Algorithms To Live By #5: Optimal Stopping And Parking In The City
So far, we have talked about the secretary problem, or more generally, about optimal stopping, and how it can be useful to model when to sell a house, or how to look for a romantic partner. There are many more applications for optimal stopping and today I would like to talk about one of them, when to stop looking for another spot and park. Parking, for anyone in an urban environment, can be a painful experience. To address this problem, Brian Christian and Tom Griffiths went to Donald Shoup, the man described by Los Angeles Times as "the parking rock star" and UCLA's Distinguished Professor of Urban Planning.
Shoup argues that while there are many factors that affect parking, the single most important one is the occupancy rate. The occupancy rate is a number representing how many spots are taken percentually in a specific area. Furthermore, he states that many of the headaches of parking are consequences of cities adopting policies that result in extremely high occupancy rates. For example, let's say that there is an event downtown and a lot of people are attending. If the cost of parking right next door to the event is the same as parking further away, then everyone will want to park there instead of parking further away and walking a little bit. This only makes people waste more time, since they need to go see for themselves that spaces right next to the event are already taken. Additionally, people burn a lot of fossil fuel while they cruise looking for a spot.
Then, what is the solution? According to Shoup, a possible solution is implementing dynamic prices in the meters so that occupancy rates are held stable at around 85%. Therefore, the higher the occupancy rate in a specific area, the higher the price to pay for a spot there. A version of this approach is already being applied in downtown San Francisco. The reason Shoup would want to keep occupancy rate at around 85% is because the time and fuel spent looking for a spot increment exponentially as occupancy rates go up. For example, when occupancy rates go from 90% to 95%, it translates into cars taking twice as much time to find a spot. Another way of saying this is as follows: With an 85% occupancy rate, you would have to start looking for a spot around two blocks before your destination. On the other hand, with an occupancy rate of 99%, you would have to start looking for a spot about a quarter mile before reaching your destination.
Of course, this is a simplified solution for a very complicated problem. When presenting this solution, other variables are being ignored such as population density in cities as well as how many cars are owned per person on average. Maybe that is why when asked what was his secret weapon to confront the parking problem, Shoup answered: "I ride my bike".
No matter what parameters are taken into account in the model, the more vacant spots, the easier life gets. Nevertheless, policy makers seem to have failed in understanding that parking is not a simple resources and maximizing of utilization problem. Parking is a continuous process. A process that consumes attention, fuel, time and one that generates pollution and congestion. Even though it may seem counter-intuitive, empty spots on a highly desirable block may indicate that things are working properly.
Sources:
If you want to check out other thoughts that this awesome book has evoked, click on these past posts:
- Algorithms To Live By #1: Expectations, Authors, And Algorithms
- Algorithms To Live By #2: Optimal Stopping - How Long To Look For And When To Stop To Find The Best (The 37% Rule)
- Algorithms To Live By #3: The 37% Rule Applied To Partner Searching
- Algorithms To Live By #4: How Does The Secretary Problem Changes With More Information?
Best,
Very interesting indeed. I frequent San Francisco and my immediate reaction was "Why are parking meter prices different in different neighborhoods?" Many consumers are trained this way. They expect to pay the same price everyone else paid. Regardless of demand or time or when then buy, etc. Dynamic pricing is now accepted in the airline industry and hotel industry. These are relatively high priced goods so there is a wide range in prices. I think the challenge with parking is that its relatively cheap and people aren't that price sensitive. Or another way to look at it is cities won't increase prices to the point that consumers start to react. Prices currently range from $0.25 per hour to $6 per hour. A big range but $6 for a parking spot during a meeting in downtown San Francisco is not a big expense. SF won't make that spot cost say $50 per hour because they will be subject to the "only the rich can afford it" argument.
Thank you for bringing the specifics! Now I know the actual range of prices. You are completely right, the prices cannot be truly dynamic because you cannot start charging $50 per hour. Specially considering how economic differences have become wider over the last decades. Furthermore, you hit on another important aspect, consumers have been trained to pay the same price for "the same" product. To the naked eye of a consumer, it is not easily perceptible what the occupancy rate of a neighborhood is. Maybe if people understood the principle behind these different prices, they would be more willing to accept it and expand it.
I know this is not a bulletproof solution, but I think it a good approach.
It is an excellent approach. Didn't mean to sound negative on the whole article. I have read or seen Shoup interviewed before...good stuff.
You got a 27.84% upvote from @minnowvotes courtesy of @capatazche!
You got a 12.02% upvote from @joeparys! Thank you for your support of our services. To continue your support, please follow and delegate Steem power to @joeparys for daily steem and steem dollar payouts!
This post has received a 17.33% upvote from @lovejuice thanks to @capatazche. They love you, so does Aggroed. Please be sure to vote for Witnesses at https://steemit.com/~witnesses.