Manufacturing-Process-Conditioned Generative 3D Modeling: A Unified Framework for Geometry, Material-State Evolution, and Additive Manufacturing Trajectories
Abstract
Recent advances in three-dimensional (3D) generative modeling have enabled the synthesis of complex geometric structures through diffusion models, neural radiance fields, 3D Gaussian Splatting, and point-cloud generative models. However, generating a geometrically plausible object does not necessarily produce a feasible manufacturing plan. Additive manufacturing introduces a sequential construction process in which geometry, material distribution, thermal or curing conditions, interlayer bonding, and manufacturing constraints evolve together. This paper proposes Manufacturing-Process-Conditioned Generative 3D Modeling (MPCG-3D), a framework for jointly generating 3D structures and the manufacturing trajectories required to construct them. The proposed representation maintains an explicit manufacturing state containing geometric occupancy, material attributes, process-dependent physical states, connectivity, and defect estimates. A conditional generative policy proposes manufacturing actions, while a learned or physics-based transition model predicts their consequences. A trajectory-level objective combines geometric fidelity, material consistency, process-state accuracy, manufacturability, and resource expenditure. Unlike approaches that treat toolpath generation as an isolated downstream operation, MPCG-3D formulates manufacturing as a sequential state-conditioned generation problem. The framework is designed to support material extrusion, photopolymerization, and powder-bed fusion through process-specific transition models and action spaces. This paper establishes the conceptual formulation, optimization objective, training strategy, and experimental protocol for evaluating the approach. Since no implementation or experimental results are reported, the framework's effectiveness, computational efficiency, and generalization remain to be empirically established.
Related
- Intelligent manufacturing system and hybrid generative-attention framework for quality prediction of high-performance aluminum alloys
- An Explainable MRI-Conditioned Generative Framework for Gene-Feature Representations in Alzheimer’s Disease
- Layerwise Deposition-Field Generative Modeling: A Process-Conditioned Framework for Three-Dimensional Generation Through Thin-Film Deposition and Surface Growth
- Meta-Generative Adversarial Networks: A Framework for the Adaptive Construction, Optimization, and Evolution of Generative Adversarial Mechanisms
- Precipitation-Driven 3D Generative Modeling: A Framework for Reaction-Guided Spatial Occupancy, Local Mineral Deposition, and Long-Term Shape Evolution
BibTeX
@misc{zhang2026manufacturingprocessconditioned,
title = {{Manufacturing-Process-Conditioned Generative 3D Modeling: A Unified Framework for Geometry, Material-State Evolution, and Additive Manufacturing Trajectories}},
author = {Jincheng Zhang},
howpublished = {Zenodo},
year = {2026},
month = oct,
doi = {10.5281/zenodo.23636259},
url = {https://zenodo.org/records/23636259},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Manufacturing-Process-Conditioned Generative 3D Modeling: A Unified Framework for Geometry, Material-State Evolution, and Additive Manufacturing Trajectories
AU - Jincheng Zhang
PY - 2026
DA - 2026/10/25
PB - Zenodo
DO - 10.5281/zenodo.23636259
UR - https://zenodo.org/records/23636259
AB - Recent advances in three-dimensional (3D) generative modeling have enabled the synthesis of complex geometric structures through diffusion models, neural radiance fields, 3D Gaussian Splatting, and point-cloud generative models. However, generating a geometrically plausible object does not necessarily produce a feasible manufacturing plan. Additive manufacturing introduces a sequential construction process in which geometry, material distribution, thermal or curing conditions, interlayer bonding, and manufacturing constraints evolve together. This paper proposes Manufacturing-Process-Conditioned Generative 3D Modeling (MPCG-3D), a framework for jointly generating 3D structures and the manufacturing trajectories required to construct them. The proposed representation maintains an explicit manufacturing state containing geometric occupancy, material attributes, process-dependent physical states, connectivity, and defect estimates. A conditional generative policy proposes manufacturing actions, while a learned or physics-based transition model predicts their consequences. A trajectory-level objective combines geometric fidelity, material consistency, process-state accuracy, manufacturability, and resource expenditure. Unlike approaches that treat toolpath generation as an isolated downstream operation, MPCG-3D formulates manufacturing as a sequential state-conditioned generation problem. The framework is designed to support material extrusion, photopolymerization, and powder-bed fusion through process-specific transition models and action spaces. This paper establishes the conceptual formulation, optimization objective, training strategy, and experimental protocol for evaluating the approach. Since no implementation or experimental results are reported, the framework's effectiveness, computational efficiency, and generalization remain to be empirically established.
ER -