Drosophila Food-Search Optimization Kedar Nath Das ⇑ , Tapan Kumar Singh Department of Mathematics, NIT Silchar, Assam, India article info Keywords: Tournament selection Redundant Search (RS) Modified Quadratic Approximation (mQA) G-protein-coupled-receptor (GPCR) Ligand abstract The method of finding optimal solution to an optimization problem is a recent challenge for the researchers. In order to solve an optimization problem many evolutionary methods have been introduced as alternate paradigms. In this paper an extensive efforts has been made to solve unconstrained optimization problems by proposing a new algorithm namely Drosophila Food-Search Optimization (DFO) Algorithm. DFO mimics the food-search mech- anism of a fly in nature based on called Drosophila Melanogaster. To maintain the diversity throughout the population during simulation, an exploration operation has been developed to generate new individuals. A set of well known benchmark function have been used to validate the better performance of DFO. The experimental results confirms that the proposed technique DFO performs better than some well known existing algorithms like Differential Evolution (DE), Intersect Mutation Differential Evolution (IMDE) algorithm, self-adaptive DE (JDE), improved Particle Swarm Optimization (PSO) algorithms, Artificial Bee Colony (ABC) algorithm and Bee Swarm Optimization (BSO) algorithm. Further two real world problems namely Gas Transmission Compressor Design and Optimal Capacity of Gas production facilities are considered and the better performance of DFO is confirmed. Ó 2014 Elsevier Inc. All rights reserved. 1. Introduction Optimization is ubiquitous and spontaneous process that performs on integral part of our day to day life. In the most basic sense, it can be defined as an art of selecting the best alternatives among a given set of options. These problems arise in var- ious disciplines such as engineering designs, agricultural sciences, manufacturing systems, economics, physical sciences, pat- tern recognition, etc. In view of practical utility in solving optimization problem, an efficient/robust computational method is essential. In spite of the existence of a number of deterministic methods, the probabilistic/evolutionary approaches based on nature inspired analogy became more popular in recent years, as they do not need any auxiliary information like differentiability and continuity of the problem in hand. Among them, the most referred algorithms in the literature are Genetic Algorithm (GA) [1], Particle Swarm Optimization (PSO) [2], Differential Evolution (DE) [3], Ant Colony Optimization (ACO) [4], Evolu- tionary Programming (EP) [5], Diversity Guided Evolutionary Programming (DGEP) [6], Self-Adaptive DE (JDE) [7], Fruit Fly Optimization Algorithm (FOA) [8], Structural Optimization (SO) [9], etc. However, most of these methods suffer with the involvement of (i) complicated mechanism, (ii) computational burdensome, (iii) premature convergence, (iv) trapping in some local minima and (v) fine tuning of many parameters. In order to minimize these shortcoming to some extent, an extensive efforts is made in this paper by introducing a new and robust technique namely Drosophila Food-Search Optimi- zation (DFO) Algorithm. DFO mimics the food-search mechanism of a fly to search the food with a minimal effort. The de- tailed description presented in Section 2. 0096-3003/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amc.2014.01.040 ⇑ Corresponding author. E-mail addresses: kedar.iitr@gmail.com (K.N. Das), tksingh1977.nits@gmail.com (T.K. Singh). Applied Mathematics and Computation 231 (2014) 566–580 Contents lists available at ScienceDirect Applied Mathematics and Computation journal homepage: www.elsevier.com/locate/amc