Evaluating lightweight 3D graphics formats for product visualization and data exchange
Nathan W. Hartman
Purdue University
Evaluating lightweight 3D graphics formats for product visualization and data exchange
ABSTRACT[1]
As companies continue to embrace product lifecycle management practices to improve business operations, the ability to create and communicate 3D product representations throughout the enterprise is becoming increasingly important. Given the geographically dispersed product design and manufacturing scenarios commonplace in industry today, companies are grappling with decisions regarding the use of specific formats and mechanisms to promote communication and collaboration processes. Companies designing and producing particularly large artifacts, whether in terms of geometry scale or in terms of object complexity, experience problems of portability and scalability when using vendor-supported file formats. Current solutions tend to center on “lightweight” file formats as one of the enabling technologies supporting this distributed collaboration. This paper presents the results of applied research examining the capabilities of three common lightweight formats and their ability to retain product information upon translation out of the native Computer-Aided Design (CAD) database.
INDEX TERMS[2]
CAD, Industry research, Lightweight formats, Product lifecycle management
I. INTRODUCTION
Data exchange typically falls into one of three categories: paper-to-CAD, part-to-CAD, or CAD-to-CAD [1]. While the issues surrounding paper-to-CAD translation are compelling (e.g., accuracy, design intent, corporate standards), similar issues also increasingly arise within the CAD-to-CAD category and must be addressed. In fact, due to interoperability considerations regarding different geometry kernels within current CAD systems, more resources are being devoted to this latter data exchange option. A critical path within the interoperability domain is the migration of CAD data from one software package to another, or between differing versions of the same vendor’s tool. This process is facilitated by more traditional technologies, such as IGES and STEP, and by contemporary technologies, such as JT and 3DXML. Indeed, many companies have begun adopting the latter options for communication within and outside their organizations, in addition to use within typical engineering design tasks. While the specific reasons for this approach may vary, the basic premise is that these technologies hold the potential to capture information in a lightweight, 3D format in a manner in which IGES and STEP cannot.
With the proliferation of emerging visualization formats, confusion exists in industry with regard to the use of these formats relative to STEP and other standard data formats. As technology vendors advance the capability of “lightweight” file formats, selecting the appropriate file format for a specific purpose is critical to the communication and collaboration process. Given the geographically dispersed product design and manufacturing scenarios commonplace in industry today, companies are grappling with decisions regarding the use of specific formats and mechanisms to promote communication and collaboration processes. Current solutions tend to center on “lightweight” file formats as one of the enabling technologies that support this distributed collaboration. The recent availability of these visualization file formats has caused confusion and uncertainty in industry relative to their use in specific situations, especially when trying to capture annotations, accurate geometry, and manufacturing information, for example. Companies are spending millions of dollars implementing product lifecycle management (PLM) toolsets, reasoning that they can communicate more effectively the design of their products. However, this thought is based on the assumption that the toolsets within the PLM environment share data in such a way that product data communication is promoted. The overall PLM market in 2002 was up slightly compared to 2001 at approximately $13.5 billion, where 33% of that total is due to Collaborative Product Definition Management (CPDM). “While the tools segment remained flat in 2002 at $9.3 billion, CPDM grew by 8% to reach $4.2 billion” [2]. End-user spending on product definition management (PDM) technologies expanded at a compound annual growth rate of 12% through 2008 [3].
Unfortunately, data sharing in the PLM environment is often sabotaged by incompatibilities between software tools and by the lack of functionality in emerging technologies within the CAD software suite, including neutral file formats. It has been estimated that over $1 billion were spent annually in the automotive supply chain to directly address interoperability issues [4]. In the architectural/engineering/construction AEC industry, annual costs are estimated at $15.8 billion [5]. In a recent survey by Kubotek USA [6], more than 2800 CAD managers and users were surveyed regarding the issues they face regarding CAD interoperability. Forty-three (43) percent of people using a history-based modeling tool, such as CATIA, NX, or Pro/ENGINEER, must rebuild their models over half of the time when completing a design task because the original model becomes unstable. In addition, 19 percent of those people surveyed are using some alternative form of direct modeling to avoid the cost and time associated with rebuilding a model. The following list is a summary of particularly salient points [6], [7]:
a. 50% of the people surveyed said that they had to redesign an object, based on current data or on 3D data they received, as frequently as once a week and sometimes more often.
