A feasible approach to the integration of CAD and CAPP
Introduction
Integrating CAD and CAM not only plays a key role in achieving digital manufacturing and computer integrated manufacturing, but also is vital to the competitiveness of mechanical manufacturing enterprises and their ability to respond rapidly to market changes, and one of the most critical links for the integration is the link between design and process planning activities [1].
The significance of CAD/CAPP integration arises from the fact that CAPP relies on the product model data provided by CAD to perform precise and consistent process planning for manufacturing. However, they tend to have different product data descriptions, i.e. CAD is usually geometry-based, whilst CAPP/CAM are feature-based and domain-dependent, which results in unsatisfactory practical implementation, or a common weakness of CAPP systems — they usually act as stand-alone functions and do not have a link with either CAD or CAM systems.
This problem could be solved by developing a feature-based CAD system to provide data directly to CAPP systems, but it inevitably imposes limitations on product design and/or modeling, and other downstream applications [2]. Consequently, although many commercial CAD systems are declared to be feature-based, in fact, the so-called feature is just a modeling macro or menu name, such as Protrusion, Revolution, Cutout, Block, etc., instead of a design feature or manufacturing feature in accordance with engineering practice. Even if features, parameters, or constraints are used in design, and CAD/CAM modules are from the same system or vendor and internally linked, consistent information of features is not communicated from CAD to CAM because vendors have treated them only as geometric modeling ‘macros’ [5], [6].
Besides the feature information, which serves as the bridge to a high level integration between design, analysis, process planning and manufacturing, being difficult to recognize and extract from CAD models, another problem is that it is also difficult to embed non-geometric technological information, such as dimensional and geometric tolerance, surface roughness and hardness, that is necessary for CAPP, in current CAD models. At a glance, CAD models seem to incorporate these data as seen in the drawings; in fact, most, if not all, of these data are stored independently in a data file, and do not connect directly to the related features, or exist as real attributes of CAD models [4], [5]. In other words, they are not represented in an integrated form to facilitate effective integration with the down-line activities of the product cycle [3], but are simply represented as technical notes for human interpretation [4].
Therefore, despite a lot of effort made in the past few decades to interlink design and process planning, sharing of design and manufacturing information still remains a bottleneck [1], [2], [3], [4], [5], [6], [7], [8], [9], [10]. It has been reported that imperfect interoperability imposes at least $1 billion per year on the members of the U.S. automotive supply chain, and the majority of these costs are attributable to repairing or re-entering data files involved in the design and manufacture of automobiles, which are not usable for downstream applications [11].
Considering the fact that the developments in the CAPP area have not kept pace with those in CAD/CAM [12], [13], [14], and a general-purpose CAPP system is far from reality, to meet the increasing requirement for such a CAPP that is customized and bi-directionally integrated with current “feature-based” CAD systems that are widely used in industry, this paper proposes a practical solution for integrating design and process planning on the basis of commercial CAD systems. The proposed methodology, system architecture and function implementation are addressed in the following sections.
Section snippets
Related research work and review
As the bridge between CAD and CAM, Computer Aided Process Planning (CAPP) is used to interpret product/part design data in terms of features, analyze the shape, size, tolerance, location, orientation and relationship of various geometric features on a part, and translate them into manufacturing operation instructions in optimal process sequences, so as to convert the stock (raw material) into a finished part economically and competitively. Generally speaking, feature recognition and technical
System architecture
At present, both CAD and CAM system technologies are quite mature, and many commercial automation island systems are widely used in industry. The main problem is the lack of an appropriate CAPP system that could interlink CAD and CAM.
To meet the urgent requirements from industry for an integrated CAD/CAPP system, and overcome the problems such as low efficiency and error proneness of traditional feature recognition systems, this paper proposes a feasible and practical approach for integrating
Feature recognition and conversion algorithm
Feature carries a mass of engineering information, both geometric and non-geometric, and is the medium of information transmission among CAD, CAPP, and CAM. However, although many commercial CAD systems are declared to be feature-based, in fact, the so-called feature is just the modeling operator, menu name (such as Protrusion, Revolution, Sweep, Cutout, Copy, etc.) or a range of primitive volumes (typically Block, Cylinder, Cone, Sphere, etc.), instead of a real form feature or design and
Process planning
Once manufacturing features and the related technical information are obtained, the following key steps are process selection and operation sequencing. In this section, a method that hybridizes knowledge engineering, a neural network and a genetic algorithm (GA) is used to seek a global optimal process plan.
Stock model generating
Mechanical parts are commonly manufactured using multiple manufacturing processes. Primary processes such as casting are usually used to provide the material stock from which the final product is machined. So, how to generate a reasonable 3D geometric model of the starting workpiece is a crucial and practical step. Unfortunately, besides simply selecting the minimal enveloping volume of a cubic block containing the part as a stock, other related research work is rarely reported. Kim and Wang
Conclusions
The basic issues towards the integration of design and manufacturing engineering have been discussed systemically. In contrast to most current academic literature and research reports which focus on theoretical study or developing a prototyping system, this paper proposed a practical approach to a total integration of CAD and CAPP based on commercial systems with the focus on key techniques such as feature parameters and constraints extraction, feature precedence tree reconstruction, technical
References (43)
- et al.
Intelligent system for extraction of product data from CADD models
Computers in Industry
(2001) - et al.
An approach for interlinking design and process planning
Journal of Materials Processing Technology
(2003) - et al.
A manufacturing process information model for design and process planning integration
Journal of Manufacturing Systems
(2003) - et al.
An integrated CAPP/CAM system for stamping die pattern machining
Computer-Aided Design
(2003) - et al.
Striving for a total integration of CAD, CAPP, CAM and CNC
Robotics and Computer-Integrated Manufacturing
(2004) - et al.
An IT view on perspectives of computer aided process planning research
Computers in Industry
(1997) - et al.
Evolution and future perspectives of CAPP
Annals of the CIRP
(1993) - et al.
Graph based heuristics for recognition of machined features from a 3-D solid model
Computer-Aided Design
(1988) - et al.
Automatic recognition of interacting machining features based on minimal condition subgraph
Computer-Aided Design
(1998) - et al.
Toward multiprocessor feature recognition
Computer-Aided Design
(1997)