Materials selection in Product design and development
Marco Leite, Arlindo Silva and Elsa Henriques
Abstract
Material selection is a live process inside the more global process of product development. The choice of a material and manufacturing technologies for a certain product is a fundamental task, often jumped over, either because the choice looks obvious, or because habits of convergence thinking lead to past proven solutions. In consequence, more often than wanted, there’s a need to do extensive rework, with obvious first order problems, man hours increase and renegotiation of contracts and other second order problems like higher time to market or pushing a bad product to market.
Concept development defines the product architecture and normally represents a point of no return to the new product. Decisions concerning materials at these stages should account for functionality, technological and economical criteria.
There´s a need to understand what methods are available, at each stage of the product development, to the decision making process of materials selection.
A review of some current methods for materials and process selection at the different stages of product design is presented and a new Materials Selection Engine that frames the decision making process at the several decision gates is proposed. The proposed Materials Selection Engine includes technical performance, predictive estimation of the cost drivers and environmental issues.
Introduction
A successful product development puts a lot of pressure in companies. They should be capable of developing more and more complex products in a shorter development time, with a high perceived quality and within a market compatible cost [1].
One of the fundamental aspects of successful product development is the choice of the right materials. Materials are responsible for function (to support loads, to assure pressure, etc.), structure (physical embodiment) and interaction with the customer/user. Materials also connect the computer calculations and drawings with a real or working design. New materials are responsible for the appearance of new products or radical changes in current products.
An effective material and process selection directly determines the perceived quality and the cost of the product and is indirectly connected to the time to market. In the phase of generating new concepts, different materials may support major changes in product architecture, thus increasing the probability of achieving a successful product. In opposition, a non effective materials selection increases the risk of downstream rework and disappointing market results. The failure rate of good technical products is alarming; well over 90% according to estimates don’t appeal to the unforgiving consumer, [2]. Although these failures are not all related to materials and processes selection, this issue certainly contributes for a significant percentage of disappointments.
On the other hand, the number of materials available to design teams is over 120 000 and this number is growing at an increasing rate, [3]. Moreover, the ability to change the materials properties in the manufacturing process increases even more the complexity of the selection process.
“Not too long ago, materials selection was considered a minor part of the design process. Today that is an unacceptable approach for all but routine and simple designs”, [4]. That vision in 1991 is more and more true nowadays: materials shortages resulting from high market demand, low cost and low weight pressures from market, the need for higher performance materials and the underlying need to do all this with “greener” materials pose a serious challenge to the design teams. In fact, design teams face a difficult challenge: to identify the “best material”, among an enormous set of alternatives, considering selection criteria ranging from technological and aesthetics to economical and environmental issues.
Fig. 1 summarizes the factors that influence product design. Material decisions must be made within this framework. The four main factors upon which the designer relies when making materials choices are the relation between material specification and technical performance, economic performance, environment performance and industrial design embedded in the product and its function, [5].
The word “selection” implies a “decision”. The selection of materials involves complex decision-making processes. This decision-making would have to balance different intricate functional, technological and economical criteria reflecting the function of the part/product, its architecture, the manufacturing process, the production volumes, the final cost and the implications to the environment through a life-cycle assessment.
There’s a myriad of methods for material selection. The next pages try to bring up some of the most relevant methods with their strengths and weaknesses and at what phase they can be used in the long product development cycle.
Materials selection engine
A methodology is proposed to address the individual components that must to be considered in an effective material selection process: technical performance, economical performance and environmental performance induced in a new product for the different available materials.
The objective of the proposed methodology is to come up with individual indicators for each of the components (technical, economic and environmental performances) and use a multi-criteria decision tool to aid the decision making process. Another objective of the methodology is to use different methods for each of the components at different stages of product development.
The top zone of Fig. 2 shows the product development flow (based on Pahl & Beitz, [6]), connected with the interactions at its successive stages with the materials selection procedures.
Concept generation phase
A new product generally arises from market needs. At this stage there is a need to identify the original system requirements (functions of the product).
After the establishment of the systems requirements, the product goes to a stage of concept generation. In this phase different concepts arise and are ranked. The original set of requirements is translated into workable concepts (engineering material properties) and a set of materials classes that adjusts to the requirements are identified.
The identification of a materials class able to fulfill the technical specifications of the new concept is the first step of the selection process. This preliminary materials identification in conceptual design is essentially function-oriented and functionalities-centered. In the concept generation stage the focus is on achieving the required systems functions [7], so the identification of viable materials supports the conception of feasible solutions. Material properties charts are probably the most common and visual way of screening material classes for a given application, [3]. This method allows for the selection of a set of candidate materials, by comparing two engineering properties at a time. A similar method can be used to perform process selection. For simple problems the method can also be used to rank the candidate materials by the use of the performance indexes.
Other authors developed software systems for materials identification. The creation and maintenance of databases to store and process data of materials can provide easy access to materials data [8]. But the access to the data is only part of the problem. In order to effectively use the data, the designer should have the right knowledge, which allows him to formulate an intelligent approach to the search and to retrieve useful data and get the better adequate materials class. A knowledge base system (KBS), as a computer system that attempts to represent human knowledge or expertise in order to provide quick and easily accessible knowledge in a practical and useful way, are the new computer aided tools capable of assisting the user in an interactive way to solve various problems in the field of materials screening, [8].
Embodiment design phase
The following phase in the Product Development is the embodiment design stage, where the level of detail increases and the concepts are translated into sketches with some level of detail. In this stage the materials selection takes a paramount importance, since the concepts details can be different depending on material selection.
At the embodiment design level, the “Materials Selection Engine 1st loop” is proposed (Fig. 2). The data available is identified and for each concept a set of candidate materials is identified. An evaluation of the materials screening is necessary. For that a preliminary estimation of part features is needed using for example CAD and even simple CAE tools. One must mention that the objective is not to have an extensive and validated design, but only to have an idea of the effect of the material on the most relevant part features and especially on part dimensions.
With a preliminary geometry found it is possible to use methods for ranking the materials in the 3 components of evaluation: technical, economical and environmental.
Each of the components can be estimated, from a given set of material attributes. For example the technical performance component can have a contribution of some mechanical properties like corrosion resistance, weight of the part, stiffness, among others. Each of these attributes has a different importance for the technical component, so the ranking of the materials is also, by itself, a multi-criteria decision making problem.
The technical performance components can be assessed by several methods. Even the use of simple methods like the simple addictive weights is often complicated, in particular the estimation of the weights that each property takes. In fact, often the biggest concern is about the subjective character of the weights. For example: for a cantilever beam, what is more important: the stiffness, the strength or the weight? All of course are very important, so 33% importance to each factor can be used. But one can argue that stiffness in a beam is far more important than strength (given that all the alternatives can support the load), or otherwise in other applications. To assess the weights there are some techniques like the entropy method, the analytical hierarchy method or the digital logic method. To rank the alternatives there are also some different techniques: ELECTRE methods, TOPSIS, GRA and others. Table 1 provides some examples of use of these methods in the literature.
Table 1 – Decision making methods used for materials selection.
| Ranking | Weights | Authors |
| ELECTRE | Entropy | [9] |
| TOPSIS | Entropy | [10] |
| TOPSIS | AHP | [11] |
| WPM | Digital logic | [12] |
| GRA | AHP | [13] |
| GRA | NSFDSS | |
| GFDA | AHP | [14] |
| GFDA | FLA |
Another technique that can be used to raise the weights of the attributes and to perform trade-off analysis for materials selection is the Multi Attribute Utility Analysis (MAUA), [5]. Utility analysis affords a rational method of materials selection which avoids many of the fundamental logical difficulties of many widely used alternative approaches. However this method is valid when there is certainty regarding the attribute levels of the alternatives. Whenever that is not the case, another operations research technique, subjective probability assessment (SPA), can be used to address this issue. SPA makes it possible to measure a probabilistic distribution describing the confidence of the decision maker in the levels of attributes for which there is a high degree of uncertainty. These probability distributions can be used in conjunction with MAUA to provide a consistent framework for making materials selection decisions, [15]. In order to obtain the SPA for new materials that the designers have uncertainty regarding attributes, a questionnaire must be used,[16].
All these methods with some advantages and disadvantage in accordance to the specific problem can be applied to the materials ranking concerning technical performance.
As regards economical performance it can be assessed based on cost metric computation. One of the attractions of cost as a basis for decision making is the apparent simplicity of the metric: an economic measure of the resources employed to undertake a set of actions, typically to yield a good or service. However, engineers are usually far less comfortable with cost when tasked with relating it to a set of specific technical or design changes. The difficulty arises from the fact that cost has traditionally been associated with accounting rather than with the engineering field, [17]
The economic performance component can be assessed with the use of predictive process based cost modeling techniques (PBCM), [17]. The use of this type of tools at this stage is of great importance because the PBCM is composed of three interrelated and interdependent models: a process model, an operations model and a financial accounting model. The PBCM explicitly offers the modeler the opportunity to incorporate engineering and operational functionality into estimates of resource requirements, whose technical interactions are largely ignored or oversimplified in other modeling approaches. This cost analysis can be extended to consider not only cost of production (the one directly supported by the producer), but also the life-cycle cost (LCC) of the product, [18]. More and more companies are trying to develop ways to lower the cost of use and disposal of products (e.a. aircraft or automotive industry) as a tradeoff for higher manufacturing costs.
The environmental performance component can be assessed using life cycle assessment tool. The eco-indicator 99 is one of the tools that can be used to rank the alternatives. The eco-indicator 99 uses a cradle to grave approach and consider 11 environmental impact categories in the following three areas: Human health, ecosystem quality and resources. [18],[19], [20] and [21] use the LCA technique for different applications to understand the environmental impact of the materials used to produce a product.
These estimations must be done as soon as possible in the early phases of product development where major improvements can be done to the design.
The major issues faced over the last decade concerning motor vehicles are: environmental constrains, economic demands and performance enhancements, [22]. The huge effort taken by the automobile companies to reduce exhaust emissions obliges to better motors and for higher performance materials that enable to lose weight and aerodynamic drag of the vehicle.
