Bayesian-Optimized Hierarchical Mixture of Experts for Steel Corrosion-Rate Prediction in Cementitious Mortars

Bayesian-Optimized Hierarchical Mixture of Experts for Steel Corrosion-Rate Prediction in Cementitious Mortars | 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 Bayesian-Optimized Hierarchical Mixture of Experts for Steel Corrosion-Rate Prediction in Cementitious Mortars Taimur Rahman, Md. Farhad Momin, Sagor Kumar Podder, Afra Anam Provasha, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7410653/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 11 Dec, 2025 Read the published version in Journal of Building Pathology and Rehabilitation → Version 1 posted 11 You are reading this latest preprint version Abstract A conditions-aware deep-learning framework based on a Bayesian-optimized Hierarchical Mixture-of-Experts (BO-HME) is introduced to predict steel corrosion rates in cementitious mortars, addressing the one-size-fits-all limitation of single global models. The approach aligns with corrosion mechanisms by using a gating network that learns exposure and chemistry regimes and routes each sample to specialized experts. A consolidated dataset of 275 specimens with 14 inputs spanning mixture, environmental/material, and electrochemical descriptors is analyzed under a leakage-safe pipeline with nested cross-validation; Bayesian optimization (Tree-structured Parzen Estimator) tunes architecture and training controls. Complementary parametric analysis combines linear screening, principal component analysis, and a Gradient Boosting explainer with SHAP to quantify nonlinear and interaction effects. On the test set, BO-HME attains R² = 0.9572, RMSE = 2.1378, MAE = 1.2243, and MAPE = 3.36 percent, outperforming multiple benchmark models under an identical pipeline; relative to leading tree-boosting alternatives, RMSE is reduced by about 20 ~ 25 percent. Predicted-versus-measured plots indicate good calibration over most of the range, with mild under-prediction confined to the highest corrosion rates. The analysis identifies electrochemical state as dominant for prediction, with pH most influential, followed by electrical resistivity, corrosion potential, and chloride-to-hydroxide ratio. A Streamlit interface reproduces the trained pipeline and enables schema-checked single-sample and batch prediction with CSV or JSON export. The approach yields an accurate and deployable tool for durability assessment, threshold-based alerts, and maintenance planning within structural health monitoring workflows. Steel corrosion-rate prediction Cementitious mortars Hierarchical Mixture of Experts Bayesian optimization Deep learning Chloride and carbonation exposure Explainable AI (SHAP) Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 11 Dec, 2025 Read the published version in Journal of Building Pathology and Rehabilitation → Version 1 posted Editorial decision: Revision requested 06 Oct, 2025 Reviews received at journal 23 Sep, 2025 Reviews received at journal 22 Sep, 2025 Reviewers agreed at journal 14 Sep, 2025 Reviewers agreed at journal 12 Sep, 2025 Reviewers agreed at journal 12 Sep, 2025 Reviewers agreed at journal 11 Sep, 2025 Reviewers invited by journal 09 Sep, 2025 Editor assigned by journal 01 Sep, 2025 Submission checks completed at journal 19 Aug, 2025 First submitted to journal 19 Aug, 2025 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-7410653","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":514833136,"identity":"453e09a0-f6c5-4b90-8567-21bda34604f6","order_by":0,"name":"Taimur Rahman","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA6ElEQVRIiWNgGAWjYFACxgYgISHHwMzAcOABkGkA4gLZYHFsgAeixcIYrCWBOC1gUJEIVkCUFnv2w62befdIpPe38xgCbTmcuJ2B+eBtHoY7sjht4Ulsu83zTCJ3xmEeA7CWnQ1sydY8DM+McTsMpOWARO4GZrYEoJbbiRsO8JhJ8wD14tTC/xCsJd0AoYX/G34tEhBbEgyYmQ/AbGHDr+XGw7abcw5IGM44DNJi8N94ZzObseUcA9x+Ye9Pf3bjzYE6ef7+g80fPlSkyW5nb354400F7hBDA6BIYYYxRsEoGAWjYBSQDQCNiVytoxnNpQAAAABJRU5ErkJggg==","orcid":"","institution":"World University of Bangladesh","correspondingAuthor":true,"prefix":"","firstName":"Taimur","middleName":"","lastName":"Rahman","suffix":""},{"id":514833137,"identity":"fd4ebf46-62da-4264-affc-145c007b98af","order_by":1,"name":"Md. Farhad Momin","email":"","orcid":"","institution":"World University of Bangladesh","correspondingAuthor":false,"prefix":"","firstName":"Md.","middleName":"Farhad","lastName":"Momin","suffix":""},{"id":514833138,"identity":"992c9376-710a-4d7d-add6-b25d203c1816","order_by":2,"name":"Sagor Kumar Podder","email":"","orcid":"","institution":"World University of Bangladesh","correspondingAuthor":false,"prefix":"","firstName":"Sagor","middleName":"Kumar","lastName":"Podder","suffix":""},{"id":514833139,"identity":"2c472742-c23f-4f0a-a897-6c3ef83e80f2","order_by":3,"name":"Afra Anam Provasha","email":"","orcid":"","institution":"World University of Bangladesh","correspondingAuthor":false,"prefix":"","firstName":"Afra","middleName":"Anam","lastName":"Provasha","suffix":""},{"id":514833140,"identity":"398dc11e-0a6e-42a4-b24b-64ce4b974557","order_by":4,"name":"Leevesh Kumar","email":"","orcid":"","institution":"World University of Bangladesh","correspondingAuthor":false,"prefix":"","firstName":"Leevesh","middleName":"","lastName":"Kumar","suffix":""}],"badges":[],"createdAt":"2025-08-19 16:53:08","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7410653/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7410653/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s41024-025-00746-x","type":"published","date":"2025-12-11T15:57:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":98243436,"identity":"c170a937-b9a3-4e5e-b5b6-bf0469525821","added_by":"auto","created_at":"2025-12-15 16:02:04","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1812689,"visible":true,"origin":"","legend":"","description":"","filename":"manuscriptJBPR.