A Multi-Layer Explainable AI Framework for Transparent Urban Budget Allocation in Smart Cities: SHAP Attribution, Causal Inference,Counterfactual Reasoning, and the Budget Explainability Score

A Multi-Layer Explainable AI Framework for Transparent Urban Budget Allocation in Smart Cities: SHAP Attribution, Causal Inference,Counterfactual Reasoning, and the Budget Explainability Score | 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 Article A Multi-Layer Explainable AI Framework for Transparent Urban Budget Allocation in Smart Cities: SHAP Attribution, Causal Inference,Counterfactual Reasoning, and the Budget Explainability Score Mariam Labib Francies, Abeer Twakol Khalil, Hanan M. Amer, Mohamed Maher Ata This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9087087/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Urban budget allocation in smart cities represents a critical Artificial Intelligence (AI) governance challenge, as deployed machine learning (ML) models remain opaque, difficult to audit, and misaligned with emerging regulatory frameworks. This study proposes a Multi-Layer Explainable AI (XAI) framework designed to improve transparency and accountability in AI-assisted budget allocation. The framework integrates SHapley Additive exPlanations (SHAP)-based feature attribution, a Structural Causal Model (SCM) with do-calculus intervention estimation, and Diverse Counterfactual Explanations (DiCE) for policy scenario generation. In addition, we introduce the Budget Explainability Score (BES), a composite governance metric that combines fidelity, interpretability, and equity to evaluate the overall transparency of AI-driven budget systems. The framework is evaluated on real open municipal budget datasets from Chicago (10,172 records, 2023–2026) and New York City (681,131 records, 2026). To avoid data leakage, prior allocation features are constructed exclusively from temporally lagged records and evaluated using a strict temporal holdout strategy, achieving an R² of 0.959 on unseen data. The results reveal strong algorithmic budget inertia, with prior allocation dominating feature importance across both cities. Causal analysis further identifies fund type assignment as the primary structural lever influencing allocation outcomes, while fairness evaluation reveals systematic prediction inequality across administrative fund categories. The BES provides a stable composite measure for evaluating governance-oriented AI budget systems, validated through policy-constrained Monte Carlo simulation confirming robustness across weight configurations. These findings demonstrate that multi-layer XAI pipeline can serve as critical infrastructure for accountable AI-supported governance, with direct implications for transparent public finance management and sustainable smart city development. Humanities/Complex networks Social science/Complex networks Physical sciences/Mathematics and computing Explainable AI Smart City Governance Urban Budget Allocation SHAP Structural Causal Model Counterfactual Reasoning AI Ethics Budget Explainability Score Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 17 May, 2026 Reviewers agreed at journal 12 May, 2026 Reviewers invited by journal 28 Apr, 2026 Editor invited by journal 16 Mar, 2026 Editor assigned by journal 12 Mar, 2026 Submission checks completed at journal 12 Mar, 2026 First submitted to journal 10 Mar, 2026 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-9087087","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":631199062,"identity":"7807ee54-bf3b-4289-921c-9144a7b424a0","order_by":0,"name":"Mariam Labib Francies","email":"data:image/png;base64,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","orcid":"","institution":"Mansoura University","correspondingAuthor":true,"prefix":"","firstName":"Mariam","middleName":"Labib","lastName":"Francies","suffix":""},{"id":631199064,"identity":"24085b91-2e48-4dd7-8ad1-1f6a2402531f","order_by":1,"name":"Abeer Twakol Khalil","email":"","orcid":"","institution":"Mansoura University","correspondingAuthor":false,"prefix":"","firstName":"Abeer","middleName":"Twakol","lastName":"Khalil","suffix":""},{"id":631199066,"identity":"5de4d15f-60f3-46f1-a333-77ed87a83a2c","order_by":2,"name":"Hanan M. Amer","email":"","orcid":"","institution":"Mansoura University","correspondingAuthor":false,"prefix":"","firstName":"Hanan","middleName":"M.","lastName":"Amer","suffix":""},{"id":631199077,"identity":"980b46a1-5a69-4694-9abf-5af78f0d6468","order_by":3,"name":"Mohamed Maher Ata","email":"","orcid":"","institution":"Zewail City of Science and Technology, 6th of October City","correspondingAuthor":false,"prefix":"","firstName":"Mohamed","middleName":"Maher","lastName":"Ata","suffix":""}],"badges":[],"createdAt":"2026-03-10 18:38:45","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9087087/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9087087/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108805729,"identity":"748755c6-613e-4955-8cb6-78f3bad4638b","added_by":"auto","created_at":"2026-05-08 15:26:44","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1002974,"visible":true,"origin":"","legend":"","description":"","filename":"AMultiLayerExplainableAIFrameworkforTransparentUrbanBudgetAllocationinSmartCitiesSHAPAttributionCausalInferenceCounterfactualReasoningandtheBudgetExplainabilityScore.