International Journal of Information, Communication and Computing Technology Jagan Institute of Management Studies, New Delhi _________________________________________________________________________________________________________ 1 Assistant Professor, Department of Computer Science & IT, Jagannath University, Jaipur 2 Professor, Department of Mathematics, Jagannath University, Jaipur 3 Assistant Professor, Department of Computer Science, St. Xavier’s College, Jaipur Email: 1 sandeep.kumar@jagannathuniversity.org , 2 vivek.sharma@jagannathuniversity.org , 3 rajanikpoonia@gmail.com Copyright ©IJICCT, Vol II, Issue II (July-Dec2014): ISSN 2347-7202 96 Self-Adaptive Spider Monkey Optimization Algorithm for Engineering Optimization Problems 1 Sandeep Kumar, 2 Vivek Kumar Sharma, 3 Rajani Kumari ABSTRACT Algorithms inspired by intelligent behavior of simple agents are very popular now a day among researchers. A comparatively young algorithm motivated by extraordinary behavior of Spider Monkeys is Spider Monkey Optimization (SMO) algorithm. SMO algorithm is very successful algorithm to get to the bottom of optimization problems. This work presents a self-adaptive Spider Monkey optimization (SaSMO) algorithm for optimization problems. The proposed strategy is self-adaptive in nature and therefore no manual parameter setting is required. The proposed technique is named as Self-Adaptive Spider Monkey optimization (SaSMO) algorithm. SaSMO gives better results for considered problems. Results are compared with basic SMO and its recent variant MPU-SMO. KEYWORDS Spider Monkey Optimization Algorithm, Swarm intelligence, Engineering optimization problems, Nature Inspired Algorithms. 1. INTRODUCTION Intelligent food foraging behavior of Spider Monkeys inspired J. C Bansal and his colleagues to develop a new algorithm. J. C. Bansal et al. [15] proposed Spider Monkey Optimization (SMO) algorithm in year 2013. It is a recent popular algorithm motivated by intelligent behavior of simple natural agents. This algorithm is stimulated by means of the extra ordinary behavior of Spider Monkeys while searching for food. This algorithm based on fission-fusion social structure (FFSS). It falls into category of Nature Inspired Algorithms (NIA) that is inspired by some natural phenomenon or extraordinary behavior of intelligent insects. NIAs include Evolutionary algorithms, Immune algorithms, neural algorithm, Physical algorithms, Probabilistic algorithms, stochastic algorithms and Swarm algorithms based on their source of inspiration and motivation. In last decade a number of new algorithms are developed by researchers that are inspired by natural phenomenon. Some recent development includes Water Wave Optimization (WWO) algorithm [2]. WWO algorithm is motivated by the shallow water wave theory. It mimics the eye catching phenomenon of water waves, such as propagation, refraction, and breaking. A. Brabazon et al. [3] proposed a new population based strategy based on social roosting and foraging behavior of one species of bird and named it Raven Roosting Optimization algorithm. Based on the living behaviors of microalgae, photosynthetic species an algorithm was developed by SA Uymaz [4] named as Artificial Algae Algorithm (AAA). S. Mirjalili et al. [6] proposed a novel nature inspired strategy for continuous optimization problems namely Grey Wolf Optimizer (GWO). GWO algorithm mimics the leadership hierarchy and hunting mechanism of gray wolves. A. Askarzadeh incepted Bird Mating Optimizer (BMO) [7] for designing most favorable searching techniques. This algorithm emulates the behavior of bird species figuratively to breed broods with superior genes. X. Li et al. [11] developed Animal Migration Optimization (AMO) Algorithm that mimics migration activities of animals. SMO algorithm consists of population of budding solutions like other population based algorithms. In this algorithm budding solutions are represented by food sources of spider monkeys. The superiority of a food source is decided by calculating its fitness. The SMO algorithm is comparatively an easy, rapid and population based stochastic search strategy. While searching for optimal solution this algorithm need to maintain balance between two basic activities named the assortment process, which make sure the exploitation of the preceding knowledge and the adaptation process, which empowers exploring diverse fields of the search space. However, it has been observed that SMO algorithm is very good in exploration of local search reason and exploitation of best feasible solutions in its immediacy [1] [5]. Therefore, to solve a complex problem like parameter estimation for frequency-modulated sound wave this paper uses a new variant of SMO algorithm. The proposed algorithm is adaptive in nature as it automatically modifies the radius of search area during local leader phase and local leader decision phase in order to update position along with fitness based position update. Fitness of a solution decides its quality. There are number of methods to calculate fitness of a function but it must include value of function. There are very few research papers on SMO algorithm in literature as it is very young algorithm. Recently S. Kumar et