14 ms·

I'm not sure how that last part plays out. If you need 100 cars active at rush hour and 10 active the rest of the day, then sure those 10 might have 90% utiliza

• IkmoIkmo 8y ago
  Well, suppose you have a fleet of 100 cars. In the morning and evening, 100 people travel for 1 hour to work and back during rushhour. That's 2 hours in total, or 100% utility of the fleet during 8% of the day. Now suppose that every hour of the rest of the day, there's only 10 people making a trip of 30 minutes. That means there's only 10% of the traffic volume compared to rushhour on the road. But you're seeing 22 hours of two groups of 10 people making a trip every hour. That's 440 trips. Now suppose that the two groups of people, rush and non-rush, all used their own cars. You'd have 100 cars for the rushhour group and 440 cars in the non-rushhour group. You get to 540 cars needed. Now suppose the two groups shared vehicles. You just need 100 cars still to make the same amount of trips. Under these (made up) assumptions, you can reduce the amount of cars needed by 5x. Then there's the notion that you can get dynamic pricing, e.g. look at Uber, but for semi-public transport. This incentivises people and companies to distribute their travel in a better manner, which also allows a system where producers of congestion pay for it. Thereby you can distribute load better, increase utilization, decrease the number of vehicles needed even further. But indeed, you're unable to get to 100% utilization because demand fluctuates during the day. But even increasing it from 10% to 20%, or 30%, would introduce massive changes to the market size.