Home > Products > CES Selector > Ashby methods

The Ashby Methods - the Power behind CES Selector

The power behind CES Selector is the Cambridge Engineering Selector methodology developed at Cambridge University and Granta Design by Professor Mike Ashby and colleagues. Here are some of the key concepts:

Functions, objectives, and constraints

Materials are sometimes chosen by trial and error or simply on the basis of what has been used before. While this approach frequently works, it does not always lead to optimization or innovation.

The Ashby approach is 'design-led'. It starts by asking 'What is the function of the component in the design?', 'What objectives need to be optimized and what constraints must be satisfied?' For instance, a car body panel (function) needs to be as light as possible (objective) for a specified stiffness and cost (constraint). Other constraints on the design are acceptable resistance to mechanical impact and to contact with various environments.

The advantage of this approach is that it is systematic and unbiased in its focus on product objectives.

Selection charts

Have you ever entered choices on the web for a product, such as a vacation package or a car - or for that matter a material - only to get back the unhelpful message: "Sorry, nothing meets your criteria. Please try again"? Ashby's selection charts avoid this problem and do much more besides. They convey information about your options in pictures and quickly answer questions like "Why isn't the material I first thought of the best?"

A materials selection chart using MaterialsUniverse data - Young's modulus is plotted against heat deflection temperature for filled and unfilled thermoplastics

Materials selection charts plot one property against another. Every material in the dataset that you are studying is represented as a point or (more typically) as an ellipse showing the range of its possible values for either property. The first thing that the chart provides is a quick, visual indication of the relative position for all of the materials being considered. This can be sufficient to focus on likely candidates, and also helps to develop an intuitive feel for the relative performance of materials. But the chart's real power is that it provides a graphical environment in which to apply and analyze quantitative selection criteria, such as those captured in performance indices, and also to make trade-offs between conflicting objectives.

Performance indices

Material performance for a specific application is rarely governed by the individual properties found in handbooks or manufacturers datasheets, but by combinations of two or more of these properties. These combinations, derived through mathematical analysis of the engineering problem, are called performance indices. Finding a material with a high value of the index maximizes the performance of the component.

Specific stiffness is not the optimum selection criterion for a panel in bending.
Click for a larger image.

This often leads to surprising results. For instance, many would think the best material for a lightweight panel such as in a car body is given by specific stiffness, modulus/density (E/r). In fact, it is given by E^(1/3)/r - a result that has significant implications for material choice. 50% glass-filled polymer is the stiffest material by weight, but, surprisingly, 30-35% of filler delivers the lightest panel for a given stiffness. (See chart, above).

Trade-offs

Trade-offs are endemic in both everyday life and materials selection. Perhaps the most common and obvious is the trade-off between performance and cost. Selection charts, with their ability to show materials performance relative to a combination of properties in a simple (yet quantitative) manner, are ideal vehicles for analyzing trade-offs and for then communicating them within an organization.

Model-based selection using functional data

Functional data (also known as multi-point or curve data) is very important within many sectors industry. For example, the mechanical properties of alloys or plastics can be highly temperature dependent, so it rarely makes sense to report a single value for a property such as tensile modulus - this is typically delivered as a curve that shows tensile against temperature.

Such data is usually only available in the form of picture graphs, meaning it cannot be used in a quantitative selection process. Instead, the user must refer to the graphs individually, manually interpolating them for the relevant conditions. This can become very time consuming.

In model-based selection, such as that available within CES Selector, the curve is stored as a function within the materials property database so that the property value at any temperature can be extracted. Selection charts can thus be constructed for any conditions - the image below shows how a plot of density v modulus, and thus the selection decisions that it could drive, changes with temperature.

Selection charts showing tensile modulus at room temperature (left) and at 400°K (right). Using model-based selection, the effect of temperature on the modulus of each material can be readily incorporated into the selection process.