Methodology and platform for integrated product design.

Banciu, Felicia ; Draghici, George ; Mazilescu, Crisanta Alina 等

1. INTRODUCTION

The products design process represents an ensemble of activities where the needed tools in order to attain the objectives, established accordingly to imposed requests, are developed and selected. Starting from a knowledge synthesis about the products design models, methods and tools briefly presented in chapter two, the objective is to elaborate a new design model, a methodology and an integrated, collaborative platform that will use the methodology based on the new design model.

We propose a holistic model, combining Pahl and Beitz's systematic model (based on design phases) and Nam Suh's axiomatic design model (based on design domains). The methodology is based on this model and the corroboration of these two models is realized through an algorithm that integrates the used design methods and tools, using a design for six sigma approach.

We considered that it is important to use also in the algorithm proposed in (Yang & El-Haik, 2003), the functional analysis, that is a common design method for both axiomatic and systematic approaches. The collaborative, integrated design platform comprises the software tools that assist the proposed methodology.

2. DESIGN MODELS

In specialized references there are many design models. The most representative are:

Models based on phase notion: Pahl & Beitz, French, Roozenburg & Eekels, March, Ulrich & Eppinger, Ullman, Pugh. These models are based on a sequential decomposition of design process, they are based on a design seen as a hierarchical, sequenced phases, are considered prescriptive models and are based on the designers own experiences.

Models based on activity notion: EVAD, Purcell, Girod, M3M. Models based on activity notion have as core the observation of the activities developed by designers during the design process.

Models based on domain notion: FBS, Rationale Design, North Europe, Value Model, Axiomatic Design. At this type of design models, we notice the existence of at least two domains. The domains are customer, functional requirements physical and process. The domain notion is a general one depending on the specific design model.

Hybrid Models: O'Sullivan, Co-Evolutionary, Evolutionary. These models are based on sequential, simultaneous, based on the concept of phase, activities and domains models evolution. This presentation's aim was to analyse the design models that can form a starting point for developing a new design model.

3. HOLISTIC MODEL

In order to develop a new design model were selected the Pahl and Beitz's systematic model (based on design phases) (Pahl & Beitz, 1996) and Nam Suh's axiomatic design model (based on design domains) (Suh, 1990). Generally, the algorithmic approach is founded on the notion that the best way to advancing in design field is to understand the process by following the best design practice.

Based on scientific rationale principles derived from the observing of successful design processes and projects, axiomatic design is characterized through its generality, it can be applied in all design fields, its rules are the same and the guidelines on how to make axiomatic design are given by the design axioms.

The proposed holistic model (Draghici & Banciu, 2007) uses as support the two fundamental models. The corroboration of these two models is realized through an algorithm that integrates the used design methods and tools, using a design for six sigma approach.

The main design methods comprised in holistic design model are Quality Function Deployment (QFD), Functional Analysis (FA), Theory of Inventive Problems Solving (TRIZ), Failure Modes and Effects Analysis (FMEA).

The functional analysis is a common design method for both axiomatic and systematic approaches and is contributing to take an account of customer's needs. TRIZ can be used to the decoupling of the design matrix in Axiomatic Design, when a technical contradiction is present in a given problem. It can be used also the su-field analyze in analyzing the open and closed streams in Internal Functional Analysis. FMEA is used in the design phase to verify a new or a modified system with a known behaviour. FMEA has in view to emphasise the possible failures aiming the improving of the design process and a technical solution validation. To apply FMEA in the system's functioning or using phases, it is necessary to know the functions for each phase in order to analyse next the failure risks, in this case FA becoming useful.

[FIGURE 1 OMITTED]

These design methods used separately, are not fully useful only if they are integrated, adapted to the need, in global design and quality approaches. Even their objectives are different; they can be applied to reach a common objective--to improve the design process. In this way, we consider that these design methods are complementary.

4. DFSS ALGORITHM

The DFSS algorithm that integrates these methods and tools has as aim to develop design entities corresponding to the expressed needs for the entire product's lifecycle at a six sigma quality level (Yang & El-Haik, 2003).The 14 steps are grouped in four phases: I-Identify, C-Characterize, O-Optimize and V-Validate. The phase represents a set of project's design activities and is limited by inputs and outputs.

The identify phase comprise the form team and determine customer expectations steps. The characterize phase comprises the following steps: understand FRs evolution (TRIZ-analyze and derive concepts), select best concept and analyze the functional structure of selected concept. It follows the optimize phase comprises the uncouple or decouple selected concept, simplify using design axiom 2, initiate scorecards and transfer function development, FMEA/PFMEA Asses Risk, DOE transfer function organization, Design for X, Tolerance design steps and finally the validate phase that comprises pilot/prototype design, validate design, launch mass production and celebrate successful completion steps.