b. 84% of those people who use a history-based modeling tool resort to editing the original feature trees within the CAD model.
c. 43% of those surveyed said they used the CAD system that the model was made in to make the changes to the object.
d. On a monthly basis, 44% of respondents said they send or receive CAD files in a format that is not the same as their preferred system.
e. On a monthly basis, 37% of those surveyed send or receive four or more CAD files in a format that is not from their preferred CAD system.
With these types of data exchange rates and disparities of CAD systems, it is clear that any improvements to the process will have measurable effects on the time and money spent on moving data between CAD systems. It is also plausible that any improvements made to this process will likely need to be supported both from the vendors and the user base of a particular toolset. One potential improvement to this process is the increasing use of XML-based technologies. They provide the ability to store (and eventually distribute) a variety of information types, including assembly instructions, materials resource planning (MRP) data, and annotations [8]. CAD software vendors are taking advantage of XML technology by creating basic viewing tools to communicate the 3D data contained in model geometry and textual data that is embedded in the geometric database. However, the technology is lacking in many areas.[3]
II. BACKGROUND[4]
Global design and manufacturing environments are becoming commonplace, with products being designed and manufactured at anytime and in any place. A key link in this process is the ability to communicate information effectively to those people who need it. The medium for communication has changed. No longer are companies sending stacks of drawings to suppliers as their sole method of communication. The use of 3D CAD tools by many product design organizations has led to the use of the 3D model as a conduit for communication and a repository for product information [9]. However, the interoperability within and between CAD systems is a well-documented issue, which causes companies to waste substantial amounts of time and money trying to overcome those problems [10].
In addition to the problems created by differing data formats from native CAD systems, it is not always desirable to share the native CAD files outside the originating organization. As these CAD models are being embedded with corporate intellectual property and design standards, a more secure scenario is necessary to provide customers (both internal and external) with the information they need without compromising the security of the information contained within the CAD model [11]. In most cases, companies might choose neutral standard file formats (e.g., STEP or IGES); however, these formats carry an overhead with them due to robust geometry representations that yield a large file size. As suggested by Ball, Ding, and Patel [12], [13] and Ding et al. [14], it is often a combination of traditional standard formats and contemporary lightweight file formats. The novelty of these new lightweight formats has led to confusion in the user community relative to specific functionality contained in the files and the most appropriate scenarios in which to use these files. This paper outlines an industry-based research project meant to address these concerns.
III. PROBLEM STATEMENT[5]
Information proliferation from the design and manufacturing areas of a business have made selecting appropriate file formats for communication within this domain difficult and potentially inaccurate. Much confusion exists regarding capabilities and usage characteristics within the industrial user base.
IV. METHODOLOGY[6]
Part 1 of the project compares selected lightweight visualization formats (3DXML, JT, and U3D) to the functionality contained with the STEP AP 203 e2 format. STEP stands for the Standard for the Exchange of Product Model Data, and it is used widely in industry for long-term archival of geometry and other design attributes, as well as for data exchange between CAD systems. 3DXML, JT, and U3D are all commercial formats being developed by various software vendors for the communication and dissemination of design information without the mathematical overhead of the native CAD format. In the absence of a standard method for comparing lightweight file formats, the criteria used for evaluating new functionality within the STEP file was used for the comparisons made in Part 1. Figure 1 illustrates the process of assessing the lightweight formats against the known metrics for standard file formats.
Figure 1. Process Outline for Comparing Lightweight File Formats
Part 2 of the project focuses on the development of a checklist that can be used to determine the applicability of particular lightweight formats to a given situation. The research team created a survey to assess the relevant characteristics of lightweight file formats and presented it to relevant experts in industry for feedback. Taking that feedback, the survey was revised and then administered to 15 industry participants that represented the aerospace, defense, manufacturing, government, and consulting industries. Upon examining relevant literature topics [12], the survey was organized into five sections: openness, extensibility, accessibility, interoperability, and security. The results of those interviews were compiled and used to generate a checklist of important characteristics to describe lightweight file formats in an industry usage scenario. Government and aerospace were the primary industry sectors represented in this survey, and it should be noted that these companies are currently the primary implementers of PLM philosophies.
