Green Vehicle Routing Problem

Hi! I hope you are well. I am asking whether the task I have at hand would be something manageable to do. I have a code that generates a hamiltonian cyclic path. What I just want to ask you is whether you can tweak it a bit. Basically, the data set it uses contains 15 or 16 refuel stations but they are represented as normal coordinates. What I want to do now to this algorithm is use these refueling stations. Imagine a car. I want to travel and visit all points. The car has a capacity of fuel, say 50 liters. It starts off from point A and starts visiting customers and if it runs out of fuel for the next customer, the car visits a refueling station and then continues with the customers. It is not required for the car to visit a refueling station obviously, if the 50 liters is enough. That is what I would like you to do.

This problem is referred to as a G-VRP (Green Vehicle Routing Problem) since the goal is to use greener methods of refueling (hence 'green') to cut down on CO2 emissions. Now, please refer to the attached image. This image example involves only one vehicle with a fuel tank capacity of say 50 gallons and fuel consumption rate of 0.2 gallons per mile. Three refueling stops (the triangles in the image) are available in the region. The vehicle begins its tour at depot D (starting point) and must visit customers C1 - C6 before returning to the depot. So it starts at the depot and ends at the depot. Lets say that the distance needed to visit all these customers is 339 miles. Then, we can appropriately compute that the vehicle would consume 67.8 gallons, which is 17.8 gallons more than the vehicle's capacity. Thus, the vehicle needs to needs to visit at least one refueling stop in order to serve all customers and and return at depot D. This problem basically takes into account the vehicle's fuel tank capacity limitation and chooses the optimal placement of refueling stop visit within the tour. Since our vehicle has now traveled to the refueling stop, the tour length is 15 miles longer than the minimum tour length, but we have managed to travel to all customers which otherwise would be impossible if we did not visit a refueling station.

Each Refueling stop may be visited more than once, or not at all again depending on the situation :) As in, if your fuel tank is enough to visit a particular set of customers then why would you go to refuel. So I am using a scientific paper and I have their data sets. Attached is the paper.

The graph is undirected. Ignore the pointy arrows. It is an undirected, complete graph.
Here is the code. I believe output of this code will be used as an input to the Green Vehicle Routing Problem.
If you can unzip "Algorithm_Path_Connection.zip", there is a file names "QuadTreeFinal.py". You can run it and see it will read the data points and save a plot with the Hamiltonian cyclic path.
The code I provided generates a cyclic Hamiltonian path so I am thinking maybe we can use that as an input for the G-VRP because to traverse the graph as per the algorithm, we need to build the graph first because the input of this task is a complete cyclic graph which that code is generating but you can check and then let me know.

I have also attached the EMH Large dataset. The txt file has 4 columns: a unique identifier field (ID), the type (d=depot, f=refueling stop, c=customer), and the longitude-latitude columns respectively.