Skip to content

Optimized route for delivering ordered using one truck and k drones

Notifications You must be signed in to change notification settings

rajatcodes/Linear-Optimization

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 

Repository files navigation

Linear-Optimization

About Project

  A transportation company is evaluating a prototype system that combines trucks
and drones for last-mile delivery services. The test run considers a set of orders that
must be delivered to known locations. A delivery truck starts from a depot and visits
“launch sites” corresponding to the customers locations. From each launch site, the
truck deploys a series of drones that deliver the orders and return back to meet the
truck at the launch site. Once all the drones are recovered, the truck moves to the
next launch site and repeats the process until all orders are delivered.

Assumptions

1. We have a euclidean distance matrix between each delivery locations.
2. Assume that the depot is located at the origin (0,0).
3. The truck can deploy up to K drones at each stop in a launch site.
4. The drones have limited cargo capacity and as a result they can only visit 1 customer (not counting the starting point) before returning to the truck.
5. Each customer should be visited at least once by a drone. The customer locations used as launch sites will be visited by the truck and also will serve as the starting and ending point of a drone tour.
6. For each unit of distance traveled by the truck there is a cost of $10, whilefor each unit of distance traveled by a drone there is a cost of $3.

About

Optimized route for delivering ordered using one truck and k drones

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published