nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

Graphique avec plusieurs facteurs et plusieurs réponses

c4rmi3r-c0r3

Bonjour à tous,

Je suis étudiant en chimie, cependant j'aimerai creer un plan d'expérience qui me donne les meilleurs valeurs de facteurs.

Cependant j'arrive a tracer les graphiques en 3D selon 2 facteurs et 1 reponse mais j'aimerai avoir clairement le point ou j'aurai les meilleures reponses.

Merci de votre aide, j'espère avoir été assez claire.

Débit (F) Concentration (F) Volume reactif (F) Temperature (F) Volume injection (F) le reste sont les reponses

1	0,025	10	37	6
1	2,5	20	37	6
0,7	2,5	20	40	6
1	0,025	20	40	2
0,7	0,025	10	40	6
1	2,5	10	40	2
1	2,5	20	37	6
0,7	2,5	10	37	2
1	0,025	20	40	2
0,7	0,025	10	40	6
0,7	0,025	20	37	2
0,7	2,5	10	37	2

Accepted answers

All comments

Bill_Worley

Bonjour,

My French is very rusty so I used an online translator to better understand your question. I hope your English is better than my French!

You state you are using 2 factors and 1 response in your 3D graph to see where you have the best outcome for your experimental design. I would suggest that you use the Analyze > Fit Model > Standard Least Squares platform and then use the Prediction Profiler to determine your best outcome. The one thing your 3D graph will not be able to show is any interactions.

If you are certain that your 3D graph is showing the best outcome based on the two factors of choice you can use a data filter to select the points of interest as shown below.

Hope this helps.

Meilleures salutations,

Bill

c4rmi3r-c0r3

Thancks for you fast answer.

I want the better point with value factor.

I want a surface with all factors and all response surface to obtain the better yield.

My english is very mad. I try to write to understand me.

Sincerely.

c4rmi3r-c0r3

Ted

<Analyze > Fit Model > Standard Least Squares > Prediction Profiler>

I did not find "Prediction Profiler" (JMP 10)

Nate_Riordan

As per Bills Instructions and using the coffee data sample:

Fit Model with specification of Standard Least Squares (hit run to get the Fit Model Report):
Prediction Profiler resides within the Fit Model Report:

Ted

Dear nateriordan, Thank you for your attention!

In a JMP 10 this is so: Fit Model > Standard Least Squares > Factor Profiling > Profiler

Ted

Bill, Thank you for pay attention to the possibility of building so beautiful 3D graph