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9087087/v1_covered_8b10dd73-285e-4e8f-a806-13deb0fb3810.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"A Multi-Layer Explainable AI Framework for Transparent Urban Budget Allocation in Smart Cities: SHAP Attribution, Causal Inference,Counterfactual Reasoning, and the Budget Explainability Score","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Explainable AI, Smart City Governance, Urban Budget Allocation, SHAP, Structural Causal Model, Counterfactual Reasoning, AI Ethics, Budget Explainability Score","lastPublishedDoi":"10.21203/rs.3.rs-9087087/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9087087/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eUrban budget allocation in smart cities represents a critical Artificial Intelligence (AI) governance challenge, as deployed machine learning (ML) models remain opaque, difficult to audit, and misaligned with emerging regulatory frameworks. This study proposes a Multi-Layer Explainable AI (XAI) framework designed to improve transparency and accountability in AI-assisted budget allocation. The framework integrates SHapley Additive exPlanations (SHAP)-based feature attribution, a Structural Causal Model (SCM) with do-calculus intervention estimation, and Diverse Counterfactual Explanations (DiCE) for policy scenario generation. In addition, we introduce the Budget Explainability Score (BES), a composite governance metric that combines fidelity, interpretability, and equity to evaluate the overall transparency of AI-driven budget systems. The framework is evaluated on real open municipal budget datasets from Chicago (10,172 records, 2023\u0026ndash;2026) and New York City (681,131 records, 2026). To avoid data leakage, prior allocation features are constructed exclusively from temporally lagged records and evaluated using a strict temporal holdout strategy, achieving an R\u0026sup2; of 0.959 on unseen data. The results reveal strong algorithmic budget inertia, with prior allocation dominating feature importance across both cities. Causal analysis further identifies fund type assignment as the primary structural lever influencing allocation outcomes, while fairness evaluation reveals systematic prediction inequality across administrative fund categories. The BES provides a stable composite measure for evaluating governance-oriented AI budget systems, validated through policy-constrained Monte Carlo simulation confirming robustness across weight configurations. These findings demonstrate that multi-layer XAI pipeline can serve as critical infrastructure for accountable AI-supported governance, with direct implications for transparent public finance management and sustainable smart city development.\u003c/p\u003e","manuscriptTitle":"A Multi-Layer Explainable AI Framework for Transparent Urban Budget Allocation in Smart Cities: SHAP Attribution, Causal Inference,Counterfactual Reasoning, and the Budget Explainability Score","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-07 13:47:43","doi":"10.21203/rs.3.rs-9087087/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-17T08:47:41+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"166791469484429386174422709284858308552","date":"2026-05-13T02:11:39+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-28T13:41:34+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-03-16T09:16:11+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-12T12:43:18+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-12T12:43:07+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-03-10T18:22:57+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"aa5907c2-870c-4e3c-bb5e-cf7d29eb5954","owner":[],"postedDate":"May 7th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-17T08:47:41+00:00","index":96,"fulltext":""},{"type":"reviewerAgreed","content":"166791469484429386174422709284858308552","date":"2026-05-13T02:11:39+00:00","index":94,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":67182161,"name":"Humanities/Complex networks"},{"id":67182163,"name":"Social science/Complex networks"},{"id":67182164,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-05-07T13:47:43+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-07 13:47:43","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9087087","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9087087","identity":"rs-9087087","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}