Dichotomous search in ABC and its application in parameter estimation of software reliability growth models

ABC (Artificial Bee Colony) is one of the most recent nature inspired algorithm (NIA) based on swarming metaphor. Proposed by Karaboga in 2005, ABC has proven to be a robust and efficient algorithm for solving global optimization problems over continuous space. In this paper, we propose a modified version of the ABC to improve its performance, in terms of converging to individual optimal point and to compensate the limited amount of search moves of original ABC. In modified version called Dichotomous ABC (DABC), the idea is to move dichotomously in both directions to generate a new trial point. The performance of the proposed algorithm is analyzed on five standard benchmark problems and also we explored the applicability of the proposed algorithm to estimate the parameters of software reliability growth models (SRGM). The proposed algorithm presents significant advantages in handling variety of modeling problems such as the exponential model, power model and Delayed S Shaped model.