This page provides a number of examples of SolverStudio being used to solve interesting problems. Please let us know if we can list your SolverStudio solution here, and, even better, share it with other users.

Two of our Engineering Science students, Michael and Oscar, have developed a SolverStudio model for forming student groups in one of their large classes. This model can balance the groups across a wide range of factors, and produces a series of graphs to visualise the solution. Their spreadsheet model is available at https://github.com/odow/group-allocator

Marilize van Buisbegen from the University of Pretoria has written a report “A Volunteer Rostering Problem: Scheduling of moderators to provide optimal coverage in an on-line chess website” on using SolverStudio with PuLP to scehdule volunteers to support an on-line chess web-site.

Hi Andrew,

I am trying to solve a MIP model (GMPL language), but I want to use the GLPSOL options to limit the search duration (by setting a gap tolerance).

I found this on the GLPK Wiki:

glpsol –cuts –fpump –mipgap 0.001 –model problem.mod –data problem.dat

How can I enter this –mingap functionality in Excel? Since I am here editing the .mod file I believe?

Thanks,

Alex

GMPL should just let you pass these parameters, so copy the line above into your SolverStudio model code. Andrew

Hi Andrew, when solving my model (language GMPL), I find a optimal LP solution. After this is achieved, the solver also tries to find the MIP optimal solution, which it does. However, I want to output the values of the decision parameters of the LP solution to my Excel file. How can I achieve this?

E.g.:

* 44: obj = 2.211556871e+05 inf = 0.000e+00 (0)

OPTIMAL LP SOLUTION FOUND

Integer optimization begins…

+ 44: mip = not found yet >= -inf (1; 0)

+ 49: >>>>> 3.066614637e+05 >= 2.211556871e+05 27.9% (3; 0)

+ 74: mip = 2.253335908e+05 >= tree is empty 0.0% (0; 17)

INTEGER OPTIMAL SOLUTION FOUND

Thanks in advance

I suggest you solve the LP first (by relaxing the problem’s integer constraints, ideally via a GMPL option), and then solve the MIP second. Hope this helps. Andrew

Hi Andrew,

I have been using opensolver on many instances with great satisfaction. Thank you for making LP available on excel. I have now to solve a problem of the type of the warehouse-bars transportation one, with a large warehouse/bar cost matrix, and would like to use solverstudio for this. In order to speed-up the calculation, I would like to fix to zero some of the elements of the warehouse-bar solution matrix – typically those corresponding to routes too long to be used. Which modeling language should I use to make sure that this variable fixing results in a reduction of the number of optimization variables? I understand that AMPL does this, but AMPL limits the number of variables in its free version. Is there another modeling language -free of use- which shares this feature?

Any modelling language can do this; just add the “x[i,j]=0” constraints as additional constraints, eg “x[3,4]=0”. Andrew

I need to start somewhere but all the language options are confusing. Currently developing an excel spreadsheet with various constraints such as found in a rolling mill (time, yield, speed, weight, lengths, inventory, product,tons, etc) to aid in optimising the throughput in tons. LP seems to be the easier option but

where do I start as a recommendation? Excel solver ad in seems to be expensive for a learner of LP. Regards,

Zach

Hello,

I am thoroughly enjoying your work on this. It’s very helpful. Forgive me if I missed it, but is there an example implementing the transshipment problem but also assuming multiple commodities (for example, the steel problem but with additional commodities sharing the same capacity constraints).

Best,

Jeff

No, we don’t have such an example, sorry. Andrew

Hello, In order to use a variable with 3 indexes or more; how should I create the table and connect it with the parameter tables?

You use row and column labels (one of which is a tuple of two values); please look at the examples, eg Python Examples, to see how this works. Andrew

Hi,

I want to be able to use this tool in my big model, to do a simple optimization.

I want to minimize a cell in excel, which is a sum of a lot of cells that are results of a sumproduct between some variables X and some constanst variables Z.

The variables X are the only things I want to change, in order to minimize my target cell.

And I have some constrains that X has to be between 0 and 1, and some varaibles in X has to be 0.

I don’t know which example to follow in order to fulfill this.

Any help would be appreciated

You have a linear programme; try using that example sheet to build your model. Hope this helps, Andrew Mason. PS: We can direct you to commercial support if you need help with your model.

Thanks, which example sheet are you referring to sorry?

How does the commercial support work?

Not at my PC but it is the pre built models workbook which has an Linear programme sheet. Andrew

Hello Mr. Mason, excellent tool.

Could you give me a simple example (template) of using the Solverstudio with GAMS?

Thanks for you attention

Thanks for your post. The “GAMS(NEOS) Examples.xlsx” file is what you want; you can open it using the “SolverStudio : SolverStudio Examples” sequence on the Excel Data ribbon. I hope this is what you are seeking. Andrew