Can Pollination run genetic algorithms for multi-objective optimization?

I would like to use octopus or wallacei with ladybug tools to perform a multi-objective optimization simulation of energy consumption and daylighting in a residential single-family building. And I need to perform a large number of iterations using genetic algorithms.
I wonder whether the pollination platform is capable of running multi-objective optimization simulations?

Hi, @omen - Pollination is capable of doing this, but I don’t think you can use it with the native Grasshopper optimization components. Here is what the process can look like.

  1. Generate the first iteration of the input geometries.
  2. Submit them to Pollination to run in parallel.
  3. Download the results, and post-process them.
  4. Stop if you have achieved your goals.
  5. Otherwise, feed them to the genetic algorithm and get the input values for the next generation.
  6. Create a new generation of models.
  7. Go to step 2.

This is a very old post that I wrote about 10 years ago and does something similar to what I’m suggesting above:

Here is a paper that describes the process in more detail: