My question is related to the use of Response Surface Methodology (RSM) for optimization of independent variables having compositional variation during substrate addition. Say, for example, i have identified 4 factors that are significantly affecting a given response (or process). These are for example pH, Temperature, substrate loading and time. Now, pH, temp, and time is governed by high and low levels, the values of which can be provided at the time of design of experiment. However, the substrate part is not only influenced quantitatively by lower and higher concentrations (that is amount of substrate added) but can also be affected by composition of substrate. The substrate which we are talking is a mixture of 3 different feed stocks.
Now, the question is that which design is best suited so that the qualitative and quantitative both variations can be singly incorporated for optimization of any response.
Hoping a positive reply.
Kshitiz
Dear Kshitiz
the best book for find your answer is :
https://books.google.com/books/about/Design_and_Analysis_of_Experiments.html?id=kMMJAm5bD34C
I'm sure you can find your answer in it
Abbas
Dear Kshitiz
the best book for find your answer is :
https://books.google.com/books/about/Design_and_Analysis_of_Experiments.html?id=kMMJAm5bD34C
I'm sure you can find your answer in it
Abbas
You can use definitive screening design. It can be used for Screening as well as Optimization. DSD is described from application viewpoint.
https://www.researchgate.net/publication/309383755_Multi_Response_Simulation_and_Optimization_of_Gas_Tungsten_Arc_Welding
Dear Kshitiz...beside above answers, the RSM design is concerns to study the interaction between variables each other. in first, if you have a lot of variables you should used plackette-burman design for screening and select the most significant variables that affected positively in your response at two levels (+1 and -1) let's say as you mention temp., ph., time substrate, in case of substrate you have mixture of substrates i don't know what you mean by 3 feeding stocks you mean different type of substrate or not .....this point need to be more clear.....also substrate concentrate should included in design.
after you select the most significant variables then you can apply RSM design like box behnken design at 3 levels (-1,0,+1) or CCD at 4 levels (-2,-1,0,+1+2)...these designs concerns with the interaction between the variables to each other and tell you the exactly concentration you should apply to obtained the maximum response (activity or productivity)..RSM applied for 3 or 4 or 5 variables but not less than 3 variable
Dear Soad,
Thank you for the answer. But, the main query is still unanswered. This is regarding substrate composition. You rightly pointed out, the substrate which I am using is already composed of 3 things: saw dust, cow dung and humus. The variation in composition of these also effect the response and is also significant.
Regarding screening of significant variables, we can simply use Factorial design also (2-level-n-factorial design, where 'n' are no. of factors affecting any given response) and on the basis of p-values we can simply eliminate the non-significant factors.
But, my question is, how 2 things can be optimized using a single design. Is there any design available, which can also optimize the composition of 3 things (saw dust, cow dung, and humus) present in substrate, side-by-side along with optimization of complete process using pH, temp, time and substrate. I think you will now understand my query.
Thank you.
yes, the vision now is clear...in case of your substrate, you can use mixture design to determine the exactly how much you should take from the mixture as all or conc., of three mixture together..mixture design will help you to get it...the attached file below will illustrate to you the idea of mixture design..usually this design used for formulation... it determine the conc., of each component in mixture which required to obtain the highest response..
- Badges
- Science topic
- Similar topics
- Biological Science
- Biochemistry
- Biochemical Techniques
- Analytical Chemistry
- pH