An improved pigeon-inspired optimisation for continuous function optimisation problems

Pigeon-inspired optimisation (PIO) is a new heuristic searching algorithm with a simple structure that requires only simple parameters. However, analogous to other intelligent algorithms, the limited optimisation method and the swarm diversity eroded its global search ability. To resolve this issue, this paper presents an improved pigeon-inspired optimisation (IPIO). First, we analyse the shortcomings of PIO systematically from its construction and use the Markov chain to quantitatively expound its convergence, proving that the algorithm can converge to the global optimum with probability one under suitable conditions. Second, a new solution generating method is introduced that tackles the limitation of the local optimum. Finally, 29 benchmark functions are used to test the performance of IPIO. The computational results show that the presented IPIO is superior to other improved versions of PIO proposed in recent literature, including MPIO, CMPIO, and HCLPIO, on most test functions.