Design of Experiment for Chemical Engineering Research
For those chemical engineering post graduate students whom are doing research, or who might plan of doing research, planning your research is very important. There will be a lot of experiments to be carried out. There must be a systematic way of doing all the tonnes of experiments. And one technique to do so is by using the Design of Experiment (DOE). My wife , Mazura, and a colleague of mine, Farizul, is going to use this tool for their pH.D research. I did used this as well during my Masters by reseach work few years ago.
DOE is a very useful technique to plan our research. From DOE we can optimize the research and know all the processing parameters that can yield optimum result. So what is actually DOE?
In an experiment, we deliberately change one or more process variables (or factors) in order to observe the effect the changes have on one or more response variables. The (statistical) design of experiments (DOE) is an efficient procedure for planning experiments so that the data obtained can be analyzed to yield valid and objective conclusions.
DOE begins with determining the objectives of an experiment and selecting the process factors for the study. An Experimental Design is the laying out of a detailed experimental plan in advance of doing the experiment. Well chosen experimental designs maximize the amount of "information" that can be obtained for a given amount of experimental effort.
The statistical theory underlying DOE generally begins with the concept of process models.
Process Models for DOE
It is common to begin with a process model of the `black box' type, with several discrete or continuous input factors that can be controlled--that is, varied at will by the experimenter--and one or more measured output responses. The output responses are assumed continuous. Experimental data are used to derive an empirical (approximation) model linking the outputs and inputs. These empirical models generally contain first and second-order terms.
Often the experiment has to account for a number of uncontrolled factors that may be discrete, such as different machines or operators, and/or continuous such as ambient temperature or humidity. Continue reading the article HERE. It's a very useful and informative one.
There a re various software/packages that offer DOE. Among them are using Statistica, Design Expert, DOE Fussion. So, incase you're interested to learn them, get hold of the software. You can start learning them from the help or tutorial available in the software. But, learning by yourself is quite slow. I know because I experienced that before. It's better for you to go through and then ask somebody who knows DOE. That will be a better way of learning. But if you/or your research group have the budget, request from your boss/supervisor to go for a dedicated course to learn DOE.
Ok, to go a little bit deeper, there are a few designs that can be choose under DOE. There are Plackett Burman Design, Response Surface Methodology (RSM) and others. I used RSM before and it was very usefull. I can have a lot of analysis generated infront of me by using RSM. It's a really fantastic tool.
posted by Kipas Repair JB @ 11:46 PM,
2 Comments:
-
At Monday, October 30, 2006,
Curious Cat said...
-
Here are some good sites with more information on design of experiments and articles on design of experiments including many from the Center of Quality and Productivity Improvement at the University of Wisconsin that was founded by William Hunter, my father and George Box (leaders in the field, in my biased opinion, but others opinions as well).
-
At Tuesday, November 07, 2006,
Kipas Repair JB said...
-
Thanks John for the input...really appreciate it
The Author
I’m Zaki. I used to be a project, process and chemical engineer. Few years ago I successfully became a Chartered Engineer (IChemE) and Professional Engineer (BEM). I'm now employed as a chemical engineering educator/researcher/consultant. Hope you like reading my blog. I welcome any feedback from you. My email: zaki.yz[alias]gmail.com. TQ!
