Surveillance camera placement optimization using Particle Swarm Optimization (PSO) algorithm and Mixed-Integer Linear Programming (MILP) model / `Ain Safia Roslan

This project explores the optimization of surveillance camera placements using Particle Swarm Optimization (PSO) and Mixed-Integer Linear Programming (MILP). PSO, inspired by the social behaviour of birds flocking or fish schooling, is a heuristic algorithm known for its flexibility and exploration capabilities. On the other hand, MILP is a deterministic optimization approach that provides precise solutions through linear programming. The project aimed to find optimal camera placements to minimize the total number of cameras used while maximizing coverage, and to perform a comparative analysis between PSO and MILP. MATLAB was chosen as the primary software due to its robust capabilities in numerical computing and optimization, enabling efficient implementation and analysis of both algorithms. The study applied these optimization techniques to various Binary Integer Programming (BIP) matrix sizes (11×9, 39×24, and 172×49) representing the same 2D layouts, to evaluate their performance in different spatial configurations. The results indicated that both PSO and MILP could achieve high coverage rates, with PSO demonstrating superior flexibility and adaptability in identifying optimal camera placements.