The Use of Original and Hybrid Flower Pollination Algorithm In Estimating The Parameters of Software Reliability Growth Models

Abstract

In order to assess software reliability, many software reliability growth models (SRGMs) have been used for estimation of reliability growth. . In this work, the parameters of (SRGMs) were estimated by using Flower Pollination Algorithm (FPA). Then, the (FPA) was hybrid with Real Coded Genetic Algorithm (RGA) to obtain Hybrid FPA (HFPA).The results that obtained from (FPA) are compared to the results of five algorithms: Particle Swarm Optimization (PSO), Artificial Bee Colony (ABC), the Dichotomous Artificial Bee Colony (DABC), Classic Genetic Algorithm (CGA) and the Modified Genetic Algorithm (MGA).The results showed that (FPA) outperformed the rest of the algorithms in parameters estimating accuracy and performance using identical datasets. Sometimes, the (DABC) showed better performance than (FPA).Other comparisons were made between (FPA) and (HFPA) and the results show that the hybrid algorithm outperformed the original one.