Community Spring Cleaning week is here! Join your fellow Maveryx in digging through your old posts and marking comments on them as solved. Learn more here!

Alteryx Designer Desktop Discussions

Find answers, ask questions, and share expertise about Alteryx Designer Desktop and Intelligence Suite.
SOLVED

Optimizing a Function Given Parallel Corresponding Constraints

Henry
5 - Atom

Hello,

 

So I'm still pretty new to using Alteryx on the predictive/optimization front and could use a little help with this problem. I am trying to build a workflow designed to optimize one value, lets call it revenue, while constrained by a second, lets call it cost.

 

Specifically, in my data set (An example one is attached here) I have ~150 "Areas" and each one has the option to select a revenue neutral option, and increase option, and a decrease option (Columns B-D). There are in turn corresponding costs associated with and dependent on each selection (Columns F-H). The goal is to maximize the sum of total revenue across all areas while costs are constrained at some given level (In this case say 0). 

 

I can kind of think of a brute force method that might work with a few area options but was hoping that someone might be able to provide a more elegant and efficient solution.

 

Thank You in advance!

 

 

1 REPLY 1
Joe_Mako
12 - Quasar

I followed the example for "Optimization with Manual Mode and Pre-Existing Data" from the Sample Workflow "1 Optimization Model Input Modes" in Prescriptive Analytics:

menu selection.png

 

And made the attached:

Optimization.png

 

- RecordID for joining and proper sorting of the columns when performing the cross tab later
- Arrange for reshaping
- Formula tools for adding in constants and renaming
- Cross tab for making the columns
- Dynamic Rename to rename them to match
- Join and Union tools to make the input matrices
- Sort, as done in the example
- Optimization configured to Maximize Objective and Dense matrix, variables in rows
- Join to bring it back to the arranged data
- Summarize tools to aggregate to different levels

Labels