alireza mokhtar and mahmoud houshmand* hand, cad and cam modules from the same vendors may have...

252 Int. J. Manufacturing Research, Vol.

Copyright © 2010 Inderscience Enterprises Ltd.

Introducing a roadmap to implement the Universal Manufacturing Platform using Axiomatic Design theory

Alireza Mokhtar and Mahmoud Houshmand* Industrial Engineering Department, Sharif University of Technology, P.O. Box: 11155-9414, Azadi Ave., Tehran, IR E-mail: [email protected] *Corresponding author

Abstract: In competitive environment of quick market changes, developing an interoperable manufacturing system is a necessity. To realise interoperability, the exchange of part and product data must be independent of their platforms. There are two basic approaches for this purpose:

i utilising interfaces

ii utilising neutral formats.

Standard for the Exchange of Product data (STEP) plays a significant role as a neutral data model to integrate design, manufacturing and other activities in product life cycle. To implement a Universal Manufacturing Platform (UMP), a systematic roadmap is required. The current paper proposes such a roadmap using the methodology of Axiomatic Design (AD). [Received 12 February 2009; Revised 2 July 2009; Accepted 25 October 2009]

Keywords: interoperability; AD; axiomatic design; UMP; universal manufacturing platform; CAD/CAM integration.

Reference to this paper should be made as follows: Mokhtar, A. and Houshmand, M. (2010) ‘Introducing a roadmap to implement the Universal Manufacturing Platform using Axiomatic Design theory’, Int. J. Manufacturing Research, Vol.

Biographical notes: Alireza Mokhtar is shortly going to finish his PhD in Manufacturing Systems at the Industrial Engineering Department of Sharif University, where he had received his BSc and MSc in Industrial Engineering. His favourite fields of research include CAD/CAM integration, STEP-based interoperability and machining processes. He also has several publications such as international journal papers, conference articles and book chapter.

Mahmoud Houshmand is a Faculty member of the Department of Industrial Engineering, Sharif University of Technology, Tehran, Iran, carrying research and teaching in the area of manufacturing systems design. He holds a PhD in Advanced Manufacturing Systems from Mechanical Engineering Department of University of Manchester Institute of Science and Technology, UK. His research interests include CAD/CAM/CAPP, design and

Introducing a roadmap to implement the Universal Manufacturing Platform 253

implementation of lean production systems using axiomatic design methodology. He is now leading a research team working on integrated and interoperable CAD/CAM/CAPP/CNC manufacturing systems based on STEP standard.

1 Introduction

An industrial enterprise needs to be quickly responsive and flexible to survive in today’s competitive markets. For this purpose, the information related to the activities such as design, manufacturing and production pertaining to the parts and products must easily and rapidly exchange among the different devices or computer systems. This exchange may also occur between the vendors providing different application protocols. In such a huge data transformation, information losses and misinterpretations between the departments, facilities and controllers would become unavoidable (Xu, 2005). The solution to tackle this problem has been widely introduced since a decade ago. Product data can be exchanged by employing standards like STEP, an international standard introduced in 1994 (ISO 10303-1, 1994). STEP stands for the Standard for Exchange of Product data model. Using EXPRESS, a neutral and universal language, all the information generated in a product life cycle could be presented, maintained and easily manipulated to support the integration (Proctor and Kramer, 1998). For example, in the environment of Computer-Aided Design, Manufacturing and Process Planning (CAD/CAM/CAPP), representation of part design features may not be identical. Therefore, different post-processors would be required to convert feature information to be compatible for the variety of machine tools. Avoiding post-processing, STEP-NC as an extension of STEP standard is able to provide a uniform data model and support an exchangeable data format (Xu and Newman, 2006). Considering manufacturing resources and environments, Newman and Nassehi (2007) have recently proposed an interoperable architecture platform called UMP. Although introduction to this platform is a big step towards the integration, it still needs to be practically developed to encompass the entire production activities. Moreover, in each manufacturing module, application and device layers are to be considered separately. In the current research, first the need for data integration in manufacturing environment is described. Second, UMP and its impediments as a solution for the entire production echelon would be highlighted. Then, one of the systematic design methodologies, entitled ‘AD Theory’, is utilised to generate a roadmap for implementation of UMP. The proposed roadmap includes all the production echelons such as manufacturing, manufacturing support systems and production control. However, the main focus in this work is on the first echelon, which deals with CAD/CAM/CAPP/CNC modules.

1.1 Integration and interoperability in manufacturing environment

The product life cycle includes all the activities required to deliver the product to the market. This life cycle includes design, manufacturing, inspection, maintenance and disposal (Joshi, 1999). The need for enterprise integration originates from the need for controlling and integrating the different functions in an enterprise (Gao et al., 2003). The aim of enterprise integration is to standardise diverse aspects of the product

254 A. Mokhtar and M. Houshmand

life cycle (Vernadat, 1996). On the other hand, each of the aforementioned activities in product life cycle may be accomplished in a different enterprise throughout the product supply chain (Jardim-Goncalves et al., 2006). Therefore, data exchange must be facilitated so that each party in the chain would be able to seamlessly communicate with its corresponding parties (White et al., 2005). Figure 1 shows how data exchange may occur between different manufacturing modules in a typical product supply chain and among the involving enterprises.

Figure 1 Modules of data communication in different enterprises (see online version for colours)

Each manufacturing enterprise may consist of a mixture of diverse echelons such as marketing, production and finance. In production echelon, there are three main levels including Manufacturing, Manufacturing Supports and Production Control, each of which can further be classified into other modules (Figure 2). Since in the current research, the main focus is on the first level of production echelon, data exchange in its consisting modules, CAD/CAM/CAPP/CNC, is discussed in Section 1.2.

Figure 2 Production echelon and its levels

1.2 Data exchange in CAx chain

Manufacturing is one of the significant modules in the production echelon of a product life cycle. CAx chain (CAD/CAM/CAPP/CNC) associates with part design, process planning and manufacturing tasks. Data integration among heterogeneous CAx systems has been widely researched (Fuh et al., 1996; Patil and Pande, 2002; Xu et al., 2005; Alvares et al., 2008). However, it has been universally accepted that using CAD data model cannot meet the requirements to realise the CAD/CAM integration. For instance, although machining features are the most appropriate representation of part information,


when two different vendors provide CAD and CAM systems, features data and parameters cannot be shared because they utilise different standards or protocols. On the other hand, CAD and CAM modules from the same vendors may have confliction since design features are not necessarily the same as manufacturing features (Miao et al., 2002). One approach to overcome such shortcoming is to benefit from a neutral data format to exchange the product data (Owen and Bloor, 1987; Qiao et al., 1993). Some of these formats are recognised as STEP, IGES, PARASOLID, STL, SET, VDA and ACIS (Li and Kota, 2001). Among all the neutral formats, STEP standard and its protocols gained more interest by the researchers as well as the practitioners (Bhandarkar et al., 2000; Asiabanpour et al., 2009). Unlike the other formats, STEP is able to specify the entire product data throughout the life cycle, independent of any particular device or application (Xu and He, 2004). Particularly, to exchange CAx data including features information, design, process planning and manufacturing, STEP has been extensively addressed by authors. STEP data model has been employed to extract the information of features (Bhandarkar and Nagi, 2000; Dereli and Filiz, 2002; Lau et al., 2005). In computer-aided design, STEP application protocols, AP203 and 214, provide all the information required to represent the part design specifications. AP 224, which entails mechanical feature definition, has been utilised to develop some process plan systems (Amaitik and Kilic, 2005). In other activities like inspection and process control, STEP has found numerous applications (Ali et al., 2005; Brecher et al., 2006; Kumar et al., 2007). Nonetheless, the majority of these researches considers prismatic components and reasonably deals with milling operations. STEP standard has also been utilised in turning processes for rotational parts (Heusinger et al., 2006). STEP AP 238 and ISO 14649, informally recognised as STEP-NC, have been extensively taken into account by researchers to integrate data in CAD/CAM/CNC environment (Mokhtar et al., 2008; Hardwick and Loffredo, 2006; Xu et al., 2005). STEP-NC presents a detailed and well-structured data model associating with feature-based programming, tools definition, machining technology and operations sequence (ISO 14649-1, 2003). A few attempts have been made in using STEP to generate tool trajectory information for CNC controllers (Suh et al., 2002, 2003).

For an interoperable CAx system, different platforms have been introduced. A platform generally consists of software and hardware structure that supports a variety of applications by enabling them to exchange information through a communication system without knowing the particulars of the destination application. In such platform, information system must be such that manufacturing resources and activities are easily communicated with each other (Iwata, 1997). For this purpose, standardisation in data transferring sounds unavoidable (Zimmerman et al., 2002). Figure 3 illustrates a fully STEP-compliant infrastructure for product data communication.

Figure 3 STEP-compliant manufacturing platforms


2 Data exchange in application and device layers

Considering Figure 2, in production echelon, two technical areas might be considered: software application and device area. Generally speaking, these two areas deal with software and hardware for CAD/CAM/CAPP systems. To realise the interoperability, two general approaches may be followed:

i using bidirectional interfaces or

ii using a neutral format.

In the former approach, an interface is designed and implemented to convert the input data from the source Application/Device (A/D) into the desired format for the target A/D. In the latter one, data could be exchanged between two sides without requiring any interfaces. Referring to Figure 1, when two CAD modules (especially from different enterprises) communicate their design specifications associating with an identical part and they meanwhile use different software applications, data exchange would become an intriguing issue. Confliction may arise owing to the loss of product or process information and incapability of one application programme to communicate with the other one. To resolve this conflict, two general solutions may be considered:

i Development of a software interface to translate data from one format into another one. In this case, both CAD modules can easily communicate through such interfaces.

ii Using a neutral format plug-in to the software application at both sides.

In this case, STEP-import/STEP-export capability might be added to both software applications. On the other side, depending on the kind of data communication (especially in e-manufacturing), a CAM file for example may directly be submitted from a computer post-processor to a CNC machine in the other enterprise, see Figure 4(b). Likewise, both enterprises should have capability of reading and manipulating the data in a bidirectional data flow. Frequently, the post-processors are not designed in the same company as the machine controllers are designed. To exchange data, the proposed approaches are:

i using a soft/hard interface compatible with both device and application programmes

ii using an intelligent controller to control devices and neutral format software to implement application programmes.

Implementation of device/device and device/application interfaces has already been experienced by some of the Japanese companies like Denso, Toyota, Mitsubishi, etc. The solution was a software development kit called ‘ORiN2’ (Mizukawa et al., 2000). The architecture provides two standardised software interfaces: Application Interface and Device Interface. It can also be extended by the add-in modules such as an application gateway and a provider (Figure 4(a)). The main purpose to employ such an interface was to provide a unified access from a PC to a lot of devices such as an industrial robot and a PLC.


Figure 4 (a) ORiN general architecture (Mizukawa et al., 2000) and (b) CAM file transferring between two enterprises

(a) (b)

In an architecture proposed by Newman and Nassehi, as shown in Figure 5, CAD/CAM systems exchange information using their interfaces such as CAD or CAM application interfaces. However, devices may not be able to exchange data with application programmes by means of an interface due to the type of computers used in devices. Moreover, most of the devices work in dusty and noisy environments of shop floor and this is a restriction in using a real computer as well as their controllers (Blecha and Bradac, 2003). Reasonably, they need their own controllers to be specifically designed for their individual applications and resistant to such inconvenient conditions. These new controllers must be intelligent enough to operate independent of any interface. Using STEP-NC neutral format, some CNC intelligent controllers have been developed to machine the simple part geometries (Lee et al., 2006; Suh and Cheon, 2002).

Figure 5 Application interface in UMP architecture (see online version for colours)

Source: Newman and Nassehi (2007)

Figure 6(a) and (b) displays schematic of different possible communications between device controllers and application programmes. Data communication between two programmes could be possible using a neutral data format or an interface programme. They are introduced as solutions A and B, respectively. This becomes practical when, for example, a CAD software exports the design data file into STEP format and the CAM application software imports it as an STEP file. Otherwise, an interface must interfere to read and write both specific formats from the two softwares. Such an interface is shown in Figure 6(b) as Application/Application (A/A) interface. Two different devices (like a CNC and Robot) can communicate in the same way (Device-to-Device interface). A/D interface may also be utilised when a computer collaborates with a device. However, to use an autonomous format like STEP and to avoid a convertor, their controllers must be fully compliant with STEP standard.


Figure 6 (a) Neutral format-based data communication, solution A and (b) interface-based data communication, solution B (see online version for colours)

(a) (b)

In production echelon, these types of communication between device controllers and application programmes are frequently happened. Alongside of numerous advantages of developing a universal platform to realise interoperability between device and application, one of the most challenging issues is intelligent controllers. Original equipment manufacturers can reduce the time they spend preparing data for their suppliers by as much as 75% if they can seamlessly share the design and manufacturing data in their databases (Suh et al., 2002). Nonetheless, for instance, CNC manufacturers need to be fully convinced with the reliability and perfectness of STEP to put the G-Code away. On the other hand, several leading CAM software companies have made it possible for STEP-NC files to be handled with their own softwares. This makes their users ready to participate in supply chains that are turning to global data exchange standards and to streamline the flow of digital information over the internet. In this paper, considering solutions A and B represented in Figure 6(a) and (b), and focusing interoperability in the manufacturing environment, a roadmap to implement the universal platform will be presented. To do this, the methodology of AD is applied. In Section 3, the principles of this theory will be briefly explained.

3 Axiomatic Design theory

In the late 1970s, based on his industrial experiences, Suh (1990) started introducing a methodology to facilitate the design procedure, which is now known and called as the AD theory. His main target was to establish solid and scientific fundamentals for product


design and manufacturing. In AD, to establish a design problem, at the first stage, to satisfy the perceived needs, the design goals must be set. At later stage, the designs for the defined requirements would be provided among which the best design is chosen (Suh, 2001). These activities occur between and in different scopes such as customer, function and design domains (Kim et al., 1991). The customer domain is wherein the Customer Needs (CNs) are specified. These needs must be mapped into the functional domain where they are translated into a set of Functional Requirements (FRs). In design domain, Design Parameters (DPs) are selected to control/satisfy the FRs. Decomposition in each domain cannot be performed autonomously of the developing hierarchies in the other domains and should follow a zigzag pattern between adjacent domains (Albano and Suh, 1994). The relationship between FR and DP domains along with the design matrix (DM) to represent how a DP can satisfy an FR is illustrated in Figure 7. The corresponding equation is expressed as,

, FR / DP .i j i jA = ∂ ∂ (1)

In AD, the FRs must be defined independently and the best design is the one with the minimum content of information. These axioms are recognised as Independence Axiom and Information Axiom, respectively (Suh, 1990). In addition to these two axioms, eight corollaries, 26 general theorems, and 14 specific theorems have been gradually developed to guide and evaluate the design problems (Thielman and Ge, 2006).

Figure 7 FRs and DPs in Axiomatic Design theory

4 AD in design problems

Axiomatic Design (AD) is quite excellent in that multiple FRs can be methodically handled, whereas most design methodologies involve with only one FR (Shin et al., 2006). During the recent decade, this methodology has been generally used in three domains of

i design for manufacture

ii manufacturing processes

iii manufacturing systems (Shirwaiker and Okudan, 2008; Kulak et al., 2005).

However, to challenge with real problems, product and enterprise requirements have been considered separately (Yien, 1998). But, it is not very well understood how the theory can work out to implement a roadmap for production systems. Cochran (1994) tackles manufacturing systems design problem in terms of AD. In the practical point of view, managers as well as engineers are inclined to focus on the solutions before the requirements are fully stated and realised (Almstrom, 2005). An analogy among different


design problems including system design, which is similar to roadmap design, has been discussed by Gebala and Suh (1992). Although this methodology is very useful to model the real design problems, explicit ontology definition and implementation of axioms measures are yet to be well thought-out by practitioners (Ge, 2001).

5 Mapping UMP with AD

On the basis of what explained in Section 2 about the basic specifications of a UMP for data integration, the FRs for implementing such a platform are defined. Then, DPs to satisfy these FRs will be provided and to represent the relationship between these two domains, DMs are determined. In the axiomatic methodology terms, the top level is the main goal of the problem and labelled as FR0. At this level, to achieve the goal of an integrated manufacturing platform, FR0 is defined as follows:

FR0: Efficient communication throughout the enterprise production echelon.

To meet this requirement, the first level in the design domain is set as follows:

DP0: Implementation of the Universal Manufacturing Platform (UMP).

Production echelon encompasses three different levels (Figure 2). Therefore, FR0 is classified into three levels as follows:

FR1: Product data integration in manufacturing level

FR2: Product data integration in manufacturing supports level

FR3: Product data integration in production control level.

To satisfy these FRs, decomposition of DP0 results in the following DPs:

DP1: Implementation of an integrated platform in CAx chain

DP2: Implementation of an integrated platform in Assembly and Automation

DP3: Implementation of an integrated platform in production applications.

As mentioned in Section 2.1, in each level of the production echelon, one or two technological layers might be considered. In manufacturing and manufacturing supports levels, both application and device layers must be covered by the proposed platform while in production control level, product data exchange will be dealt with only in application layer. Consequently, the set of subsequent FRs along with their corresponding DPs are as follows (Figures 8 and 9):

FR11: Integration in Application to Application → DP11: Utilising the direct data exchange

FR12: Integration in Device to Application → DP12: Utilising the convertor interface

FR13: Integration in Device to Device → DP13: Utilising the intelligent controllers

FR21: Integration in Application to Application → DP21: Utilising the direct data exchange


FR22: Integration in Device to Application → DP22: Utilising the convertor interface

FR23: Integration in Device to Device → DP23: Utilising the intelligent controllers

FR31: Integration in Application to Application → DP31: Utilising the direct data exchange.

Figure 8 FR’s decomposition (see online version for colours)

Figure 9 DP’s decomposition (see online version for colours)

As previously discussed, in this paper we mainly consider the integrated platform in manufacturing level and since the other two levels almost follow the same solutions represented in Figure 6, they would be disregarded for further analysis. DPs for FRij (i > 1) are the same as DPs for FRij (i = 1). Hence, hereafter, manufacturing level including all the activities in CAx chain will be investigated. Figure 10 demonstrates how FRs (for i = 1, j = 1, 2, 3) depend on their corresponding DPs.

To mathematically represent the relationship between FRs and DPs, DMs are given by,

1 11 12 13 1

2 21 22 23 2

3 31 32 33 3

FR DPFR DP ; 0 for FR DP

ij

A A AA A A A i jA A A

= × = ≠

(2)


11 11 12 13 11

12 21 22 23 12 13 21 31

13 31 32 33 13

FR DPFR DP ; ( 0);FR DP

0{FR } 0 {DP }.

0ij ij

B B BB B B B B BB B B

X XX XX X

= × = = =

=

(3)

Figure 10 Relationship between FR and DP domains, a coupled design

To satisfy the independence axiom, the DM must be decoupled or uncoupled. Coupled designs cannot be accepted since they violate the first axiom (Suh, 2001). Therefore, considering the diagonal matrix A, the set of FRi=1,2,3 is fully independent and the first axiom has been fulfilled or the design is uncoupled. At the next stage, as shown by matrix B, it is observed that this design is coupled because the matrix is not triangular. Since based on AD, coupled designs are undesirable; matrix B must be converted to a decoupled one. The matrix would become a triangular one in two different ways: setting B32 = 0 and obtaining an upper triangular matrix or setting B12 = B23 = 0 and obtaining a lower triangular matrix. In fact, eliminating B32 implies that in integration of device to device, utilising a convertor interface is not allowed (∂FR13/∂DP12 = 0). In the other way, applications cannot use convertor interface to communicate (∂FR11/∂DP12 = 0) while integrating the applications with device would not be applicable (∂FR12/∂DP13 = 0). Consequently, to decouple this design, following the first way seems to be more reasonable since less change would be applied in the manufacturing system (Suh, 2001). Therefore, the new set of equations of FRs is as follows:

11 11 11 12 12 13 13FR *DP *DP *DPB B B= + + (4)

12 22 12 23 13FR *DP *DPB B= + (5)

13 33 13FR *DP .B= (6)

This means by satisfying FR13, FR12 and FR11, respectively, the independence axiom becomes fulfilled and the uncoupled design alters to a decoupled one. The new FR–DP relationships are illustrated in Figure 11.


