It would be more meaningful to consider a specific property of material that is important to your problem as a parameter than the material type so that you can design your experiments better. The selected property should be the most influential for your response, for example material roughness, hardness, modulus or strength. By selecting the materials as controlling parameters in Taguchi optimization methods, it would be impossible to quantify them into levels of the input parameters, whereas by selecting a specific property as the controlling parameter, it would be easy to categorize the materials into different levels.

My answer is in the stream line and I would suggest you to quantify your material properties in terms of its Hardness, Conductivity, melting point etc. The software handling Taguchi DOE do not have such provision to provide non-numerical material properties as an input parameter and it is illogical to.

A very big thank you for the kind answers given.

But I would like to ask, if the type of materials (types of nanoparticles) does have a significant influence on the final output that I wish to observe, would it be better for me to enter this as one of the parameters in the design?

As one might know, different types of nanoparticles behaves quite differently from one another and this technology is still relatively new.

Thank you.

Taguchi designs are based upon categorical levels of factors. So, you can use numerical and non-numerical factors as your levels. However, Taguchi designs tend to be cookie cutter. You need to force your experiment to fit your design. 

If you use an optimal factorial design or an optimal response surface design, you fit your design to your experiment. You customize it as you see fit. In most cases, optimal response surfaces require fewer samples and deliver better results. At worst, an optimal response surface will give you the exact same design as Taguchi. 

Yes. You can use Taguchi  technique for analyzing categorical factors too and is possible with Design Expert Software.

Dear Chai Yee Ho   

I used Taguchi design in my last work and I explain that step by step in my article. You can see my new publication in below link.

I hope my article, would help you in your research filed.

Good Luck

http://www.tandfonline.com/doi/abs/10.1080/19443994.2015.1092893

You can find the article “Shahavi M. H., Hosseini M., Jahanshahi M., Meyer R. L. and Najafpour G. D. (2015) Clove oil nanoemulsion as an effective antibacterial agent: Taguchi optimization. Desalination and water treatment, Taylor & Francis, Impact Factor: 1.173, ISSN: 1944-3994, DOI: 10.1080/19443994.2015.1092893, “ARTICLE IN PRESS”” through below Researchgate link.

Article Clove oil nanoemulsion as an effective antibacterial agent: ...

Yes, as much time that quality characteristic can be measured in a certain manner, or attribute assessed (with a variable assessment result) … the lattice is not influenced as much time as the parameters and values are coded …

LM

Yes. You can use Taguchi method for attribute data also. You can find such examples in 

Taguchi Techniques for Quality Engineering by Philip Ross.