Part 3 was meant to develop outlines for specific use cases to aid industry in the selection and implementation of lightweight file formats for key tasks, including collaborative design evaluation, request for quote from supplier, and information transfer from design to manufacturing. These use cases were influenced by the work done in Parts 1 and 2 as a basis for completion. An online survey was created to gather data, which included three sections corresponding to the aforementioned preliminary use-case scenarios. Twelve (12) participants completed all sections of the survey, while 59 participants completed the section of the survey that most closely aligned with their specific job functions. Job functions included design engineering, manufacturing engineering, analysis, manufacturing planning, and customer support. Participants represented the aerospace OEM, aerospace supplier, automotive, heavy equipment, and government industry sectors. Each of the use cases posed a scenario that involved an initial sender of information, a translation/export process, and a receiver of information (Figure 2).
Figure 2. Use Case Data Translation Process
V. RESULTS AND ANALYSIS[7]
The results for Part 1 of the study yields a comparison of JT, 3DXML, and U3D to the STEP AP 203 e2 standard using defined evaluation criteria for the process. Figure 2 provides an overall view of the process, with detailed results found in Hartman and Lim [15]. Note that this type of research is a combination of evaluating file formats and viewing technology currently available. As software vendors improve their technology offerings, these results may vary. Tests of the lightweight file formats show many negative results when compared to the STEP file features. Many STEP AP203 ed2 features are not available with the lightweight formats (consistent with supporting format documentation), such as form features, construction history, and drafting capabilities. It is conceivable that some features are not available by default and need to be extended manually. This suggests a difference in the fundamental roles these lightweight formats are built upon compared to the STEP file format. Currently these formats would likely support collaboration and visualization, but need to be enhanced in order to support long-term storage or archival scenarios. The focus of Part 2 of the research study is to develop a checklist that could be used for determining which of the commercial lightweight formats in question would be most appropriate to use in a given situation. The importance of this checklist is summarized in the following points:
a. No standard method of assessing visualization formats.
b. Industry looking for a way to display/store/retain data in lightweight formats.
c. Some “lightweight” formats are not lightweight.
d. Visualization formats used in different ways.
e. The final checklist can be found with additional detail in Hartman and Lim [15].
Part 3 of this research focuses on the development of use-case scenarios for determining the suitability of using a particular format in a specific situation. Three of the most common scenarios for industry application are collaborative design evaluation, request for bid/quote, and communication from design to manufacturing. In a collaborative design evaluation scenario, the designers present their CAD design to other engineers, company decision makers, or perhaps customers. This could take place within a company or between companies. The original designers could either send out their native CAD file or the appropriate exported formats to the receiver end. The receivers would load the file, interrogate, annotate, propose changes, and eventually finalize the design for manufacturing. In a request for quote process, a company invites several suppliers to present and bid on their design based on the company’s requirements. The company will then select the best design at the best price. In other cases, the company provides their own design to the best manufacturer. Several important factors during the transactions of CAD data in this bidding process are the protection of intellectual property of the supplier or the company and design review capabilities such as the ones for a collaborative design evaluation process. In the process of design to manufacturing, design engineers communicate their design to the manufacturers for prototyping or large-scale manufacturing. The manufacturer in this case could be in-house or outsourced to another party. The designers present their CAD data in the appropriate format that contains manufacturing information such as material properties, mechanical properties, geometric dimensioning and tolerance, and important annotations. In some cases, a highly accurate geometry data is also provided for the manufacturer to perform a manufacturing analysis. The Methodology section of this paper contains a procedural figure outlining the nature of the data translation. The intent of this portion of the research was not to specify a specific lightweight format for a specific situation but to expose the characteristics of the scenario in which the format would be used. By combining results from Part 3 with the results from Parts 1 and 2, a user can make an informed assessment of a specific lightweight format for use in a specific situation.