Figure 11 Relationship between FR and DP domains, a decoupled design applying the first axiom

6 Implementation of UMP, the proposed roadmap

In this section, a typical roadmap is presented to designate how a UMP may be implemented. As shown in Figure 11, a decoupled design is the outcome of applying ADs independence axiom. Considering the design equations, the first two FRs could be fulfilled by employing any of the two DPs; FR11 could be satisfied by DP11 or DP12 and FR12 could be satisfied by DP12 or DP13. The third functional requirement, (FR13), can only be fulfilled by DP13. To demonstrate the roadmap in practice, a simple production cell consisting of machine tools and material-handling devices supported by application software is taken (Figure 12). In this cell, the parts are designed by CAD software programme and their process plans are prepared by CAPP application software. To manufacture the parts, boring and face milling operations are carried out on CNC machine tools and an industrial robot transfers the finished products. The operator prepares a CAD file of the parts using a computer-aided design application software programme uploaded on computer A. The CAD file is submitted to computer B wherein a process-planning software is uploaded and the final process plan is automatically generated. The output as a piece of data is to be first transferred to the CNC milling machine 1. Upon completion of boring task in this machine, the semi-finished part is completed on the CNC milling machine 2. The information of boring and milling operations are required for the succeeding operations. For instance, in the illustrated part, after boring operation of the cylinder, the related data on-machine inspection should be transferred to the second CNC controller to select the desired cutting tool. On the other hand, sometimes it would be necessary to send the information of inspection to the robot, which is waiting to pick the part up and drop it on the checked items conveyor. The defected items, referring to these data, need to be stored in a separate container.

To exchange data from the CAD file to the process-planning application software and vice versa, two approaches are applicable:

i Direct data exchanging (DP11). In this approach, the first step is to export the CAD data file to STEP file (i.e., Application Protocol 203) in the form of Part 21 file (ISO 10303). To do this, a feature must be embedded in CAD software to save the draft file as a .stp extension file. On the other side, computer B should be able to read and import the STEP CAD file as well as having the possibility of sending it back to computer A if any design modification is required.

ii Using an interface that is able to convert the design specification and part tolerance to a desirable format for computer B (DP12).


In case of any modification and feedback from the process planning to design stage, this interface should be able to work bilaterally. This means the exchange of data within CAD and CAM application software programmes is feasible either by direct exchange of data via STEP or utilising an interface system. To exchange manufacturing data from CAPP application software programme to CNC machine tools, there are two possibilities:

i the exchange of data via an interface

ii direct exchange of data.

Figure 12 Data exchange in a typical production echelon, the shop floor level (see online version for colours)

For example, consider the manufacturing system shown in Figure 12, manufacturing information in terms of machining process plan is passed to the first CNC machine tools for boring operation. The CNC machine tool must be able to read the data received from Computer B and send it back if needed. Therefore, an interface (DP12) or an intelligent controller (DP13) could be employed to realise such collaboration. The intelligent controller is an STEP-compliant controller, which has the capability of loading the process-planning file in the format of STEP standard (i.e., Application Protocol 238) and transform it to tool trajectory codes required for cutting operations and other NC functions. This means the exchange of data within devices and application software is feasible either by applying an interface system (DP12) or directly by an intelligent controller (DP13). As realised by the decoupled design, Figure 11, to integrate the CNC milling machine tools and the industrial robot, the only way would be utilising the STEP-based intelligent controller. It is required to select one of the two approaches presented earlier to satisfy FR11 and FR12. To select one, a quantitative comparison may be performed based on the second axiom of the AD theory. The selected design (approach) would be the one with the least content of information. To satisfy this axiom, cost of data exchange is considered as the comparison criteria. On the basis of the AD theory, a system and desired design intervals are to be measured and the overlaps should


be quantitatively determined (Figure 13). The total content of information for a design having n FR is calculated through equation (7) (Suh, 2001).

21

Info log ( / ) ; (SI, DI and CI denote the system,

Design and their common intervals, respectively).

n

T ii

SI CI=

=∑ (7)

In this case, as shown in Figure 13, the calculation of InfoT is to be done for FR11 and FR12. This will result in searching for the minimum content of information of DP11, or DP12 for FR11, and DP12 or DP13 for FR12, correspondingly. The minimum of InfoT corresponds to the solution, which requires the least possible change in the initial design (Suh, 2001). Apparently, the value of log2 (SI/CI) would become the minimum when the design and system intervals, DI, SI have through overlap equation (7). To satisfy the FR11, i.e., the integration of application software programmes, there are several commercial software packages supporting STEP-in/STEP-out feature. They export CAD STEP file in one side and importing it in the same format in the other side to meet the requirement of CAD/CAPP data integration. The bidirectional flow of product information will also be applicable if CAPP software could send the file back in STEP format. In this approach, an invisible software interface makes the integration in application-device format. PRO/E, CATIA and STEP-Developer are the successful examples of exporting CAD in .stp extension while the last one is also supporting AP224 mechanical definition of features for process planning (ISO 13030-224, 2001). In CAPP module, the aforementioned solutions might be employed to retrieve, manipulate and manage the process-planning file. In CAPP/CAM/CNC side, the existing solutions based on interface data converter include ORiN2 and Elysium (Dean, 2005). Finally, to satisfy FR13, according to the first axiom, using the intelligent controllers, like the ones that have already been developed by Suh et al. (2002, 2003), sounds feasible. The final arrangement of FRs and DPs is depicted in Figure 14. This fully uncoupled design promises an impeccable solution based on AD theory.

Figure 13 Design, system and common intervals

Finally, it should be mentioned that a variety of stakeholders of such UMP will definitely benefit communicating in an interoperable environment. Designers, CAD/CAM engineers as well as CNC operators will experience much less iterations and reworks with the upstream and downstream stages. Moreover, utilising the intelligent controllers, CNC will operate bidirectionally and its operator enjoys manipulating a data model (e.g., STEP AP238) rather than struggling with complicated G-Codes to make further modification of process plans.


Figure 14 Recommended DPs applying the second axiom

7 Discussions and conclusions

In this paper, using the methodology of AD, a roadmap to implement the integrated manufacturing platform was introduced and verified through a case study. Referring to the decoupled design shown in Figure 11, two application programmes are able to seamlessly communicate when they utilise a convertor interface or the direct data exchange. This dilemma led to a final selection after applying the second axiom of AD. Interoperability in device to application is also realised by employing a convertor interface or an intelligent controller. The only requirement that is fulfilled by one DP is integrating the two device layers. As discussed in Sections 1 and 2, the acceptable neutral format to be used as a direct data exchange is STEP and its relevant application protocols. On the other hand, STEP-based CNC controllers and STEP-based Robots are yet to be developed and invested in near future to realise the use of intelligent and format-independent controllers in device side. While developing comprehensive and integrative convertor interfaces like ORiN2 is mostly followed by the practitioners, fully STEP-based communication will become more efficient in the future. This is due to its simplicity and integrity along with the huge cost and time for developing numerous interfaces. Despite these advantages, utilising adaptive controllers in CNC robots and other devices is a challenging issue. CNC adaptive controllers are to be installed so that bidirectional data streamlines would be applicable. They need an almost huge investment to be designed and embedded. However, this would become economically justified when the number of CNCs using STEP-in/STEP-out controllers notably increase. In this case, comparing the solutions based on the second axiom of AD, it would be more reasonable to implement the fully STEP-based UMP. As mentioned earlier, the main challenge is still to invest on the new controllers for a variety of devices and equipments such as Robots, Conveyors and PLCs. In this paper, the goal of an interoperable manufacturing system is stipulated as efficient data communication in manufacturing level and then decomposed to some other requirement, which is more tangible in CAx modules. To achieve this, DPs are then developed based on AD methodology and these results in a design problem. The roadmap is eventually proposed as the solution of the corresponding problem. Another potential field of developing AD theory in this application could be enhancing it with Taguchi method (Renbin and Xianfu, 2008). This may lead to a robust design by making connection between the information content and quality loss function. Finally, it should be mentioned that the universal platform boundaries may go further to


capture the entire organisational activities. Although marketing, product development, sales, etc., are recognised as the crucial functions requiring data integration and interoperability, they were not dealt with in this work.

References Albano, L. and Suh, N.P. (1994) ‘Axiomatic design and concurrent engineering’, Computer-Aided

Design, Vol. 26, No. 7, pp.499–504. Ali, L., Newman, S.T. and Petzing, J. (2005) ‘Development of a STEP-compliant inspection

framework for discrete components’, Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, Vol. 219, No. 7, pp.557–563.

Almstrom, P. (2005) Development of Manufacturing Systems A Methodology Based on Systems Engineering and Design Theory, PhD Thesis, ISBN 91-7291-634-6, University of Chalmers, Sweden.

Alvares, A.J., Ferreira, J.C.E. and Lorenzo, R.M. (2008) ‘An integrated web-based CAD/CAPP/CAM system for the remote design and manufacture of feature-based cylindrical parts’, Journal of Intelligent Manufacturing, Vol. 19, No. 6, pp.643–659.

Amaitik, S.M. and Kilic, S.E. (2005) ‘STEP-based feature modeller for computeraided process planning’, Int. J. Prod. Res., Vol. 43, pp.3087–3101.

Asiabanpour, B., Mokhtar, A., Hayasi, M. and Kamrani, A. (2009) ‘An overview on five approaches for translating CAD data into manufacturing information’, Journal of Advanced Manufacturing Systems, Vol. 1, No. 8, pp.1–26.

Bhandarkar, M.P. and Nagi, R. (2000) ‘STEP-based feature extraction from STEP geometry for Agile Manufacturing’, Journal of Computers in Industry, Vol. 41, pp.3–24.

Bhandarkar, M.P., Downie, B., Hardwick, M. and Nagi, R. (2000) ‘Migrating from IGES to STEP: one to one translation of IGES drawing to STEP drafting data’, Computers in Industry, Vol. 41, No. 3, pp.26–277.

