The charts we’ve just discussed give an overview, but the number of materials that can be shown on any one of them is obviously limited. Selection using them is practical when there are very few constraints, but when there are many, as there usually are, checking that a given material meets them all is cumbersome. Both problems are overcome by a computer implementation of the method.
The CES material selection software2 is an example of such an implementation. Its database contains records for materials, organized in the hierarchical manner. Each record contains property data for a material, each property stored as a range spanning its typical (or, often, permitted) values. It also contains limited documentation in the form of text, images, and references to sources of information about the material. The data are interrogated by a search engine that offers the search interfaces shown schematically in Figure 8.16.
On the left is a simple query interface for screening on single attributes. The desired upper or lower limits for constrained properties are entered; the search engine rejects all materials with attributes that lie outside the limits. In the center is shown a second way of interrogating the data: a bar chart, constructed by the software, for any numeric property in the database. It, and the bubble chart shown on the right, are ways of both applying constraints and ranking. For screening, a selection line or box is superimposed on the charts with edges that lie at the constrained values
The operation and outputs of typical selector software.
of the property (bar chart) or properties (bubble chart). This eliminates the materials in the shaded areas and retains the materials that meet the constraints. If, instead, ranking is sought (having already applied all necessary constraints), an index line like that shown in Figure 8.8 is positioned so that a small number—say, 10—materials are left in the selected area; these are the top-ranked candidates. The software delivers a ranked list of the top-ranked materials that meet all the constraints.
8.7 Summary and conclusion
There is a broad strategy that works for selecting anything—products, services, or materials. Decide on the attributes the entity must (or must not) have, defining a set of constraints. Apply them, leaving a list of the entities that meet them. Decide on an objective—a measure of excellence.
It could be price (the cheaper, the better), weight (the lighter, the better), or eco-impact (the lower, the better), or some other measure of performance. Use this measure to rank the list of survivors. Then get to work researching the top three or four on the list, gathering as much information as possible to make a well-informed final choice.
There are, almost always, many constraints, but this does not create a difficulty: simply apply them sequentially, retaining only those entities that meet them all. Often, too, there are two or more objectives, and that does create a difficulty; the entity that best satisfies one is unlikely to also be the best choice by the other. Then tradeoff methods become useful, either graphical ones (plotting the alternatives, identifying the tradeoff line, and then using judgment to select an entity on or near the line) or analytical ones, by formulating a penalty function and seeking the entitites that carry the lowest penalty.
Now we have a set of tools. In Chapters 9 these tools will be used to analyze and select materials to design for the environment.
8.8 Further reading
Ashby, M. F. (2005), "Materials selection in mechanical design", 3rd ed.,
Butterworth Heinemann. Chapter 4, ISBN 0-7506-6168-2. (A text that develops the ideas presented here in more depth, including the derivation of material indices, a discussion of shape factors, and a catalog of simple solutions to standard problems.)
Ashby, M. F., Shercliff, H. R. and Cebon, D. (2007), "Materials: engineering, science, processing and design", Butterworth Heinemann. ISBN-13: 978-0-7506-8391-3. (An elementary text introducing materials through material property charts and developing the selection methods through case studies.)
Bader, M. G. Proc of ICCM-11, Gold Coast, Australia, Vol. 1: Composites applications and design, ICCM, 1977. (An example of tradeoff methods applied to the choice of composite systems.)
Bourell, D. L., Decision matrices in materials selection, ASM Handbook Vol. 20, Materials selection and design, G. E. Dieter (Ed.), ASM International, 1997, 291-296, ISBN 0-87170-386-6. (An introduction to the use of weight factors and decision matrices.)
Dieter, G. E. (2000), "Engineering design, a materials and processing approach", 3rd ed., McGraw-Hill. 150-153 and 255-257, ISBN 0-07-366136-8. (A well – balanced and respected text, now in its third edition, focusing on the role of materials and processing in technical design.)
Field, F. R., and Neufville, R. de, "Material selection: maximizing overall utility, Metals and Materials", June 1998, 378-382. (A summary of utility analysis applied to material selection in the automobile industry.)
Goicoechea, A., Hansen, D. R. and Druckstein, L. (1982), "Multi-objective decision analysis with engineering and business applications", Wiley. (A good starting point for the theory of multiobjective decision making.)
Keeney, R. L. and Raiffa, H. (1993), "Decisions with multiple objectives: preferences and value tradeoffs", 2nd ed., Cambridge University Press. ISBN 0-521-43883-7. (A notably readable introduction to methods of decision making with multiple, competing objectives.)
