Evolutionary Algorithms - Multi-Objective Problems 

My pleasure - thank you.

You're welcome!

Typically speed and handling are the objectives, and the variables are what determine these objectives. You can optimise the objectives without defining any relationships between them, however, it's often beneficial if you have auxiliary knowledge about the problem (e.g. knowledge about the desired trade-off).

Thank you Dr. Rostami. I have 7 related variables - 4 discrete, 3 continuous, each of the same size n. They are intuitively related but it's hard to formalize that relationship quantitatively a priory without introducing subjective bias. I am comfortable with lexicographic preference ordering though. I am interested in identifying a set of non-dominated, pareto-optimal solutions where 2 of these variables are optimally maximized and 5 of these variables are optimally minimized. How do I go about: a) identifying appropriate optimization algorithm for this problem, and b) how do I formulate / setup this problem in Scipy/PyGMO etc. all of which seem to want a functional formulation? Thank you so much for your time and help.

I'm glad you found it useful. I've had a few requests for a video on NSGA-II - is there a particular part of the algorithm which is difficult to understand?

thanks for your prompt response! sure, I need am not clear on NSGA-II, especially for more than 2 objectives with more than 3 decision variables for each. thanks again

Hello Ênio. Your comment prompted me to create and publish an unplanned video, you can find it here ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-YFgtbxt7yhE.html. I hope you find it useful! All the best.

@@ShahinRostami 100% usefull, like always. Your book are amazing 😍