission checks completed at journal 18 Jun, 2024 First submitted to journal 09 Jun, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4553585","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":321443427,"identity":"ce67ccc4-c0e7-4fba-950d-6b66144ae8de","order_by":0,"name":"Xiexin Zou","email":"","orcid":"","institution":"The Hong Kong Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Xiexin","middleName":"","lastName":"Zou","suffix":""},{"id":321443428,"identity":"df9911a8-f406-411e-aac3-098baff70184","order_by":1,"name":"Edward Chung","email":"data:image/png;base64,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","orcid":"","institution":"The Hong Kong Polytechnic University","correspondingAuthor":true,"prefix":"","firstName":"Edward","middleName":"","lastName":"Chung","suffix":""},{"id":321443429,"identity":"d6270ced-dc2f-4358-b68a-d2bc0ce79fe0","order_by":2,"name":"Hongbo Ye","email":"","orcid":"","institution":"The Hong Kong Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Hongbo","middleName":"","lastName":"Ye","suffix":""},{"id":321443430,"identity":"3a7b1588-ed7c-4ed6-91e1-2cadd709abbb","order_by":3,"name":"Haolin Zhang","email":"","orcid":"","institution":"The Hong Kong Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Haolin","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2024-06-09 11:27:20","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4553585/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4553585/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s42421-024-00112-2","type":"published","date":"2024-10-04T15:58:06+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":66096946,"identity":"88063773-ad2f-449f-8dfc-e92a01c94e36","added_by":"auto","created_at":"2024-10-07 16:12:04","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5402737,"visible":true,"origin":"","legend":"","description":"","filename":"latexDST20240609.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4553585/v1_covered_f6aa12f7-ff57-4a44-ae51-f24005763ee1.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Deep Learning for Traffic Prediction and Trend Deviation Identification: A Case Study in Hong Kong","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"data-science-for-transportation","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Data Science for Transportation](https://www.springer.com/journal/42421)","snPcode":"42421","submissionUrl":"https://submission.nature.com/new-submission/42421/3","title":"Data Science for Transportation","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"snapp","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"traffic volume prediction, traffic speed prediction, deep learning, clustering method, detector profile","lastPublishedDoi":"10.21203/rs.3.rs-4553585/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4553585/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"This paper presents a robust method for predicting traffic volume and speed on major strategy routes in Hong Kong, leveraging data from data.gov.hk and employing deep learning models. The proposed approach provides future predictions ranging from 6 minutes to one hour, considering the reliability of traffic detectors. The method involves extracting hidden deep features from the historical data of each detector to derive the detector profile. Subsequently, detectors are grouped into clusters based on the similarity of their profiles. A CNN-LSTM prediction model is provided for each cluster. The proposed approach demonstrates its resilience to detector failures. Various tests evaluating prediction performance across detector failure rates ranging from 1% to 20% validate its capability to deliver accurate predictions despite the presence of failed detectors. Without failed detectors, MAE, RMSE, and MAPE for traffic volume prediction for the next 6 minutes are 5.17 vehicles/6 minutes, 7.64 vehicles/6 minutes, and 14.07%, respectively. For traffic speed prediction are 3.70 km/h, 6.32 km/h, and 6.33%. Considering a failure rate of approximately 6% in the Hong Kong dataset, when 6% of the detectors failed, the MAE, RMSE, and MAPE for traffic volume prediction are 5.24 vehicles/6 minutes, 7.81 vehicles/6 minutes, and 14.21%, respectively. The corresponding values for traffic speed prediction are 3.87 km/h, 6.55 km/h, and 6.68%. However, the limitation of the proposed method is its potential to underperform when predicting rare or unseen scenarios, indicating the need for future research to incorporate additional data sources and methods to enhance predictive performance.","manuscriptTitle":"Deep Learning for Traffic Prediction and Trend Deviation Identification: A Case Study in Hong Kong","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-07-04 06:19:09","doi":"10.21203/rs.3.rs-4553585/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-07-08T15:29:52+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-07-08T15:20:59+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-07-02T00:58:47+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"213489755628682546735664608450824701193","date":"2024-06-19T12:31:22+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"212235606470989166334818539176959220350","date":"2024-06-19T11:54:27+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-06-19T08:11:53+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-06-18T12:57:50+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-06-18T12:57:21+00:00","index":"","fulltext":""},{"type":"submitted","content":"Data Science for Transportation","date":"2024-06-09T11:26:01+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"data-science-for-transportation","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Data Science for Transportation](https://www.springer.com/journal/42421)","snPcode":"42421","submissionUrl":"https://submission.nature.com/new-submission/42421/3","title":"Data Science for Transportation","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"snapp","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"0a51b81c-2ac8-43af-8934-efeb7908b768","owner":[],"postedDate":"July 4th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2024-10-07T16:04:20+00:00","versionOfRecord":{"articleIdentity":"rs-4553585","link":"https://doi.org/10.1007/s42421-024-00112-2","journal":{"identity":"data-science-for-transportation","isVorOnly":false,"title":"Data Science for Transportation"},"publishedOn":"2024-10-04 15:58:06","publishedOnDateReadable":"October 4th, 2024"},"versionCreatedAt":"2024-07-04 06:19:09","video":"","vorDoi":"10.1007/s42421-024-00112-2","vorDoiUrl":"https://doi.org/10.1007/s42421-024-00112-2","workflowStages":[]},"version":"v1","identity":"rs-4553585","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4553585","identity":"rs-4553585","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}