 |  |  | |  | |  | |  | |  |
Taguchi statistical analysis is one means of providing an analytical solution to the calculation of the effects of process variables.
Taguchi experimentation is often used as an aid in calculating the effects of changing processing variables on manufacturing production processes. This method creates an orthogonal array where the effect of each variable can be measured in combination with other variables. This avoids having to carry out costly experiments identifying the effect of changing one processing individually.
The method involves refined changes of the inputs to a process or an activity in order to observe the corresponding outputs. Taguchi stresses simplicity in its approach and avoids statistical complexity in the analysis of its results.
The National Physical laboratory has managed a programme on developing materials measurement methods for adhesives in which the Taguchi method has been applied to aid companies in the manufacturing process. This programme was sponsored by the Department of Trade and Industry.
To assist development of experiments involving several variables. To assess the significance of individual variables. To optimise the setting of individual variables in a multi-variable problem.
A technique for designing and performing experiments to investigate processes where the output depends on many factors (variables; inputs). By systematically choosing certain combinations of variables it is possible to separate their individual effects. Simpler and more economic alternative to having to run the process using all possible combinations of values of those variables.
The technique often referred to as Taguchi experimentation was used initially in Japan. Through using an orthogonal array the effect of each variable can be measured in combination with other variables. Properly designed experiments enable the array to separate the effect of each variable within a matrix of manageable size.
A special variant of Design of Experiments (DOE) that distinguishes itself from classic DOE in the focus on optimizing design parameters to minimize variation before optimizing design to hit mean target values for output parameters.
In the DTI sponsored MTS programme the Taguchi method was used to optimise process variables for the adhesive process. Analysis showed factors to be ranked in the following order, with best levels as indicated.
|
Variable Significance |
Variable |
Setting |
|
1 |
Oven temperature |
100°C |
|
2 |
Time to cure |
120 min |
|
3 |
Time in solvent |
5 minutes |
|
4 |
Time in oven |
60 minutes |
|
5 |
Time post oven |
4 hours |
|
6 |
Time in fixture |
15 minutes |
|
7 |
Parts temperature |
20°C |
E1 Rail bonding on a computer hard disc assembly
For further information on design of experiments and Taguchi please contact:
Dr Abayomi Olusanya
Centre for Materials Measurement and Technology
National Physical Laboratory, Queens Road
Teddington, Middlesex, TW11 0LW
Telephone: 0181 943 6798, Fax: 0181 943 6046, E-Mail: ao@npl.co.uk
Other links: