📈 #13 Finding the sweet spot: when to choose exact vs heuristic algorithms in optimization

From 3 different perspectives: business, software development, and Operations Research.

According to Aristotle, moral virtue is the golden mean between two extremes.

That's why I've always thought that Aristotle laid the foundation for optimization. When you face a problem of this type, you have to know how to find the balance, that middle ground, between different perspectives.

Today on Feasible, I'm talking about the importance of balancing the workload in optimization from three different perspectives:

• The business.

• Software development.

• Operations research.

Let's go for it!

As someone who solves optimization problems, one of the main tasks is to choose how to solve the problem you are facing.

Life makes it easy for you: either you choose an exact method or you give a solution with a heuristic. Usually you know what to do.

The problem is: when do you choose one and when do you choose another?

💼 Act I: the business

It is the business itself that has optimization problems and needs to solve them. With this, it improves its financial results. One of the key questions that you always have to ask yourself is: will I be able to translate business needs into a mathematical model?

If not, then it will be very difficult for you to be able to continue on the side of the exact algorithm. Okay, you can always write your own exact algorithm, but I'm going to be honest with you: a solver like Gurobi encapsulates decades of knowledge, so I doubt that you or I are going to do better.

Reinventing the wheel is not the best thing you can do, and writing -again- a Branch & Bound with some heuristic to know which branch of the tree is the most promising is neither going to be faster nor more efficient than any solver on the market.

So if you can't translate those business needs into mathematical formulas...