When the 3 components are calculated, the decision can be done using a ternary graphic, where each axis is associated to the importance of each performance component. At this point, the problem is also to determine the weights of each of the components. This decision must be made by each company according to their decision values of the long term revenue.
Detail design phase
When the product reaches the detail design (Fig. 2), a “Materials Selection Engine 2nd loop” is needed. The objective of this phase is essentially to refine and validate the “best material”. The natural evolution of the product development in the phase of detail design can change the ranking obtained in the materials selection 1st loop. In this stage, for questions of time expenditure, a smaller set of materials can run the materials selection 2nd loop.
Conclusions
Materials and process selection is a very complex decision making process by itself. There are technological, economic and environmental issues that must be taken in account in the early stages of product design.
The use of material set-based product development strategy increases the number of operations necessary to deliver a final product in the early stages of product design. However, this effort is of paramount importance because it is at these stages that fundamental decisions are made. If the selected material is frozen to early the result can be a poor solution too far from the “best one”.
The proposed methodology is very simple, but hard work is hidden inside, mainly when product complexity increases. When a new product contains different parts made with different materials the work load increases exponentially. The outcome of the proposed methodology is a strategic tool, or a designer thinking guide, for informed decision making regarding materials selection at the different stages of product development.
Acknowledgments
The authors would like to acknowledge the MIT-Portugal Program.
Keywords
Materials selection; Product development, Decision making tools
Bibliography
[1] Ulrich, K.T. and S.D. Eppinger, Product design and development. 3rd ed. 2004, McGraw-Hill.
[2] Patrick, J., How to develop successful new products. 1997. American Marketing Association.
[3] Ashby, M.F., Materials selection in mechanical design. 3rd ed. 2005: Elsevier Butterworth-Heinemann.
[4] Dieter, G.E., Engineering design: a materials and processing approach, 2nd ed. 1991, McGraw-Hill.
[5] Field, F.R., J.P. Clark, and M.F. Ashby, Market drivers for materials and process development in the 21st century. Mrs Bulletin, 2001. 26(9): p. 716-725.
[6] Pahl, G., et al., Engineering design: a systematic approach. 3rd ed. 2007, Springer.
[7] Deng, Y.M. and K.L. Edwards, The role of materials identification and selection in engineering design. Mat.& Des. 2007. 28(1): p. 131-139.
[8] Sapuan, S.M., A knowledge-based system for materials selection in mechanical engineering design. Mat.& Des. 2001. 22(8): p. 687-695.
[9] Shanian, A. and O. Savadogo, A material selection model based on the concept of multiple attribute decision making. Mat.& Des. 2006. 27(4): p. 329-337.
[10] Jee, D.-H. and K.-J. Kang, A method for optimal material selection aided with decision making theory. Mat.& Des. 2000. 21(3): p. 199-206.
[11] Rao, R.V. and J.P. Davim, A decision-making framework model for material selection using a combined multiple attribute decision-making method. Int. J.of Adv.Man. Tech. 2008. 35(7-8): p. 751-760.
[12] Dehghan-Manshadi, B., et al., A novel method for materials selection in mechanical design: Combination of non-linear normalization and a modified digital logic method. Mat.& Des. 2007. 28(1): p. 8-15.
[13] Chan, J.W.K. and T.K.L. Tong, Multi-criteria material selections and end-of-life product strategy: Grey relational analysis approach. Mat.& Des. 2007. 28(5): p. 1539-1546.
[14] Kuo, T.-C., S.-H. Chang, and S.H. Huang, Environmentally conscious design by using fuzzy multi-attribute decision-making. The Int. J. of Adv. Man. Tech. 2006. 29(3): p. 209-215.
[15] Roth, R., F. Field, and J. Clark, Materials selection and multi-attribute utility analysis. J. of Computer-Aided Mat.Des. 1994. 1(3): p. 325-342.
[16] Roth, R., Materials substitution in aircraft gas turbine engine applications, in Department of Materials Science and Engineering 1992, MIT: Boston. p. 152.
[17] Field, F., R. Kirchain, and R. Roth, Process cost modeling: Strategic engineering and economic evaluation of materials technologies. Jom, 2007. 59(10): p. 21-32.
[18] Ribeiro, I., et al., Life cycle engineering methodology applied to material selection, a fender case study. J. Cleaner. Prod. In Press, Corrected Proof.
[19] Kurk, F. and P. Eagan, The value of adding design-for-the-environment to pollution prevention assistance options. J. of Cleaner Prod. 2008. 16(6): p. 722-726.
[20] Bovea, M.D. and A. Gallardo, The influence of impact assessment methods on materials selection for eco-design. Mat.& Des. 2006. 27(3): p. 209-215.
[21] Matos, M.J. and M.H. Simplicio, Innovation and sustainability in mechanical design through materials selection. Mat. & Des. 2006. 27(1): p. 74-78.
[22] Edwards, K.L., Strategic substitution of new materials for old: Applications in automotive product development. Mat. & Des. 2004. 25(6): p. 529-533.
Illustrations
Fig. 1. The driving forces that influence the product design [5]
