Yes, you are correct. Excel guessed that my objective was $E$4 and I didn't change it from the default. In this case, it doesn't matter if the objective is set correctly (it could be any random cell) because I added conditions that the equations had to be equal to zero. This creates a unique solution with equal number of equations and variables (zero degrees of freedom). An objective is important when there are multiple possible solutions that satisfy the equations. The objective is used to pick the best one. Here is an optimization problem: apmonitor.com/che263/index.php/Main/ExcelSolver Thanks for the great comment!
Is it just me or did that example minimize E4, which is a constraint rather than the objective function? Either way, I did appreciate this tutorial. Thanks!
+Jehan Music, you can either minimize the sum of the squared errors (E6 or E7, later) or add the constraints individually. I added the constraints individually so it didn't really matter what I included as the objective because there were two equations and two unknowns (or 3 eqns / 3 vars later). For some nonlinear problems with multiple potential solutions, the objective function can give guidance on which feasible solution is best.
Thank you for the video! When I try this I get an error in model, please verify that all cells and constraints are valid. This same error comes up with any equation I try to use solver on. Is there something I need to change in my settings?
great, I need minimize this function " sum( ( ( VR - VM ) .* QI ./ ( VM + eps ) ) .^2 ) " where VR is a value that I want to find and VM is a value that I have and QI I have too. The constraints are the mass balance. Is possible ? could you help me please ?
+Diogo Oliveira, you can add the constraints in the solver dialog box. You'll want to pose it as an optimization problem like I've shown here apmonitor.com/che263/index.php/Main/ExcelSolver You will want to use Matlab or Python if you have more than 50 variables.
Thanks a lot. I am studying chemical engineering in brazil. I like studying optimization problems. I use your videos to learn more. In my internship I'm trying to optimize this function. Sorry for my english I am learning too. do you have an email for contact? Thanks
+mohamed hassan, the Excel solver produces a local solution so you'll need to solve two instances with different initial guesses or constraints that will give the two local answers.
The solver returns the limit value, in this case 1 or 0 when I actually wanted it to be between 0 and 1. I need this solution. I need this solution, but thank you in advance.
Entendendo a Engenharia, if the solver report says that it converged then it is likely the correct solution or at least a local optimum. You can try to change the value of x to see if you get a better solution. With continuous variables, x>0 and x>=0 are the same because of machine precision in storing values.
![ILLSLICK X YOUNGOHM - We're The same [Official Video]](http://i.ytimg.com/vi/yUJNkobYR0E/mqdefault.jpg)
Hi ! Why do you set objective cell E4 and not E6 ? Isn't sum of E3 and E4 (Cell E6) the function that we want to minimaze ?
Thanks for the upload
Yes, you are correct. Excel guessed that my objective was $E$4 and I didn't change it from the default. In this case, it doesn't matter if the objective is set correctly (it could be any random cell) because I added conditions that the equations had to be equal to zero. This creates a unique solution with equal number of equations and variables (zero degrees of freedom). An objective is important when there are multiple possible solutions that satisfy the equations. The objective is used to pick the best one. Here is an optimization problem: apmonitor.com/che263/index.php/Main/ExcelSolver Thanks for the great comment!
Very helpful educational videos sir! Thank you! I learned a lot from your videos.
Is it just me or did that example minimize E4, which is a constraint rather than the objective function? Either way, I did appreciate this tutorial. Thanks!
+Jehan Music, you can either minimize the sum of the squared errors (E6 or E7, later) or add the constraints individually. I added the constraints individually so it didn't really matter what I included as the objective because there were two equations and two unknowns (or 3 eqns / 3 vars later). For some nonlinear problems with multiple potential solutions, the objective function can give guidance on which feasible solution is best.
Thank you for the video! When I try this I get an error in model, please verify that all cells and constraints are valid. This same error comes up with any equation I try to use solver on. Is there something I need to change in my settings?
Maybe try it again, exactly copying what I had?
helped me to solve my assignments! Thank you!!
I'm glad that it helped. Check out apmonitor.com/che263 for more content.
Helped me so much! Thanks, bro!
Glad it helped!
Thank you, really helped!
Thank you so much
Thank you
great, I need minimize this function " sum( ( ( VR - VM ) .* QI ./ ( VM + eps ) ) .^2 ) " where VR is a value that I want to find and VM is a value that I have and QI I have too. The constraints are the mass balance. Is possible ? could you help me please ?
+Diogo Oliveira, you can add the constraints in the solver dialog box. You'll want to pose it as an optimization problem like I've shown here apmonitor.com/che263/index.php/Main/ExcelSolver You will want to use Matlab or Python if you have more than 50 variables.
Thanks a lot. I am studying chemical engineering in brazil. I like studying optimization problems. I use your videos to learn more. In my internship I'm trying to optimize this function. Sorry for my english I am learning too.
do you have an email for contact?
Thanks
Thank you!
how can i show the 2 possible answers not 1 answer ?
+mohamed hassan, the Excel solver produces a local solution so you'll need to solve two instances with different initial guesses or constraints that will give the two local answers.
Thank you :)
You're welcome, Can.
How to do it when x 0
Entendendo a Engenharia, at 6:30 you can add additional constraints such as your inequality constraints.
The solver returns the limit value, in this case 1 or 0 when I actually wanted it to be between 0 and 1. I need this solution. I need this solution, but thank you in advance.
Entendendo a Engenharia, if the solver report says that it converged then it is likely the correct solution or at least a local optimum. You can try to change the value of x to see if you get a better solution. With continuous variables, x>0 and x>=0 are the same because of machine precision in storing values.
I will do this. Thank you for your help