← Back

8. Simplex Stepping

The term “simplex stepping” is inspired by a concept from a professor on my college campus named George Dantzig. It made a big impression on me when I first learned about it in a mathematical sciences course. It wasn’t the math mechanics that caught my attention, but the sheer elegant beauty of Dantzig’s seminal insight and what struck me as its philosophical implications.

Consider the following challenge: You want to find an optimal outcome to a huge complex problem that has many variables; furthermore, each variable has many possible values. One approach might be to calculate every possible permutation of the mix of values and then select its best result. But suppose there are too many possibilities for even the most powerful computer to process them all. How could you find an optimal result? Dantzig showed that it is possible to move in an iterative series of small steps that will lead you to an optimal outcome without ever needing to calculate the total set of possibilities. His approach proved reliable, and it revolutionized optimization. He called it the “simplex algorithm