Value Driven Design Logo

6.06 - Taguchi Approach to Experimentation


Taguchi is an off-line Quality Engineering approach which complements on-line quality control systems such as Statistical Process Control (SPC). The methodology encompasses a range of techniques for experimental design that are applicable to most areas of product design and manufacture. Taguchi experiments are a fast and efficient way of optimising new products and a ‘fast track’ to solving manufacturing problems. One approach is a total quality tool box combined with other quality tools used by Quality practitioners which makes it very effective.

The Taguchi approach

Taguchi is the next level after the application of FMEA processes when a particular process element requires a more in-depth attack for resolution.

A Taguchi set of experiments minimises the number of parameter changes required to each variable when searching for the optimum combination. Standard factorial tables are used that replace traditional one variable at a time experiments. This minimises effort and identifies the performance plateau that avoids high sensitivity of performance to small changes in a variable (i.e. a robust product or process design is produced economically). It does, however, require multiple experiments in which variable change and performance are tabulated.


  • Price and hence target costs are set by the market place
  • Within this cost the engineer must deliver the best quality consistent with performance
  • Quality is re-defined on a monetary scale which evaluates the losses associated with poor quality.
  • The engineer simultaneously considers both the variability and the output achieved with a particular choice of settings or dimensions
  • The optimum solution can be selected from knowledge of costs and the associated variability’s.

Background – Product / Process Design

The design process is split into three stages:-

System Design – a non-statistical process of surveying and selecting appropriate design technology and concepts to produce a prototype design that possess the functions required by the product plan.

Parameter Design – experimental design methods to find the optimum levels of the individual system parameters which were determined during the system design (e.g. dimensions and material types).

Tolerance Design – experimental design methods, used only after parameter, to set the tolerances of the parameters if necessary. Narrow tolerances should be the weapon of last resort to be used only when parameter design gives insufficient results.

Key Concepts

Quality Loss is a continuous function, relating quality and cost, which approaches a minimum at the nominal value (i.e. parts within tolerance still incur losses).

Quality Loss Function

Noises are parameters which are difficult, expensive or impossible to control (e.g. ambient temperature, humidity, user handling).

Robustness is the quality characteristic of a product or process which describes its immunity to ‘noises’.

Orthogonal Array is a balanced arrangement of experimental conditions, which permits efficient multi-factor testing whilst retaining the ability to perform single factor analyses.

Signal-to-Noise Ratio is a combined measure of variability and output associated with a particular situation (i.e. a quantitative measure of robustness).

Note: Because of its specialised nature, Taguchi should not be used indiscriminately. It should only be applied after a priority selection has been used (e.g. FMEA – see guide 4.13).

Stages of for Taguchi Design of Experiments

1. Planning

Careful planning for the course of experimentation before embarking upon the process of testing and data collection is important. A thorough and precise terms of reference identifying the need to conduct the investigation including an assessment of time and resources available to achieve the objective and integration of prior knowledge to the experimentation procedure are a few of the considerations for this stage. A multi-function team of individuals from different disciplines related to the product or process should be used to identify possible factors to investigate and determine the most appropriate response(s) to measure. A team-approach promotes synergy that gives a richer set of factors to study and thus a more complete experiment. Careful planning of experiments leads to increased understanding of the product or process being examined.

2. Screening

Screening experiments are used to identify the key factors that affect the system under investigation. These experiments must be carried out in conjunction with prior knowledge of the system to eliminate unimportant factors and focus attention on the key factors that require further detailed analyses. Screening experiments are usually efficient designs requiring a few executions where the focus is not on interactions but on identifying the vital few factors.

3. Optimization

Once attention is narrowed down to the important factors affecting the process, the next step is to determine the best setting of these factors to achieve the desired objective. Depending on the product or process under investigation this objective may be to maximize, minimize or achieve a target value of the response.

4. Robustness Testing

Once the optimal settings of the factors have been determined, it is important to make the product or process insensitive to variations that are likely to be experienced in the application environment. These variations result from changes in factors that affect the process but are beyond the control of the analyst. Such factors as humidity, ambient temperature, variation in material, etc. are referred to as noise factors. It is important to identify sources of such variation and take measures to ensure that the product or process is made insensitive (or robust) to these factors.

5. Verification

Verification means validation of the best settings of the factors by conducting follow-up experiments to confirm the system functions as desired and all objectives are met.

Taguchi Examples

Check this link

Further Reading