Elsevier

Computer-Aided Design

Volume 59, February 2015, Pages 140-154
Computer-Aided Design

Simplification of feature-based 3D CAD assembly data of ship and offshore equipment using quantitative evaluation metrics

https://doi.org/10.1016/j.cad.2014.03.003Get rights and content

Highlights

  • Shipyards need simplified equipment 3D CAD assembly data for system-level design.

  • Three simplification operations applicable for 3D CAD assembly data are developed.

  • Evaluation metrics considering multiple simplification criteria are proposed.

  • The evaluation metrics allow discriminatory priority of specific criteria.

Abstract

In the design process for ship outfitting and offshore plants, an equipment catalog database is compiled in order for shipyards to reutilize data effectively. However, the current procedure for building such a catalog causes wastage of time because the modelers in the shipyard must perform manual modeling of the 3D CAD data in order to decrease the size of 3D CAD data and adopt a different level of detail (LOD) depending on the purpose of its use. This problem arises because equipment suppliers are not willing to give all of their 3D CAD data to shipyards, out of fear of the loss of intellectual property. Moreover, the 3D CAD data of equipment suppliers have a high LOD, while a shipyard’s 3D CAD data have a relatively low LOD. Therefore, it is necessary to introduce an automated method to simplify the 3D CAD assembly data for equipment that is received from the equipment supplier. In addressing this problem, this study first proposes criteria for a simplification process and quantitative evaluation metrics for the simplification of 3D CAD assembly data, considering the characteristics of equipment data in the shipbuilding industry. Based on these findings, a simplification system was developed, and, four experimental test cases that were conducted on-site were used to verify the proposed system. The results showed that the data to be stored could be reduced to at least 25% of the original 3D CAD assembly data while ports, outer boundaries, and connectivity between CAD parts could be maintained.

Introduction

In the shipbuilding industry, three-dimensional (3D) CAD systems have been introduced and are widely used in the process of design, production, and delivery  [1]. Using these systems, it becomes possible to check and modify design results in real time. For example, connections and interference among different pieces of equipment can be checked in advance before installing them. Moreover, building a system database allows many designers to share information and receive feedback from each other, thus reducing the number of errors and improving the productivity of the entire process  [2]. This has led to a reduced production time, and has significantly increased sales.

In general, equipment suppliers create 3D CAD data with a high level of detail (LOD) to manufacture equipment. Shipyards, on the other hand, focus mainly on installing the equipment that is provided by equipment suppliers. From the view point of system-level design by shipyards, detail shape of equipment is not important but exact port and principle dimension data should be provided. Due to this reason, shipyards need 3D CAD data that has been relatively simplified compared to the original data, and add additional non-geometric information such as ports, principal dimensions, and detailed specifications to the simplified 3D CAD data. Besides, the LOD in simplified 3D CAD data differs depending on the purpose of its use. Fig. 1, for example, shows the same engine model with different LODs. Moreover, since a ship or an offshore plant is a huge system consisting of millions of parts  [3], [4], storing simplified 3D CAD data in a shipbuilding 3D CAD system has benefits in data storage and computing speed.

To create such simplified 3D CAD data, shipyards mostly subcontract the modeling work to a small company, whose modelers then create the simplified data manually based on a limited number of 2D drawings. These modelers need to follow the modeling requirements given by the shipyard, such as preserving the ports and the principal dimensions of the equipment. This practice involves extra time and expense for shipyards, and the resulting data can differ from modeler to modeler. These problems could be solved if there were a standard method, to be implemented in a computer program, for simplifying the 3D CAD data that is created by equipment suppliers to the LOD that shipyards require. Equipment suppliers usually create feature-based 3D CAD assembly data. Thus, a method of simplification has to consider not only such data types, but also the modeling requirements of the shipyard. Ultimately, there is a need for a standard, computerized method for adjusting the LOD of 3D CAD data to save time and efforts. The resulting model should then be exported to various formats without a modeling history, because equipment suppliers hesitate to give shipyards their detailed 3D CAD data for fear of the loss of intellectual property.

