Multimodal AI-Assisted Crack Provenance Assessment Using Surface-Engineered Polymer Marker Fibers in Cementitious Composites

Abstract

This technical report presents MIOPR-KC-MM, a multimodal material-information framework for crack provenance assessment in cementitious composites. The core of the system consists of surface-engineered polymer marker fibers that combine crack-bridging functionality with a local diagnostic signal in the crack zone. The material signal is treated as one input modality and is fused with crack-image morphology, object-card data, monitoring records and the chronology of geotechnical actions. The proposed framework introduces three main indices: WSM-AI for rapid material verification, WPPR-KC-MM for integrated crack provenance assessment, and WPR-GEO for the geotechnical component of crack history. The output is a probabilistic crack provenance vector rather than a single binary class, enabling a structured assessment of shrinkage, thermal, load-induced, technological, service-related and geotechnical crack origins. MIOPR-KC-MM draws on several scientific and technical areas: self-sensing cementitious composites, surface engineering of polymer fibers, visual detection methods for microcracks, AI in structural diagnostics, material passports and life-cycle object documentation. The integration of these areas forms a coherent diagnostic framework for crack provenance assessment in cementitious composites. [1] Bekzhanova, Z.; Memon, S.A.; Kim, J.R.; et al. Self-Sensing Cementitious Composites: Review and Perspective. Nanomaterials 2021, 11, 2355. https://doi.org/10.3390/nano11092355 [2] Jung, S.; Lee, S.; Yu, J. Ontological Approach for Automatic Inference of Concrete Crack Cause. Applied Sciences 2021, 11, 252. https://doi.org/10.3390/app11010252 [3] Yu, K.-K.; Zhao, T.-Q.; Luo, Q.-L.; Ping, Y. Recycled PET Fibers with Dopamine Surface Modification for Enhanced Interlayer Adhesion in 3D Printed Concrete. Materials 2024, 17, 5126. https://doi.org/10.3390/ma17205126 [4] Song, Y.K.; Lee, K.H.; Kim, D.M.; Chung, C.M. A Microcapsule-Type Fluorescent Probe for the Detection of Microcracks in Cementitious Materials. Sensors and Actuators B: Chemical 2015, 222, 1159-1165. https://doi.org/10.1016/j.snb.2015.08.011 [5] Jiang, L.; Wu, M.; Du, F.; Chen, D.; Xiao, L.; Chen, W.; Du, W.; Ding, Q. State-of-the-Art Review of Microcapsule Self-Repairing Concrete: Principles, Applications, Test Methods, Prospects. Polymers 2024, 16, 3165. https://doi.org/10.3390/polym16223165 [6] Yamane, T.; Chun, P.-J. Bridge Damage Cause Estimation Using Multiple Images Based on Visual Question Answering. arXiv:2302.09208, 2023. [7] Ekanayake, B.; Thengane, V.; Wong, J.K.-W.; Ling, S.H. CracksGPT: Exploring the Potential and Limitations of Multimodal AI for Building Crack Analysis. Buildings 2025, 15, 4327. https://doi.org/10.3390/buildings15234327 [8] EN 14889-2:2006. Fibres for concrete - Part 2: Polymer fibres - Definitions, specifications and conformity. European Committee for Standardization. [9] Senarathne, H.N.Y.; et al. Developing a Standardized Materials Passport Framework to Unlock the Full Circular Potential in the Construction Industry. Sustainability 2025, 17, 6337. https://doi.org/10.3390/su17146337