Experimental Design Lecture - Revision history

Admin: /* Full Factorial Design */

2017-12-17T06:18:30Z

Full Factorial Design

← Older revision		Revision as of 06:18, 17 December 2017
Line 243:		Line 243:
	===Full Factorial Design===		===Full Factorial Design===

	The following discussion assumes use of two ~~factors~~. Using more than 2 ~~factors~~ can potentially make factorial experimental design very complex.		The following discussion assumes use of two factor levels. Using more than 2 factor levels can potentially make factorial experimental design very complex.

	The full factorial algorithm is as follows:		The full factorial algorithm is as follows:

2017-12-16T12:43:34Z

Links to Papers

← Older revision		Revision as of 12:43, 16 December 2017
Line 521:		Line 521:
	==Links to Papers==		==Links to Papers==

			<s>
	* Box Behnken 1960: http://files.charlesmartinreid.com/BoxBehnken1960.pdf		* Box Behnken 1960: http://files.charlesmartinreid.com/BoxBehnken1960.pdf
	* Plackett Burman 1946: http://files.charlesmatinreid.com/PlackettBurman1946.pdf		* Plackett Burman 1946: http://files.charlesmatinreid.com/PlackettBurman1946.pdf
			</s>

2011-06-29T09:25:59Z

Monte Carlo Sampling

← Older revision		Revision as of 09:25, 29 June 2011
Line 166:		Line 166:
	{{Main\|Monte Carlo Experimental Design}}		{{Main\|Monte Carlo Experimental Design}}

	[[Image:Roulette.jpg\|400px\|Roulette illustrates the random nature of the function samples used in Monte Carlo methods.]]		{\|class="wikitable" style="width:400px;"
			\|-
			\|[[Image:Roulette.jpg\|400px]]
			\|-
			\|Roulette illustrates the random nature of the function samples used in Monte Carlo methods. A good random number generator (for example, a roulette table) is an essential part of any Monte Carlo method.
			\|}

	Monte Carlo sampling is essentially a brute-force technique in which random samples are taken until confidence that the entire space has been sampled is satisfactory.		Monte Carlo sampling is essentially a brute-force technique in which random samples are taken until confidence that the entire space has been sampled is satisfactory.

2011-06-29T09:23:30Z

Monte Carlo Sampling

← Older revision		Revision as of 09:23, 29 June 2011
Line 165:		Line 165:

	{{Main\|Monte Carlo Experimental Design}}		{{Main\|Monte Carlo Experimental Design}}

			[[Image:Roulette.jpg\|400px\|Roulette illustrates the random nature of the function samples used in Monte Carlo methods.]]

	Monte Carlo sampling is essentially a brute-force technique in which random samples are taken until confidence that the entire space has been sampled is satisfactory.		Monte Carlo sampling is essentially a brute-force technique in which random samples are taken until confidence that the entire space has been sampled is satisfactory.

2011-06-29T09:21:31Z

← Older revision	Revision as of 09:21, 29 June 2011
(No difference)

2011-06-29T09:20:35Z

← Older revision		Revision as of 09:20, 29 June 2011
Line 516:		Line 516:
	* Box Behnken 1960: http://files.charlesmartinreid.com/BoxBehnken1960.pdf		* Box Behnken 1960: http://files.charlesmartinreid.com/BoxBehnken1960.pdf
	* Plackett Burman 1946: http://files.charlesmatinreid.com/PlackettBurman1946.pdf		* Plackett Burman 1946: http://files.charlesmatinreid.com/PlackettBurman1946.pdf






	{{ExperimentalDesign}}		{{ExperimentalDesign}}

2011-06-26T02:42:22Z

Monte Carlo Sampling

← Older revision		Revision as of 02:42, 26 June 2011
Line 163:		Line 163:

	==Monte Carlo Sampling==		==Monte Carlo Sampling==

			{{Main\|Monte Carlo Experimental Design}}

	Monte Carlo sampling is essentially a brute-force technique in which random samples are taken until confidence that the entire space has been sampled is satisfactory.		Monte Carlo sampling is essentially a brute-force technique in which random samples are taken until confidence that the entire space has been sampled is satisfactory.

2011-06-26T02:38:06Z

← Older revision		Revision as of 02:38, 26 June 2011
Line 514:		Line 514:
	* Box Behnken 1960: http://files.charlesmartinreid.com/BoxBehnken1960.pdf		* Box Behnken 1960: http://files.charlesmartinreid.com/BoxBehnken1960.pdf
	* Plackett Burman 1946: http://files.charlesmatinreid.com/PlackettBurman1946.pdf		* Plackett Burman 1946: http://files.charlesmatinreid.com/PlackettBurman1946.pdf



			{{ExperimentalDesign}}

2011-06-23T17:49:07Z

← Older revision		Revision as of 17:49, 23 June 2011
Line 502:		Line 502:
	}}		}}
	</ref>		</ref>

	~~===Koshal Design===~~

	~~Koshal 1933~~

	==Sequential Assembly==		==Sequential Assembly==
Line 513:		Line 509:

	<references />		<references />

			==Links to Papers==

			* Box Behnken 1960: http://files.charlesmartinreid.com/BoxBehnken1960.pdf
			* Plackett Burman 1946: http://files.charlesmatinreid.com/PlackettBurman1946.pdf

2011-06-20T20:55:15Z

Composite Design

← Older revision		Revision as of 20:55, 20 June 2011
Line 410:		Line 410:
	Thinking about a three-dimensional factorial design, the factorial design sample points looks like a cube in parameter space. Composite design makes a "star" design.		Thinking about a three-dimensional factorial design, the factorial design sample points looks like a cube in parameter space. Composite design makes a "star" design.

	The additional sample points consist of the new lower <math>-2</math> and upper <math>+2</math> levels for each factor, while keeping all other factors constant at the <math>0</math> level. It also includes a center point, at which all factors have a factor level of <~~math0~~</math>.		The additional sample points consist of the new lower <math>-2</math> and upper <math>+2</math> levels for each factor, while keeping all other factors constant at the <math>0</math> level. It also includes a center point, at which all factors have a factor level of <math>0</math>.

	For a two-parameter experimental design, this looks like:		For a two-parameter experimental design, this looks like: