Optimal Swimming Slot

14 Mar 2025 Comments

As Most of u probably know I live in Munich/Germany. And since we lived in Vietnam I got hooked on swimming - maybe not really hooked, but I enjoy it. I learned to freestyle for over 1 km in the sea in Vietnam, and from time to time I work on keeping alive this skill here in Munich.

The problem is, in Munich u need a public pool (because I don’t have a private one 😉), and public pools tend to be full & crowded. Luckily, the SWM (the Munich public services) provide a website that tells us how busy the different public swimming pools are.

Even thow I work 40 hours / week (or so…), I might have some flexibility as of when I go swimming: before work, after work, maybe even at lunchtime. And the question arises, when best to go. When are the pools the less crowded?

For example: I suspect that going as early as possible in the morning is not the smartest, as many sportive white-collar worker do so. So maybe it’s smarter to have tea with my wife in the morning, and then go for a swim and to the office.

The best about this problem: it’s a typical machine learning problem 😉

So this would be the plan:

• Build a scraper that gathers the pool occupancy every 10 minutes and stores it somewhere
• Train a machine learning model on that data
• Build a UI that asks when u could go, and that gives u advice when u should go
• Extra features: take into account week-ends and public holidays, pool-features (i.e. I prefer swimming in a 50m pool)

Anyone up for building such a tool? Send me an email if u want to hack 😉