Step standardized product data representation for product development and automated process planning.

Schumann, Christian ; Militzer, Joerg ; Teich, Tobias 等

1. INTRODUCTION

Masses of variable customer demands require cost estimates as fast and exact as possible during preliminary stages to obtain potential orders and execute them cost-effectively afterwards. Especially in the area of machining parts very individual types of customer inquiries are possible. If there were already developed similar products then product developers can perform an intuitive estimation. Otherwise time-consuming analysis and planning are required that occasion not only costs but also include risks of miscalculation.

For a fast estimation and of course for optimizing the whole product development, enhanced product data models are needed. We focus on providing required information during the whole product lifecycle to achieve a complete integration of very different applications. In our scope of product development the feature technology is a very promising approach. Combined with the Standard for the Exchange of Product model Data (STEP), a standardized solution for the integration of different application systems within different application areas is specified by STEP's AP 224.

For us, the AP 224--Mechanical product definition for process planning using machining features--plays an important role. Additionally in the area of process planning we need advanced product models for developing products faster and more effectively to reach the goal of automated offers to customers. In this paper we present one detail of our model for providing quotations automatically by using STEP.

Therefore this paper is organized as follows: First we will deal with several problems relating the whole product lifecycle and especially the phases of product design and development within section two. Afterwards we are going to present STEP and our planned way of using STEP in section three. Finally we describe our current work in section four. The paper concludes with a pre--and a review of work that was done and to be done.

2. BACKROUND

2.1 Product Development

Design is a building block of the product lifecycle for developing new products. It pursues the goal of creating the description of a new product. It does not matter how the designer goes ahead. The only important aspect is the result of his work. For instance, to represent a new design draft, a simple pencil sketch can be sufficient (Cross, 2000).

Particularly during the product development developers should also consider the following phases of the product lifecycle for an optimized conjunction. (Ehrlenspiel et al., 2005) emphasized in this context that developers do not respect this requirement in practice. Cross (2000) also mentioned that a huge amount of decisions made by designers is intuitive or relies on experience. These decisions cannot be mapped into traditional product data models although they are constitutive for realizing a product and first of all an effective process planning.

During the last decades virtual product development became more and more important. As a consequence many Computer Aided Design (CAD) systems came up to stay.

The well known systems like CATIA thereby rely on their own, proprietary formats, but also support neutral ones. Own research shows in this context a good representation of surfaces only. The history of steps needed for the completion of the final shape is still missing. But this history is essential, if one wants to automatically generate process-plans from this draft. Next to an interpretable format of the CAD-drawing the method of designing a model is essential.

2.2 Process Planning

A process plan is a concrete guideline for the manufacturing process. Furthermore process planning describes a systematic sequence of manufacturing steps for the creation of the product based on a defined raw stock.

[FIGURE 1 OMITTED]

As shown in figure one, there are many parameters that should be respected during the development of a process plan. Beside the pure geometry, we need additional information about the product like dimensions, material properties, tolerances or the surface finish.

Furthermore the complete manufacturing environment has to be considered. Process planning creates a document for manufacturing a product that is satisfying all requirements in an economic manner. Computer aided process planning can be defined as a computer-based set of functionality supporting the decisions of a process planner and the development of a process plan. The two primary approaches are variant and generative planning (Nasr & Kamrani, 2006). For generative process planning, which is the core of our approach, all parameters of a manufacturing environment have to be described and provided in a database. Such parameters are for instance available raw materials, existing machines and their executable operations. Another criterion is to acquire the decision logic of a process planner and the mapping into an expert system, a knowledge database or into the program logic (Shome & Sinha, 2004). The coherences are illustrated in figure two.

It is fundamental for such a system to use high level product models to retrieve essential parameters automatically. Features are one step to provide models that can be integrated within different application areas but there are still many problems caused by the only geometric semantics of such features. Therefore we want to present STEP which offers an optimized feature modeling approach we want to use for automated process planning.

