CVRP_SCLM
Description
The inclusion of M1 (uncollected load) and M2 (vehicle idle capacity) penalties in the CVRP solution, as well as CVRP-SCLM split deliveries, makes route design for the collection of agricultural waste such as poultry manure more flexible. The data present the geospatial location of 63 poultry farms capable of delivering manure to feed a biorefinery associated with microalgae cultivation. The distance matrix between each location is also presented. The classic CVRP, PC-VRP, and SD-VRP (.ypynb) models from the Google Collaboratory are presented, with which the proposed model was compared using boxplots.
Files
Steps to reproduce
Google My Maps was used to identify the geospatial location of 63 poultry farms where poultry manure is generated. A distance matrix was generated between each node and a biorefinery, whose location was simulated at the coordinate (3.95587; -76.29103) in the Municipality of Buga, Valle del Cauca, Colombia. Using information from Fenavi, the average amount of chicken manure generated by each poultry farm was estimated, determining an average of 10 tons/day. Taking into account cost reduction, a vehicle fleet (22 vehicles) with two axles and a maximum load capacity of 31 tons was identified. Using OR-Tools, a model was formulated to respond to the need to transport low-value items such as agricultural waste. The obtained model was compared with other base models to see its effectiveness and relevance.
Institutions
- Universitaria Agustiniana