To simplify 3D CAD data, a simplification operation and evaluation metrics for the geometric elements of the 3D CAD data are necessary. The simplification operation removes selected model elements including vertices, faces, or features. Evaluation metrics measure the priority of the model elements, and select low priority model elements to be removed. Therefore, it is important to establish the appropriate evaluation criteria and metrics that can be used to evaluate the relative importance of model elements  [5]. In previous studies  [6], [7], [8], geometric information was widely used as simplification criteria. Only a few studies have been carried out that consider non-geometric information, such as the modeling requirements as related to a specific industry, or the characteristics of an assembly model.

In this research, we review related studies in Section  2. In Section  3, we analyze the requirements for the simplification of 3D CAD assembly data, and propose a new set of evaluation metrics. The detailed procedure for the proposed simplification method is described in Section  4. The experimental results that were obtained with test cases in which the method was applied are described in Section  5. Finally, we present our conclusion in Section  6.

Section snippets

Review of related studies

Many methods to simplify and reduce 3D CAD data have been proposed in earlier studies, and these can be divided into the following three approaches according to the data type: polygon-based  [9], [10], [11], [12], b-rep-based  [13], [14], [15], and feature-based approaches  [1], [6], [7], [8], as shown in Fig. 2. The polygon-based approach focuses mainly on reducing the number of triangles in a mesh that does not entail much shape deformation. The b-rep-based approach simplifies shapes by using

Requirements for the simplification of 3D CAD data

The requirements for the simplification of 3D CAD data for ship and offshore plant equipment were defined by conducting interviews with shipyard employees. The purpose of the interviews was to investigate current practices of equipment 3D CAD data simplification in the shipbuilding industry, and to derive user requirements for automatic simplification of the process. The interviewees were department heads in charge of constructing equipment catalogs, and information system executives at the “D”

Application scenario

Equipment suppliers hand over 2D drawings or B-rep model of their equipment deliverables to shipyards. However, they do not want to provide feature-based model which contains a lot of design knowledge. Therefore, it is practically impossible for shipyards to simplify the feature-based model of equipment suppliers.

For this reason, this study assumes the following application scenario, as shown in Fig. 8. Shipyards, firstly, provide the simplification system proposed in this study. Equipment

Development environment and system structure

A prototype system has been developed in the following environment.

  • -

    OS: Windows 7 64-bit

  • -

    Programming language: C++, MFC

  • -

    XML parser: MSXML 4.0

  • -

    Geometric modeling kernel: ACIS R23

  • -

    CAD adaptor: 3D InterOp R23

  • -

    Visualization engine: HOOPS 1919

The system structure was designed as shown in Fig. 15. The application function layer consists of the most important modules as regards the simplification, management, and manipulation of geometric and non-geometric information. The system can import and export the

Conclusion

In this study, we first discussed the difference between shipyards and equipment suppliers with regard to the complexity of 3D CAD data. In the process of building an equipment catalog database, shipyards need to make simplified 3D CAD data to reduce the amount of data to be stored and manipulated in large 3D CAD models of the total ship including all systems installed, which leads to a significant waste of time and money.

Previous studies on the simplification of 3D CAD data were reviewed, and

Acknowledgments

This research was supported by the Industrial Strategic Technology Program (Project ID 10040162) and the Technology Innovation Program (Project ID 2011T100200145) funded by the Ministry of Knowledge Economy of the Korean government. The authors gratefully acknowledge these supports.

References (29)

  • A. Thakur et al.

    Improving performance of rigid body dynamics simulation by removing inaccessible regions from geometric models

    Computer-Aided Design

    (2012)
  • D. Mun et al.

    Knowledge-based part similarity measurement utilizing ontology and multi-criteria decision making method

    Adv. Eng. Inform.

    (2011)
  • D.H. Mun et al.

    A set of standard modeling commands for the history-based parametric approach

    Computer-Aided Design

    (2003)
  • Y. Kang et al.

    Method to simplify ship outfitting and offshore plant equipment three-dimensional (3-D) computer-aided design (CAD) data for construction of an equipment catalog

    J Marine Sci Technol

    (2013)
  • Cited by (0)

    View full text