Parameter Estimation of Nonlinear Systems Using L` evy Flight Cuckoo Search Walid M. Aly and Alaa Sheta Abstract Metaheursitc algorithms are used to solve hard optimization problems which can not be solved using traditional approaches within reasonable time and using feasible resources. One of the new natural inspired metaheursitc algorithms is the Cuckoo Search (CS) which is stimulated by the brood parasitism of some Cuckoo species. In this research, we explore the application of CS to solve the problem of parameter estimation of a nonlinear manufacturing process model. An industrial metal cutting system is used to examine the effectiveness of the proposed approach and also to compare CS with other metaheuristic approaches like genetic algorithm and particle swarm optimization. Results shows the high efficiency and robustness of CS when applied to the problem of parameter estimation of a nonlinear system model. 1 Introduction Nature inspired algorithms use inexact approaches to solve computationally hard problems, they include different approaches like genetic algorithms (GAs), neural networks (NN), particle swarm optimization (PSO), Bat Algorithm, Water drops algorithm, fuzzy logic and Cuckoo Search (CS). As theses algorithms proved to be efficient, many researches investigated their usage to solve various complex in- dustrial and engineering problems, among these problems is the system modeling problem. System modeling is the concept of representing the behavior of a system by presenting it in a form of a mathematical equation or set of equations. Parame- ter estimation of a model is a complex optimization problem that standard analytic approaches might fail to solve [7]. To estimate these parameters, search procedures like least square estimation, likelihood and instrumental variable methods [1] can be applied. These procedures aim to minimize the error between the actual model and the predicted model, although they can usually provide good results, they have no exact solution and they suffer from efficiently reduction in the presence of noise. As traditional techniques would fail to reach satisfactory solutions for the parame- ter estimation problem, different nature inspired algorithms have been investigated Walid M. Aly College of Computing and Information Technology, Arab Academy for Science, Technology & Maritime Transport, Egypt. e-mail: walid.ali@aast.edu Alaa F. Sheta Computers and Systems Department, Electronics Research Institute (ERI), Cairo, Egypt. e-mail: asheta66@gmail.com