Blecha, R. and Bradac, Z. (2003) ‘Distributed control system for robotic manipulators’, Proceedings of the IEEE International Conference on Industrial Technology, Vol. 1, pp.169–172.

Brecher, C., Vitr, M. and Wolf, J. (2006) ‘Closed-loop CAPP/CAM/CNC process chain based on STEP and STEP-NC inspection tasks’, International Journal of Computer Integrated Manufacturing, Vol. 19, No. 6, pp.570–580.

Cochran, S.D. (1994) The Design and Control of Manufacturing Systems, PhD Thesis, Auburn University, Auburn.

Dean, A. (2005 December) ‘Elysium™ CAD doctor 5.2’, MCAD Magazine, EDA Publications. Retrieved July 28, 2007, from http://www.mcadonline.com/index.php?option=com _content&task=view&id=179&Itemid=1)Dean A, (2007), Elysium CAD doctor 5.2, MCAD magazine, EDA publication.

Dereli, T. and Filiz, H. (2002) ‘A note on the use of STEP for interfacing design to process planning’, CAD Computer Aided Design, Vol. 34, No. 14, pp.1075–1085.

Fuh, J.Y.H., Chang, C-H. and Melkanoff, M.A. (1996) ‘The development of an integrated and intelligent CAD/CAPP/CAFP environment using logic based reasoning’, Computer Aided Design, Vol. 28, pp.217–232.

Gao, J.X., Aziz, H., Maropolous, P.G. and Cheung, W.M. (2003) ‘Application of product data management technologies for enterprise integration’, Int. J. Comput. Integr. Mfg., Vol. 16, pp.491–500.

Ge, P. (2001) Collaborative Negotiation for Early Stage Parametric Design of Mechanical Systems, PhD Thesis, University of Southern California, Department of Mechanical Engineering, Los Angeles, CA.


Gebala D.A. and Suh, N.P. (1992) Research in Engineering Design Theory, Applications, and Concurrent Engineering, Vol. 3, Springer-Verlag, New York Inc., pp.149–162.

Hardwick, M. and Loffredo, D. (2006) ‘Lessons learned implementing STEP-NC AP-238’, International Journal of Computer Integrated Manufacturing, Vol. 19, No. 6, pp.523–532.

Heusinger, S., Rosso Jr., R.S.U., Klemm, P., Newman, S.T. and Rahimifard, S. (2006) ‘Integrating the CAx process chain for STEP-compliant NC manufacturing of asymmetric parts’, International Journal of Computer Integrated Manufacturing, Vol. 19, No. 6, pp.533–545.

ISO 10303-1 (1994) Industrial Automation and Systems and Integration-Product Data Representation and Exchange, Part 1: Overview and Fundamental Principles.

ISO 13030-224 (2001) Industrial Automation Systems and Integration; Product Data Representation and Exchange, Part 224. Application Protocol: Mechanical Product Definition for Process Plans Using Machining Features.

ISO 14649-1 (2003) Data Model for Computerized Numerical Controllers, Part 1: Overview and Fundamental Principles.

Iwata, K. (1997) ‘Virtual manufacturing systems as advanced information infrastructure for integrating manufacturing resources and activities’, Annals of the CIRP, Vol. 46, No. 1, pp.335–338.

Jardim-Goncalves, R., Figay, N. and Steiger-Garcao, A. (2006) ‘Enabling interoperability of STEP application protocols at meta-data and knowledge level’, International Journal of Technology Management, Vol. 36, No. 4, pp.402–421.

Joshi, S. (1999) ‘Product environmental life-cycle assessment using input-output techniques’, Journal of Industrial Ecology, Vol. 3, Nos. 2–3, pp.95–120.

Kim, S-J., Suh, N.P. and Kim, S-G. (1991) ‘Design of software system based on axiomatic design’, CIRP Annals – Manufacturing Technology, Vol. 40, No. 1, pp.165–170.

Kulak, O., Durmusoglu, M.B. and Tufekci, S. (2005) ‘A complete cellular manufacturing system design methodology based on axiomatic design principle’, Computers and Industrial Engineering, Vol. 48, No. 4, pp.765–787.

Kumar, S., Nassehi, A., Newman, S.T., Allen, R.D. and Tiwari, M.K. (2007) ‘Process control in CNC manufacturing for discrete components: a STEP-NC compliant framework’, Robotics and Computer-Integrated Manufacturing, Vol. 23, No. 6, pp.667–676.

Lau, H.C.W., Lee, C.K.M., Jiang, B., Hui, I.K. and Pun, K.F. (2005) ‘Development of a computer-integrated system to support CAD to CAPP’, International Journal of Advanced Manufacturing Technology, Vol. 26, Nos. 9–10, pp.1032–1042.

Lee, W., Bang, Y-B., Ryou, M.S., Kwon, W.H. and Jee, H.S. (2006) ‘Development of a PC-based milling machine operated by STEP-NC in XML format’, International Journal of Computer Integrated Manufacturing, Vol. 19, No. 6, pp.593–602.

Li, Z. and Kota, S. (2001) ‘Virtual prototyping and motion simulation with ADAMS’, J. Comput. Inf. Sci. Eng., Vol. 1, No. 3, pp.276–280.

