AI-Driven Integrated Cadastre and Land Tenure Platform for Smart Cities

Thesis Abstract

This study addresses the growing inefficiencies and inequities in urban land administration by developing an AI-driven integrated cadastre and land tenure platform tailored for smart cities, combining cadastral data, title records, zoning constraints, and beneficiary-based tenure analytics to support transparent and secure land governance. The aim is to design, implement, and evaluate a scalable ICT solution that integrates heterogeneous data sources, automates tenure verification, enhances spatial decision-making, and promotes inclusive land access. Specific objectives include (1) to synthesize a multi-layered data architecture that ingests parcel, ownership, zoning, and transactional data from municipal, state, and private sources; (2) to develop AI-enabled modules for automated title verification, boundary reconciliation, and anomaly detection to improve accuracy and reduce adjudication time; (3) to implement a smart contract-ready governance layer that supports transparent transfer, leasing, and encumbrance workflows; (4) to evaluate user acceptance, system usability, and decision-support effectiveness among urban planners, surveyors, and property owners; and (5) to assess potential social, economic, and regulatory impacts of platform adoption on tenure security and urban land markets. The methodology adopts a mixed-methods research design conducted in three phases. Phase one involves a quantitative survey of 320 stakeholders (surveyors, property agents, urban planners, and residents) to assess current pain points and requirements, complemented by structured interviews with 25 officials from the cadastral and land registry offices. Phase two implements the platform prototype in a metropolitan district of 50 square kilometers with 32,450 parcels, using a sample of 1,200 parcels for validation and 80 land registries for governance evaluation. Data collection instruments include structured questionnaires, semi-structured interview guides, system usage logs, and documentation of policy frameworks. Phase three employs quantitative and qualitative analyses to evaluate performance and acceptance. Statistical methods include regression analysis to identify predictors of user satisfaction and system adoption, ANOVA to compare performance across user groups, and time-to-decision analysis to measure adjudication efficiency. The AI components are evaluated using precision, recall, and F1 scores for boundary detection and title verification, and ROC-AUC for risk scoring models. The governance layer is analyzed through process mining to assess workflow conformance and through thematic analysis of interview data to extract perceived barriers and enablers. Theoretical grounding draws on the Institutional Theory to examine normative and regulatory influences on platform adoption, and the Technology Acceptance Model extended with Trust and Perceived Risk to explain user uptake. Expected findings include (1) improved accuracy and speed of boundary reconciliation and title verification, (2) measurable reductions in adjudication time and error rates, (3) enhanced transparency and stakeholder trust in land tenure processes, and (4) positive shifts in tenure security indicators and informal market normalization. The study contributes to knowledge by delivering an end-to-end, AI-enabled cadastral platform that harmonizes disparate data silos, introduces automated governance workflows, and provides empirical evidence on the socio-technical implications of digital land administration in dense urban settings. Policy-relevant insights are anticipated regarding data governance, interoperability standards, and scalable deployment models for municipal land administration agencies. The conclusion anticipates that the platform will deliver significant efficiency gains, reduce disputes related to boundaries and ownership, and support more equitable access to land resources in smart city contexts. Recommendations emphasize aligning data standards across jurisdictions, investing in trusted AI governance to mitigate bias, ensuring open yet secure APIs for interoperability, and pursuing phased rollouts with continuous evaluation to maximize adoption and impact.