TOC

• Authors
• Dual-Objective Scheduling of Rescue Vehicles to Distinguish Forest Fires via Differential Evolution and Particle Swarm Optimization Combined Algorithm
  • Project statement
  • Experiments Data
  • Algorithm ingredients
  • Results

Authors

• Guglielmi Matteo (Repo Owner) : @github
• Davide Guidolin (Collaborator) : @github, @website

Dual-Objective Scheduling of Rescue Vehicles to Distinguish Forest Fires via Differential Evolution and Particle Swarm Optimization Combined Algorithm

This repository contains a possible implementation of a "Dual-Objective Scheduling of Rescue Vehicles to Distinguish Forest Fires via Differential Evolution and Particle Swarm Optimization Combined Algorithm" with reference to the homonymous paper.
This project has been carried out as practical part of the course Bio-Inspired Artificial Intelligence course.
This project caught out attention due to its inner nature of being very close to an actual emergency situation where everything needs a very fast decision-making procedure. This algorithm allows to do so in a manner of seconds evaluating different possible solutions.

Project statement

This work has been carried out as application of Evolutionary and Differential algorithms to gain practical insights and make some experiences with bio-inspired algorithms.
Technically, this project is devised to provide a solution (i.e. scheduling plan) to an emergency scheduling of forest fires problem.
In the development of this work, several realistic factors have been taken into account such as :

• terrain conditions
• wind conditions (force and direction)
• number of available fire engines in the nearest fire-station
• extinguishing power of a vehicle and arrival time
• distances between points
• temperature of a fire point. The formulated problem is a complex non-linear issue since it consists in an integer programming problem whose complexity increases non-linearly with the number of fire points (and the possible solutions grow accordingly) and the constraints impact each other.

Experiments Data

The data used to achieve the results are the ones used by the authors of the reference paper. Anyway, for further detail please refer to the section "V. RESULTS AND ANALYSIS" of the project's report.

Algorithm ingredients

The flow chart followed during this work can be analyzed in the following figure:

Results

The results obtained with our implementation resembles the authors' ones (don't be tricked by the different scale in the images). It is worth mentioning that during the development of this project we've met some difficulties due to unclear explanations and missing of important variables' values. Despite that, the results turned out to be significant and satisfying.

Final Mark

The final mark for this project is $30/30$.