Elsevier

Computers in Industry

Volume 56, Issue 3, April 2005, Pages 289-304
Computers in Industry

A framework for fast 3D solid model exchange in integrated design environment

https://doi.org/10.1016/j.compind.2004.11.003Get rights and content

Abstract

Exchanging 3D solid models across engineering applications has become increasingly important to integrated design environments (IDEs). However, transferring models among distributed locations via computer networks usually consumes large amounts of network bandwidth, requires powerful machines, and introduces potentially unacceptable latency. This paper presents a software framework to realize fast solid model exchange in IDEs. In particular, this framework integrates two key components: (a) incremental editing that allows solid models to be edited, modified, and propagated in an incremental form, and (b) progressive streaming that enables solid models to be stored, retrieved, transmitted, and reconstructed in a progressive manner. The benefits of this framework include obtaining real-time interaction with complex geometry, and obviating the burden of transferring large amounts of data over networks.

Introduction

With the advent of Internet-based computing technologies, network-centric integrated design environments (IDEs) have been emerged as a systematic solution to deal with multi-disciplinary design tasks that today's industry is facing. In general, an IDE can be regarded as a heterogeneous computing environment that provides engineers with a platform to collaborate various spatially distributed CAD/CAM/CAE (or CAX in short) applications through computer networks. Compared with conventional product developing practices, some advantages brought by IDEs may be concluded as below (but are not limited to):

  • Parallel development cycles replace serial development cycles in order to shorten product development time;

  • Distributed workplaces replace centralized workplaces in order to efficiently and flexibly utilize product development resources;

  • Network-connected applications replace isolated applications in order to freely share product development information.

Solid modeling plays a key role in IDEs. It provides CAX applications with tools of creating, modifying, querying, and exchanging product information. Traditionally, solid modeling systems are single-user driven, stand-alone systems with limited interaction with multiple participants. However, under the philosophy of IDE, it becomes possible that the data of solid models can be shared conveniently and solid modeling operations can be executed simultaneously between various CAX applications. Fig. 1 illustrates an example of solid modeling in IDEs. Here, a CAD system takes the responsibility of creating solid models. Based on these models, two CAM systems are set up to conduct the tasks of process planning and NC machining, respectively. A CAE system is also used to perform finite element analysis. All these systems are connected together via computer networks. As a result, engineers using these applications can collaboratively work together, and hence, the productivity and efficiency of product developments can be greatly improved.

IDEs require quick and efficient exchange of solid models, not only to maintain the integrity of model data across a variety of CAX applications, but also to dynamically distribute data of interest to engineers who do not directly interact with a specific engineering application. Typically, IDEs are constructed on network-connected computing environments, i.e., geographically dispersed software systems communicate with each other through local area networks (LAN), Internet/Intranet and so on. Therefore, the issue of fast exchanging solid models between applications via networks is very important and needs to be carefully considered.

Among all the information of solid models conveyed between applications, model shape information has always been the most important type of data that need to be delivered quickly, reliably and with high-quality. This is because geometric properties are the most fundamental information for any solid model. Traditional methods of uploading/downloading model geometry between applications involve repeated transfers of entire solid models using ftp-like network protocols. Considering the (oftentimes) large amounts of geometry data and the limited network bandwidth, these approaches usually result in long response time and delayed interaction, and hence, severely negate the benefits of IDEs. In order to alleviate such problems, this paper introduces a software infrastructure known as Fast eXchange Framework (FXF) which integrates various approaches to efficiently access solid models in a wide range of network and computing environments.

The rest of the paper is structured as follows. A review of previous literature is presented in Section 2. In Section 3, the overall infrastructure of the FXF is described. The approach of incrementally editing solid models is proposed in Section 4. Section 5 introduces the approach of progressively streaming solid models. Some test results are reported in Section 6. Finally, a summary of the present research is given in Section 7.

Section snippets

Literature review

An integrated design environment can be viewed as a computing environment that links various computer aided activities related to product life cycle, e.g., design, analysis, manufacturing, inspection, and visualization. There are several different types of IDEs have been proposed in previous literature.

Hoffmann and Joan-Arinyo [1], [2] introduced the master model. In this research, the net shape information (e.g., faces, edges) of a solid model is created and stored at a master application. For

Infrastructure

The Fast eXchange Framework is based on a client–server architecture that enables one application (denoted as the server) to send solid model information via networks and other applications (denoted as the client) to receive the information from the server. Note that any application in IDEs, at any instant, could be a server and/or client depending on whether it is sending or receiving models. The two main objectives of the FXF include:

  • Reducing the amount of data transferred over networks when

Incremental editing

In CAX applications, typical (shape) modification of B-reps includes global and local edits. Global edits (assuming in a feature-based modeling system, which has been widely accepted in industry and academia) include feature addition/deletion and feature re-parameterization. Local edits involve directly manipulating the B-reps, e.g., by blending, chamfering, etc. Since any B-rep can be treated as a cell complex1

Progressive streaming

Incremental editing alleviates the need for repeatedly transferring entire solid models during the course of editing. However, it is not able to help increasing the speed of solid model transmission. In other words, if a huge solid model has to be directly transmitted from the server to client, or a complex cell change model (i.e., the CCM which is the resultant model of editing) is due to be exchanged, conveying large amount of model data over bandwidth-limited networks may still result in

Results

A prototype system was developed to implement the Fast eXchange Framework. The server was installed on an Dell Dimension L700CX desktop (Intel Celeron 700MHz processor). The client was installed on a Toshiba Satellite 1805 laptop (Intel Celeron 800MHz processor). All of these PCs are running Windows 2000 and connected by dial-up networks. Here, the reason to use the relatively “slow” dial-up connection and PCs is to test the efficiency of the proposed FXF approach in a worst-case scenario of

Conclusions

As computer networks technologies have been tremendously improved, more and more integrated design practices begin to require CAD/CAM/CAE systems to not only provide powerful solid modeling functionality, but also be able to exchange solid models over networks quickly and reliably. This paper proposes the Fast eXchange Framework that addresses two critical issues of exchanging solid models: incremental editing and progressive streaming. This framework improves the communication in integrated

Acknowledgements

Part of the proposed work was conducted in the PEARL laboratory at the University of Toledo. The author would like to thank Swati Bhargava for her useful comments and suggestions.

Di Wu is a Software Engineer at SolidWorks Corporation, working on product design software. He received his MS degree in Mechanical Engineering from the University of Toledo in 2001. Prior to that, he also earned BE and MS degrees in Mechanical Engineering in 1994 and 1997, respectively. His research interests include: geometric and solid modeling, computer graphics, and scientific visualization.

References (19)

There are more references available in the full text version of this article.

Cited by (6)

Di Wu is a Software Engineer at SolidWorks Corporation, working on product design software. He received his MS degree in Mechanical Engineering from the University of Toledo in 2001. Prior to that, he also earned BE and MS degrees in Mechanical Engineering in 1994 and 1997, respectively. His research interests include: geometric and solid modeling, computer graphics, and scientific visualization.

Dr. Radha Sarma completed her PhD in mechanical engineering from The University of Michigan. She worked as assistant and associate professor at Iowa State University and University of Toledo, respectively. She also taught at The University of Michigan. Her research interests include computer-aided design and manufacture, surface and solid modeling, and MEMS.

View full text