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7410653/v1_covered_5340757a-4c8c-4487-93a4-3a7e5bfca397.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Bayesian-Optimized Hierarchical Mixture of Experts for Steel Corrosion-Rate Prediction in Cementitious Mortars","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":"journal-of-building-pathology-and-rehabilitation","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bpar","sideBox":"Learn more about [Journal of Building Pathology and Rehabilitation](http://link.springer.com/journal/41024)","snPcode":"41024","submissionUrl":"https://submission.nature.com/new-submission/41024/3","title":"Journal of Building Pathology and Rehabilitation","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Steel corrosion-rate prediction, Cementitious mortars, Hierarchical Mixture of Experts, Bayesian optimization, Deep learning, Chloride and carbonation exposure, Explainable AI (SHAP)","lastPublishedDoi":"10.21203/rs.3.rs-7410653/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7410653/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eA conditions-aware deep-learning framework based on a Bayesian-optimized Hierarchical Mixture-of-Experts (BO-HME) is introduced to predict steel corrosion rates in cementitious mortars, addressing the one-size-fits-all limitation of single global models. The approach aligns with corrosion mechanisms by using a gating network that learns exposure and chemistry regimes and routes each sample to specialized experts. A consolidated dataset of 275 specimens with 14 inputs spanning mixture, environmental/material, and electrochemical descriptors is analyzed under a leakage-safe pipeline with nested cross-validation; Bayesian optimization (Tree-structured Parzen Estimator) tunes architecture and training controls. Complementary parametric analysis combines linear screening, principal component analysis, and a Gradient Boosting explainer with SHAP to quantify nonlinear and interaction effects. On the test set, BO-HME attains R\u0026sup2; = 0.9572, RMSE\u0026thinsp;=\u0026thinsp;2.1378, MAE\u0026thinsp;=\u0026thinsp;1.2243, and MAPE\u0026thinsp;=\u0026thinsp;3.36 percent, outperforming multiple benchmark models under an identical pipeline; relative to leading tree-boosting alternatives, RMSE is reduced by about 20\u0026thinsp;~\u0026thinsp;25 percent. Predicted-versus-measured plots indicate good calibration over most of the range, with mild under-prediction confined to the highest corrosion rates. The analysis identifies electrochemical state as dominant for prediction, with pH most influential, followed by electrical resistivity, corrosion potential, and chloride-to-hydroxide ratio. A Streamlit interface reproduces the trained pipeline and enables schema-checked single-sample and batch prediction with CSV or JSON export. The approach yields an accurate and deployable tool for durability assessment, threshold-based alerts, and maintenance planning within structural health monitoring workflows.\u003c/p\u003e","manuscriptTitle":"Bayesian-Optimized Hierarchical Mixture of Experts for Steel Corrosion-Rate Prediction in Cementitious Mortars","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-16 11:44:22","doi":"10.21203/rs.3.rs-7410653/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-06T11:38:15+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-23T06:49:17+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-22T18:32:15+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"271826313683242998597943641686529708770","date":"2025-09-14T13:18:33+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"51703901029857981269477059897505272882","date":"2025-09-12T08:11:07+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"67092577283657519671329583948397636882","date":"2025-09-12T04:11:49+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"314620086544329087812248886665537678009","date":"2025-09-11T20:11:39+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-09-09T13:48:37+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-01T20:14:12+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-08-19T19:06:49+00:00","index":"","fulltext":""},{"type":"submitted","content":"Journal of Building Pathology and Rehabilitation","date":"2025-08-19T16:37:28+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"journal-of-building-pathology-and-rehabilitation","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bpar","sideBox":"Learn more about [Journal of Building Pathology and Rehabilitation](http://link.springer.com/journal/41024)","snPcode":"41024","submissionUrl":"https://submission.nature.com/new-submission/41024/3","title":"Journal of Building Pathology and Rehabilitation","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"d0435589-7af0-4dbd-83f8-81d2a19453e6","owner":[],"postedDate":"September 16th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-12-15T15:59:58+00:00","versionOfRecord":{"articleIdentity":"rs-7410653","link":"https://doi.org/10.1007/s41024-025-00746-x","journal":{"identity":"journal-of-building-pathology-and-rehabilitation","isVorOnly":false,"title":"Journal of Building Pathology and Rehabilitation"},"publishedOn":"2025-12-11 15:57:00","publishedOnDateReadable":"December 11th, 2025"},"versionCreatedAt":"2025-09-16 11:44:22","video":"","vorDoi":"10.1007/s41024-025-00746-x","vorDoiUrl":"https://doi.org/10.1007/s41024-025-00746-x","workflowStages":[]},"version":"v1","identity":"rs-7410653","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7410653","identity":"rs-7410653","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}