Implementing solvers for finding optimal planning schedules for periodic supply vessels based on research on the field.
Currently implementing a solution for finding the optimal amount of supply vessels and corresponding optimal sailing schedule for rigs based on supply demand.
Based on the paper by Borthen, Thomas, et al. [1].
array of shape (number of vessels, number of days in schedule, number of installations) representing on which day which vessel visits which installation and in what order.
First index is vesse, then day in schedule period. That array contains the order of which installations are visited, and in which order padded with zeros at the end. Zero also represents the depot. Only zeros represents the vessel staying at the depot that day.
In the following example the second vessel (index 1) does not sail on day one, visits installation 1 on day 2 and visits installation 2, 3 and 4 in that order on day 7.
array([[[0, 0, 0, 0],
[2, 0, 0, 0],
[3, 0, 0, 0],
[0, 0, 0, 0],
[0, 0, 0, 0],
[0, 0, 0, 0],
[0, 0, 0, 0]],
[[1, 0, 0, 0],
[0, 0, 0, 0],
[0, 0, 0, 0],
[4, 0, 0, 0],
[1, 2, 0, 0],
[0, 0, 0, 0],
[2, 3, 4, 0]]], dtype=int8)
Table of Contents
psvpp-solver
is distributed under the terms of the MIT license.
[1] Borthen, Thomas, et al. "A genetic search-based heuristic for a fleet size and periodic routing problem with application to offshore supply planning." EURO Journal on Transportation and Logistics 7.2 (2018): 121-150.