LAUSR

In pigeon-inspired optimization (PIO) algorithm [1], the computational process consists of two different stages that originate from the tools pigeons depending on during a flight. Based on this, its variants have been widely used in different domains and have achieved excellent results. However, most improvements were limited to the separate manipulation of the two independent iterative cycles; thus, the number of iterations for each stage must be set empirically. In [2], two independent computations were merged by using a transition factor to obtain the global optimum, but the linear conversion between the two processes was conducted in a rigid pattern, which created an issue wherein the coordination and allocation between the operators and coefficients would not be considered. Additionally, because it is typically used for a specific model, the internal parameters rely on the problem to be optimized, which results in a lack of adaptability.