"FabSearch" : A 3D CAD Model Based Search Engine for Sourcing Manufacturing Services
Not on the map. Why?
Abstract (arXiv)
In this paper, we present "FabSearch", a prototype search engine for sourcing manufacturer service providers, by making use of the product manufacturing information contained within a 3D digital file of a product. FabSearch is designed to take in a query 3D model, such as the .STEP file of a part model which then produces a ranked list of job shop service providers who are best suited to fabricate the part. Service providers may have potentially built hundreds to thousands of parts with associated part 3D models over time. FabSearch assumes that these service providers have shared shape signatures of the part models built previously to enable the algorithm to most effectively rank the service providers who have the most experience to build the query part model. FabSearch has two important features that helps it produce relevant results. First, it makes use of the shape characteristics of the 3D part by calculating the Spherical Harmonics signature of the part to calculate the most similar shapes built previously be job shop service providers. Second, FabSearch utilizes meta-data about each part, such as material specification, tolerance requirements to help improve the search results based on the specific query model requirements. The algorithm is tested against a repository containing more than 2000 models distributed across various job shop service providers. For the first time, we show the potential for utilizing the rich information contained within a 3D part model to automate the sourcing and eventual selection of manufacturing service providers.
Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). First posted on arXiv 17 Sep 2018.
Builds on
No reference list yet.
Built on by
Citing papers per year: 2021: 2; 2022: 0; 2023: 0; 2024: 3; 2025: 1; 2026: 1.
7 papers.
- 2024BrepGen: A B-rep Generative Diffusion Model with Structured Latent Geometrycited by 103
- 2021BRepNet: A topological message passing system for solid modelscited by 95
- 2024Geometric Deep Learning for Computer-Aided Design: A Surveycited by 14
- 2025Bringing Attention to CAD: Boundary Representation Learning via Transformercited by 9
- 2021Classifying Component Function in Product Assemblies with Graph Neural Networkscited by 2
- 2024Constraint-Aware Feature Learning for Parametric Point Cloudcited by 1
- 2026Brep2Shape: Boundary and Shape Representation Alignment via Self-Supervised Transformerscited by 0
- 2026Brep2Shape: Boundary and Shape Representation Alignment via Self-Supervised Transformers7 Feb 2026
- 2025Bringing Attention to CAD: Boundary Representation Learning via Transformer7 Apr 2025
- 2024Constraint-Aware Feature Learning for Parametric Point Cloud12 Nov 2024
- 2024Geometric Deep Learning for Computer-Aided Design: A Survey27 Feb 2024
- 2024BrepGen: A B-rep Generative Diffusion Model with Structured Latent Geometry28 Jan 2024
- 2021Classifying Component Function in Product Assemblies with Graph Neural Networks8 Jul 2021
- 2021BRepNet: A topological message passing system for solid models1 Apr 2021