Miao, H.K., Sridharan, N. and Shah, J.J. (2002) ‘CAD–CAM integration using machining features’, International Journal of Computer Integrated Manufacturing, Vol. 15, No. 4, pp.296–318.

Mizukawa, M., Matsuka, H., Koyama, T. and Matsumotom, A. (2000) ‘ORiN: open robot interface for the network, a proposed standard’, International Journal of Industrial Robot, Vol. 27, No. 5, pp.344–350.

Mokhtar, A., Xu, X. and Lazcanotegui, I. (2008) ‘Dealing with feature interactions for prismatic parts in STEP-NC’, Journal of Intelligent Manufacturing, Vol. 20, No. 4, pp.431–445.

Newman, S.T. and Nassehi, A. (2007) ‘Universal manufacturing platform for CNC machining’, CIRP Annals, Vol. 56, No. 1, pp.459–462.

Owen, J. and Bloor, M.S. (1987) ‘Neutral formats for product data exchange: the current situation’, CAD Computer Aided Design, Vol. 19, No. 8, pp.346–443.


Patil, L. and Pande, S.S. (2002) ‘An intelligent feature-based process planning system for prismatic parts’, International Journal of Production Research, Vol. 40, No. 17, pp.4431–4447.

Proctor, F.M. and Kramer, T.R. (1998) ‘Feature-based machining system using STEP’, Proceedings of SPIE – The International Society for Optical Engineering, Vol. 35, No. 18, pp.156–163.

Qiao, L-H., Zhang, C., Liu, T-H., Ben Wang, H-P. and Fischer, G.W. (1993) ‘A PDES/STEP-based product data preparation procedure for computer-aided process planning’, Computers in Industry, Vol. 21, No. 1, pp.11–22.

Renbin, X. and Xianfu, C. (2008) ‘An analytic approach to the relationship of axiomatic design and robust design’, International Journal of Materials and Product Technology, Vol. 31, Nos. 2–4, pp.241–258.

Shin, M-K., Kim, Y-I., Kang, B-S. and Park, G-J. (2006) ‘Design of an automobile seat with regulations using axiomatic design’, Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering, Vol. 220, No. 3, pp.269–279.

Shirwaiker,·R.A. and Okudan, G. (2008) ‘Triz and axiomatic design: a review of case-studies and a proposed synergistic use’, J. Intell. Manuf., Vol. 19, pp.33–47.

Suh, N.P. (1990) The Principles of Design, Oxford University Press, New York. Suh, N.P. (2001) Axiomatic Design: Advances and Automations, Oxford Univ. Press, New York. Suh, S.H., Cho, J.H. and Hong, H.D. (2002) ‘On the architecture of intelligent STEP-compliant

CNC’, Int. J. Comput. Integr. Mfg., Vol. 15, pp.168–177. Suh, S.H., Lee, B.E., Chung, D.H. and Cheon, S.U. (2003) ‘Architecture and implementation

of a shop-floor programming system for STEP-compliant CNC’, Computer Aided Design, Vol. 35, No. 12, pp.1069–1083.

Suh, S-H. and Cheon, S-U. (2002) ‘A framework for an intelligent CNC and data model’, International Journal of Advanced Manufacturing Technology, Vol. 19, No. 10, pp.727–735.

Thielman, J. and Ge, P. (2006) ‘Universal linking of engineering objects’, Journal of Engineering Design, Vol. 17, No. 1, pp.1–16.

Vernadat, F.B. (1996) Enterprise Modeling and Integration, Chapman and Hall, London. White, A., Daniel, E.M. and Mohdzain, M. (2005) ‘The role of emergent information technologies

and systems in enabling supply chain agility’, International Journal of Information Management, Vol. 25, No. 5, pp.396–410.

Xu, X. and He, Q. (2004) ‘Striving for a total integration of CAD, CAPP, CAM and CNC’, Robotics and Computer Integrated Manufacturing, Vol. 20, No. 2, pp.101–109.

Xu, X. and Newman, S.T. (2006) ‘Making CNC machine tools more open, interoperable and intelligent; A review of the technologies’, Computers in Industry, Vol. 57, No. 2, pp.141–152.

Xu, X.W. (2005) ‘Novel surface and volumetric feature interactions for process planning’, International Journal of Computer Applications in Technology, Vol. 24, No. 4, pp.185–194.

Xu, X.W., Wang, H., Mao, J., Newman, S.T., Kramer, T.R., Proctor, F.M. and Michaloski, J.L. (2005) ‘STEP-compliant NC research: the search for intelligent CAD/CAPP/CAM/CNC integration’, Int. J. Prod. Res., Vol. 43, pp.3703–3743.

Yien, J.T. (1998) Manufacturing Systems Design Methodology, PhD Thesis, Hong Kong University of Science and Technology, Hong Kong.

Zimmerman, J.U., Haasis, S. and van Houten, F.J.A.M. (2002) ‘Session on design (Dn)-Dn/l ULEO-Universal Linking of Engineering Objects’, Annals of the CIRP, Vol. 51, No. 1, pp.99–102.

alireza mokhtar and mahmoud houshmand* hand, cad and cam modules from the same vendors may have...

Documents