The above presented algorithm (Yang & El-Haik, 2003) follows an axiomatic design approach and uses QFD, TRIZ, FMEA. We considered that it is important to use also the functional analysis, a common design method for both axiomatic and systematic approaches. It is contributing to take an account of customer's needs (QFD), to know the functions for each product's using or functioning phase in order to analyze next the failure risks (FMEA).

5. FUNCTIONAL ANALYSIS

According to the AFNOR X50-150 standard (AFNOR, 1988): "Functional Analysis is an action which consists of searching, set in order, characterization, hierarchy and/or set a value (quantifying) for the functions". Connected to functional analysis it has to be specified that there are many methods of Functional Analysis (FA) that are more or less adapted to the analysis that we want to realize. Functional Analysis can be seen as a bridge between the needs expressed by the customers and the functions that have to render the products, processes, or organizations. Through these features, FA is essentially a way to conduct to the obtaining of the products' quality and it can be realized for every specific utilization context of the studied object. Functional Analysis has two possibilities to be unfolded: External Functional Analysis and Internal Functional Analysis.

FA acts on the relation customer-purveyor defining the needs and the exigencies to satisfy. In this stage of need's formalization, the services that has to be accomplished are emphasized, referring to objective data regarding the product's use and the product's design will result from this analyze. The system (product) is considered taking into account its finality; it takes into account all the factors regarding the system and its environment, emphasise the product's quality through its value expression for the customer. Thus, consulting the customer's objective and subjective expectations result what they wants and using the functional analysis, it can be used the qualitative parts of value criterions to complete the functions. Using next the internal functional analysis the product can be structured and organised in a functions ensemble (how). Functional analysis take an account of customer's needs that are part of the first matrix of the first phase of QFD (Quality Function Deployment). In this phase, QFD dwells on the relation between customer's expected needs and different technical solutions that can conduct to customer's needs satisfaction.

In this way the functional analysis will help to the clearly defining of the functional requirements.

6. DESIGN PLATFORM

The collaborative integrated design platform contents the software tools that assist the proposed methodology. The platform contains two software tools provided by TDC Software (***, 2009) and Axiomatic Design Solutions Inc. (***, 2009). The TDC Software Products allows the integration of the functional analysis in collaborative integrated design platform. TDC Need tool helps to classifies, to give value to the generic functional requirements for a specific need and next using an internal functional analysis to find the main functions, the constraint functions with their importance degree. The next step is to use them as entrances in TDC Structure module the result being a more detailed functional structure and next the solutions (ways to provide the functions) list. Acclaro DFSS is an axiomatic design tool that supports a better concept and design process development through its hierarchical decomposition in the four domains: customer needs domains, functional requirements, physical domain and process domain, and modules like those that risk analyses module, cost analyses module, FMEA module.

7. CONCLUSION

The proposed research methodology refers to corroborate the Pahl and Beitz's systematic model and Nam Suh's axiomatic design through an algorithm that integrates the common design methods and tools, using a design for six-sigma approach. We considered that it is important to use also the functional analysis, a common design method for both axiomatic and systematic approaches, and in this way a new design algorithm will be proposed in our further research work and also we will apply this proposed methodology to a design task, The proposed methodology will be assisted by the integrated, collaborative product design platform, which comprises software tools provided by TDC Software (http://www.tdc.fr) and Axiomatic Design Solutions, Inc. (http://www. axiomaticdesign.com).

8. REFERENCES

AFNOR (1988), Recommandation pour obtenir et assurer la qualite en conception. Norme X50-127, AFNOR, Paris

Draghici, G. & Banciu, F. (2007). Parallelism Between Algorithmic (Systematic) Design and Axiomatic Design, Annals of the Oradea University, Fascicle of Management and Technological Engineering, Volume VI (XVI), ISSN 1583--0691, pp.1457-1468

Pahl, G. & Beitz, W. (1996). Engineering Design, Second edition, Springer, ISBN 3-540-19917-9

Suh, N. P. (1990). The Principles of Design, Oxford University Press, ISBN 0-19-504345-6

Yang, K. & El-Haik, B. (2003). Design for Six Sigma--A Roadmap for Product Development, McGraw-Hill, ISBN 007-141208-5

*** (2009) http://www.tdc.fr--TDC Software, Accessed on: 2009-05-13

*** (2009) http://www.axiomaticdesign.com--Axiomatic Design Solutions Inc., Accessed on: 2009-05-13