An experiment involving sweeping floors, simulated annealing, and why algorithms that optimize for the wrong thing ruin your life.

Programming Digest is a curated collection of articles, tutorials, and resources for software developers covering a wide range of programming languages, frameworks, and technologies. With a focus on practical tips, code snippets, and industry insights, Programming Digest provides a resource for developers looking to stay updated with the latest trends and advancements in software development.

Programming Digest

A computer science student turned a supermarket floor sweeping job into an optimization problem using simulated annealing and the traveling salesman problem. The initial solution minimized distance but created an impractical path with excessive turns. Adding a turn penalty to the cost function produced a more realistic, human-friendly route. This experiment illustrates how optimizing for easily measurable metrics (distance, engagement, profit) instead of actual goals (usability, wellbeing, sustainability) leads to technically correct but practically useless or harmful outcomes in algorithms, social media, AI, and business.

I got paid minimum wage to solve an impossible problem.

Step 1: Turn reality into a (overly simple) model.

Step 6: Realize life is not so good after all.

Actually you were paid minimum wage to sweep the floor, and apparently should have been paid less since you used some of that time to tinker with a problem whose solution is well-known and called a Roomba – you even joked about it in the article.
What’s fascinating is that you had a whole we-live-in-a-society epiphany from this and managed to miss that the company was already optimizing for the right cost function, which is not shaving a couple minutes off a janitor’s task.

I like when people passionately overcomplicate things just because they are curious. Nice job 👌

neat exploration! This was a fun journey leading to what I think is an often-needed reminder

If every janitor was that dedicated, we’d have flying cars by now.