L3S Research Center, Leibniz University of Hannover Appelstraße 9a, 30167 Hannover, Germany E-mail: ghodsi@l3s.de Date: February 20, 2026. Dear Editor-in-Chief, Please find enclosed our manuscript entitled "Integrating Neurosymbolic Systems in Advanced Product Design: A Comprehensive Review" by S. Ghodsi, G. Rabby, F. Engel, S. Münker, and S. Auer, submitted for publication in Neurosymbolic Artificial Intelligence as a survey paper for the Special Call for Survey Papers. Modern product design increasingly relies on data-driven generative and optimization methods, yet engineering practice demands hard constraint satisfaction, traceability, and verification readiness across the CAD-to-manufacturing lifecycle. While neurosymbolic AI has matured rapidly, existing surveys remain largely domain-agnostic and do not systematically connect integration patterns to the operational requirements of product design workflows. Our review addresses this gap by synthesizing neurosymbolic methods across generative CAD, topology optimization, manufacturing planning, assembly, and sustainability-oriented design. Key contributions include: - Systematic review with living resources: Following a PRISMA-inspired protocol, we survey 55 papers and provide both a curated snapshot and a continuously updatable comparison in the Open Research Knowledge Graph (ORKG) for community extension. - Product-design-focused taxonomy: Building on Kautz's integration types, we classify methods by integration pattern (I–VI), symbolic substrates, neural functions, and evaluation domains—revealing systematic patterns and engineering trade-offs. - Lifecycle "survey lens": We map dominant integration patterns to canonical design stages and identify transferable cross-domain architectures (e.g., differentiable solvers, shielding, explainable NS-RL). - Evaluation and deployment perspective: We synthesize evaluation frameworks and summarize technological, ethical, and governance challenges for real-world adoption. We believe this manuscript aligns well with the journal's scope and will interest readers working on neurosymbolic integration, constraint-aware learning, and trustworthy AI for engineering. This manuscript is original, has not been published previously, and is not under consideration elsewhere. All authors have approved the submission and declare no conflicts of interest. Thank you for your consideration. Yours sincerely, Siamak Ghodsi (on behalf of all authors) PhD Candidate, L3S Research Center & TIB, Leibniz University of Hannover ghodsi@l3s.de