Behavioral Study of Drosophila Fruit Fly and Its Modeling for Soft Computing Application

Behavioral Study of Drosophila Fruit Fly and Its Modeling for Soft Computing Application

Tapan Kumar Singh (National Institute of Technology, India) and Kedar Nath Das (National Institute of Technology Silchar, India)
DOI: 10.4018/978-1-4666-9885-7.ch003
OnDemand PDF Download:
No Current Special Offers


Most of the problems arise in real-life situation are complex natured. The level of the complexity increases due to the presence of highly non-linear constraints and increased number of decision variables. Finding the global solution for such complex problems is a greater challenge to the researchers. Fortunately, most of the time, bio-inspired techniques at least provide some near optimal solution, where the traditional methods become even completely handicapped. In this chapter, the behavioral study of a fly namely ‘Drosophila' has been presented. It is worth noting that, Drosophila uses it optimized behavior, particularly, when searches its food in the nature. Its behavior is modeled in to optimization and software is designed called Drosophila Food Search Optimization (DFO).The performance, DFO has been used to solve a wide range of both unconstrained and constrained benchmark function along with some of the real life problems. It is observed from the numerical results and analysis that DFO outperform the state of the art evolutionary techniques with faster convergence rate.
Chapter Preview


Optimization is an art of selecting best alternatives amongst the set of individuals. Henceforth, the collective intelligent behavior of insect groups such as flocks of birds, colonies of ants, school of fish’s and swarms of bees have attracted the attention of researchers. The aggregative behavior of insects is known as swarm behavior and the collective phenomenon of collective model swarms have been studied by Entomologists, while engineers have applied these models for solving complex real-world problems, as because, it has some advantages like scalability, fault tolerance, adaption, speed, modularity, autonomy and parallelism (Kassabalidis, EI-Sharkawi, Marks, Arabshai, & Gray 2001) etc.

Now a day, a number of complex real life problems arising in the field of Engineering, Science, Industry and Finance which were turn out to be nonlinear optimization problems associates with many constraints. For this reason, the need for designing efficient, reliable and robust optimization techniques is required. According, to No Free Launch Theorem (Wolpert, & Macready, 1995), no single algorithm exists in order to solve all optimization problems and therefore, a number of optimization techniques are available in literature to deal with non-linear problems both constraint and unconstraint in nature.

The mathematical form of the general nonlinear optimization problem is


subject to

The set 978-1-4666-9885-7.ch003.m03 is called the feasible region and 978-1-4666-9885-7.ch003.m04 is the objective function, 978-1-4666-9885-7.ch003.m05are inequality constraints, 978-1-4666-9885-7.ch003.m06 are equality constraints and978-1-4666-9885-7.ch003.m07, 978-1-4666-9885-7.ch003.m08 are the lower and upper bounds of the decision variables978-1-4666-9885-7.ch003.m09. A point 978-1-4666-9885-7.ch003.m10 is called local minima of 978-1-4666-9885-7.ch003.m11 if 978-1-4666-9885-7.ch003.m12 where 978-1-4666-9885-7.ch003.m13 is an978-1-4666-9885-7.ch003.m14 neighborhood of 978-1-4666-9885-7.ch003.m15 and 978-1-4666-9885-7.ch003.m16is called the global minima of 978-1-4666-9885-7.ch003.m17if 978-1-4666-9885-7.ch003.m18for all 978-1-4666-9885-7.ch003.m19.

Complete Chapter List

Search this Book: