The Drivers Cooperative: New York’s driver-owned ride-hailing app
1–10 of 137 posts
Re: The Drivers Cooperative: New York’s driver-owned ride-hailing app
#2>Scheduling trips is a tricky calculation, and it all comes down to routing — you want drivers to spend as much of their shift as they can actually driving clients around, but you have to leave enough time in-between trips so the driver isn’t late for their next client. So far, the only way to do this successfully has been by using a team of human dispatchers, though the co-op is working on an automated solution using artificial intelligence.
I worry about this. Travelling salesman is a solved problem. Throwing "AI" into the mix sounds like someone was sold a bridge.
Re: The Drivers Cooperative: New York’s driver-owned ride-hailing app
#3>Scheduling trips is a tricky calculation, and it all comes down to routing — you want drivers to spend as much of their shift as they can actually driving clients around, but you have to leave enough time in-between trips so the driver isn’t late for their next client. So far, the only way to do this successfully has been by using a team of human dispatchers, though the co-op is working on an automated solution usin…
Maybe by AI they mean it uses big data as well like historical traffic to more accurately set the weights of the vertices.
Re: The Drivers Cooperative: New York’s driver-owned ride-hailing app
#4>Scheduling trips is a tricky calculation, and it all comes down to routing — you want drivers to spend as much of their shift as they can actually driving clients around, but you have to leave enough time in-between trips so the driver isn’t late for their next client. So far, the only way to do this successfully has been by using a team of human dispatchers, though the co-op is working on an automated solution usin…
Maybe by AI they mean it uses big data as well like historical traffic to more accurately set the weights of the vertices.
Possibly, but today EVERY nontrivial piece of software is "AI". It's the new "synergy" for dipshits spouting nonsense.
FYI: in MBA speak, "AI" is now anything that can't be hacked together on top of airtable, outsourced, or done by someone with a GED and six months of bootcamping. If you need someone who isn't perceived as interchangeable cog labor to help build the thing then it's now called AI. Was just in a meeting where a major component for an optimizing compiler was called "the AI".
Re: The Drivers Cooperative: New York’s driver-owned ride-hailing app
#5>Scheduling trips is a tricky calculation, and it all comes down to routing — you want drivers to spend as much of their shift as they can actually driving clients around, but you have to leave enough time in-between trips so the driver isn’t late for their next client. So far, the only way to do this successfully has been by using a team of human dispatchers, though the co-op is working on an automated solution usin…
https://en.wikipedia.org/wiki/Travelling_salesman_problem
Traveling Salesman is NP-Hard. Don't think it is solvable with current technology.
Re: The Drivers Cooperative: New York’s driver-owned ride-hailing app
#6>Scheduling trips is a tricky calculation, and it all comes down to routing — you want drivers to spend as much of their shift as they can actually driving clients around, but you have to leave enough time in-between trips so the driver isn’t late for their next client. So far, the only way to do this successfully has been by using a team of human dispatchers, though the co-op is working on an automated solution usin…
Travelling salesman is a largely solved problem, but I'd say this problem is larger.
As soon as humans and the real world are involved, seemingly contradictory criteria need to be met.
Algorithms like A* search may not be able to capture these criteria robustly enough.
Example: a driver may be assigned the optimal route, but because they are human will not want to sit and wait for 15 minutes between several of their rides
A* is great for path finding but it's not a global panacea for all subproblems related to ridesharing
Re: The Drivers Cooperative: New York’s driver-owned ride-hailing app
#7>Scheduling trips is a tricky calculation, and it all comes down to routing — you want drivers to spend as much of their shift as they can actually driving clients around, but you have to leave enough time in-between trips so the driver isn’t late for their next client. So far, the only way to do this successfully has been by using a team of human dispatchers, though the co-op is working on an automated solution usin…
Travelling salesman is a largely solved problem, but I'd say this problem is larger. As soon as humans and the real world are involved, seemingly contradictory criteria need to be met. Algorithms like A* search may not be able to capture these criteria robustly enough. Example: a driver may be assigned the optimal route, but because they are human will not want to sit and wait for 15 minutes between several of their…
This is true, however these known criteria can be added in, thats the whole point of the heuristic in A*.
Likely one of Uber/Lyfts greatest strengths in their algos is discovering the E[X] of those heuristics. This could probably be captured from the human dispatchers as well.
Re: The Drivers Cooperative: New York’s driver-owned ride-hailing app
#8>Scheduling trips is a tricky calculation, and it all comes down to routing — you want drivers to spend as much of their shift as they can actually driving clients around, but you have to leave enough time in-between trips so the driver isn’t late for their next client. So far, the only way to do this successfully has been by using a team of human dispatchers, though the co-op is working on an automated solution usin…
https://en.wikipedia.org/wiki/Travelling_salesman_problem Traveling Salesman is NP-Hard. Don't think it is solvable with current technology.
Surprisingly you can solve massive TSPs using heuristics. The solutions are not guaranteed to be optimal, but they can get very close to optimal without a crazy amount of computing power.
The Concorde TSP solver can apparently solve instances with 85.6k cities to optimality. Pretty amazing!
Re: The Drivers Cooperative: New York’s driver-owned ride-hailing app
#9>Scheduling trips is a tricky calculation, and it all comes down to routing — you want drivers to spend as much of their shift as they can actually driving clients around, but you have to leave enough time in-between trips so the driver isn’t late for their next client. So far, the only way to do this successfully has been by using a team of human dispatchers, though the co-op is working on an automated solution usin…
https://en.wikipedia.org/wiki/Travelling_salesman_problem Traveling Salesman is NP-Hard. Don't think it is solvable with current technology.
Only in the worst case and large N. For practical problems, it’s solvable for tens of thousands of nodes.
Re: The Drivers Cooperative: New York’s driver-owned ride-hailing app
#10One difference I noticed is that drivers are basically employees - they sign up for a shift and take all rides assigned to them.
I tried to understand why they are almost profitable and Uber/Lyft not and as best as I can tell it's simply because Uber/Lyft are trying to grow and expand, and this co-op is not.