[FIGURE 2 OMITTED]

3. STEP

During the mid-eighties it was already recognized that there is a demand for standardized product data representation that should support the complete product lifecycle. Therefore the project ISO 10303--STEP was initiated. Since the beginning, there are several international teams working on this standard (Scheer, 1997). STEP is fundamentally based on application protocols, each of them satisfying special demands of special industries. The standard defines 40 application protocols and 22 are already declared as an international standard (SCRA, 2006). Application protocols are situated on the highest level of the STEP architecture and they define complex data models for the application specific description of products. We have identified the application protocol (AP) 224--Mechanical Product Definition for Process Planning Using Machining Features. This protocol supports the deployment of information required for manufacturing single piece parts through machining operations with high level features.

In contrast to design products with traditional geometric methods or general form features the modeling degree of freedom is very limited while using features defined by AP 224. This limitation causes the great advantage of this protocol: A part can only be modeled with features that allow the determination of suitable machining operations.

Based on the semantics of all features used in a product model you can check whether there are corresponding machines in your manufacturing environment that are able to manufacture all the required aspects of a part. Additionally we can divide features used to define a single piece part into dependent and independent features. Independent features can be employed directly to a given base shape (which represents the raw material).Against this, dependent features require to employ other features in the forefront. This could happen if a related surface is defined by another feature or the removal of volume is prevented by volumes that should be removed before. For instance a chamfer could be employed only, if the corresponding corner is already available. In this case we find a set of geometric dependencies which have to be checked before. These dependencies are then used for generating a process plan.

4. IMPLEMENTING STEP

Based on the recognition of missing implementations we went to implement a feature library for using STEP AP 224. The reason for this was the need to create design models based on AP 224 and the focus on the use of these models for the development of a process planning tool.

Therefore we have implemented an EXPRESS interpreter for the early binding approach which is fully independent from any application protocol and can be used for implementing all protocols of the STEP suite. The idea is to generate classes corresponding to the EXPRESS declaration including specialized logic for the mapping into a neutral exchange structure. This can be done by mapping entities into a STEP or XML file. Therefore we read an EXPRESS scheme and parse the syntax to build a logical structure e.g. of the entities, relations and properties.

Afterwards we can select a previously implemented source code generator. Currently we have implemented one generator for C# and one for generating an HTML documentation of an EXPRESS file. We can extend the generators also for generating classes e.g. for involving the C++ or JAVA programming language.

5. CONCLUSIONS

Through our researches we have realized that there is still a surprising lack of specialized implementations of STEP although there exists huge potential how we could use STEP to optimize the whole product lifecycle which is essential for mass customization of individual parts. As shown in our purposes the AP 224 has a great potential for implementing an automated process planning system for machining parts to support automated offers to a potential customer. Currently we have implemented a solution to get several application protocols of STEP into. To accept the methodology of the AP 224 in the industry the STEP implementation is not enough. We require persuasive applications like an easy going modeler based on AP 224 or our intended process planning system. Another problem still remains because AP 224 forces a change of thinking concerning to the way of modeling. Instead of designing a part by describing the final part shape the destructive approach starts with the description of a raw volume. The advantages are the huge correspondence to machining operations and the associated semantics of the features which are essential for automated process planning.

6. REFERENCES

Cross, N. (2000). Engineering Design Methods--Strategies for product design. 3rd Edition, John Wiley & Sons Lidt, Milton Keynes UK

Ehrlenspiel, K., Kiewert, K. & Lindemann, U. (2005). Kostengunstig entwickeln und konstruieren--Kostenmanagement bei der integrierten Produktentwicklung, 5th Edition, Springer, Berlin

Nasr, E. A. & Kamrani, A. K. (2006). Computer-Based Design and Manufacturing, 1st Edition, Springer, US

Scheer, A.-W.(1997). Wirtschaftsinformatik: Referenzmodelle fur industrielle Geschdftsprozesse, 7th Edition, Springer, Berlin

South Carolina Research Authority (2006). STEP Application Handbook--ISO 10303., 3rd Edition, SCRA/ISG, North Charleston

Shome, S. N., Basu, J. & Sinha, G. P. (2004). Proceedings of the National Conference on Advanced Manufacturing & Robotics, 1st Edition, New Delhi: Allied Publishers