Hi there, I'm not much of a traveler myself but I visit this board fairly often.
I'd like to ask if anyone uses software for planning their trips?
I'm a software developer with intrest in logistic algorithms, and I've been wondering would people be interested in software that would allow them to pick a bunch of towns and find an optimal route for several criteria(cost, least time spent enroute).
The thing is, also it sounds easy, but finding a good solution for a problem of covering a set of cities is difficult and optimal routes cannot be guaranteed for trips with more than 65 cities(and finding a optimal route for a route with say 15-60s cities will take up a bunch of time).
So would anyone use a free piece of software(haven't thought about the platform yet) that would allow to generate sub-optimal routes?
This is a project I'm consider doing in my free time
I'd be very interested in that as a traveler and frankly there's probably some monetization potential if you could make that as an app and slap some ads on it.
I'm still a CS student so this might be impractical, but that sounds like something that a genetic algorithm would excel at for the larger problem sizes. Semi-randomly generate a first generation based on physical distance, give the user the option to choose between faster and more optimized pathing (fewer or more generations). I've never actually implemented a genetic algorithm but I've studied them in concept.
>>1064055
>soliciting quasi-solutions to the Traveling Salesman Problem on /trv/
You're in way over your head
>>1064055
This isn't the board for tackling Traveling Salesman Problem on. Talk to /g/.
But research more first, you clearly haven't thought about this enough to start asking for help.
Secondly, no one goes on 65 city holidays anyway. I've travelling to every major city in China and that's only 40. Where the fuck else would you be talking about? Every state capital in the USA is less than 65 and who does that kind of trip?
There are all sorts of heuristic simplifications that you could do to make the problem smaller but you're not even sure what you want to do, so you're not even ready to define the problem that you want to solve.
>>1064059
If you like GAs, you'd also be interested in Particle Swarm Optimization.
>>1064063
>>soliciting quasi-solutions to the Traveling Salesman Problem on /trv/
>You're in way over your head
This.
Unless you actually look at how long is typical in each city not gas/mileage, this ideal is a big fat fail and utterly useless.
I'd say there is some marketability for programs that might draw or plot on wold maps and record the route a person took, maybe with some mouseover abilities and website plug-in. Think Prezi slideshow on a big canvas that advances, but something better than a clickable map (which has the java code already done and reusable). Figure that out!
Routing is stupid, and it's never the issue at all, nor is it an integral part of the traveler outside of the US whatsoever. The issue is allll the other expenses of travel and interest draws that make the logistics of where to go and when to go more relevant. Watch any documentary on logistics.
