Search

Abstract: A modified Artificial Bee Colony algorithm to solve constrained numerical optimization problems is presented in this paper. Four modifications related with the selection mechanism, the scout bee operator, and the equality and boundary constraints are made to the algorithm with the aim to modify its behavior in a constrained search space. Six performance measures found in the specialized literature are employed to analyze different capabilities in the proposed algorithm such as the ability and cost to generate feasible solutions, the capacity and cost to locate the feasible global optimum solution and the competency to improve feasible solutions. Three experiments, including a comparison with state-of-the-art algorithms, are considered in the test design where twenty four well-known benchmark problems with different features are utilized. The overall results show that the proposed algorithm differs in its behavior with respect to the original Artificial Bee Colony algorithm but its performance is improved, mostly in problems with small feasible regions due to the presence of equality constraints. [Copyright &y& Elsevier]

Abstract: The filled function method is an efficient approach for finding global minimizers of multi-dimensional and nonlinear functions in the absence of any restrictions. In this paper, we give a new definition of filled function and the idea of constructing a new filled function, and then a new class of filled functions with one parameter on the basis of the new definition, which possesses better quality, is presented. Theoretical properties of the new class of filled functions are investigated. A new algorithm is developed from the new filled function method. The implementation of the algorithm on seven test problems with dimensions up to 30 is reported, and comparisons with other filled function methods demonstrate that the new algorithm is more efficient. [Copyright &y& Elsevier]

Abstract: In this paper, a nonmonotone trust region algorithm for unconstrained optimization problems is presented. In the algorithm, a kind of nonmonotone technique, which is evidently different from Grippo, Lampariello and Lucidi’s approach, is used. Under mild conditions, global and local convergence results of the algorithm are established. Preliminary numerical results show that the new algorithm is efficient. [ABSTRACT FROM AUTHOR]

Abstract: The aim of this paper is to incorporate the preconditioned gradient path in a nonmonotone stabilization algorithm for unconstrained optimization. The global convergence and locally superlinear convergence are established for this class of algorithms. Finally, we report in details the numerical results which show the effectiveness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Abstract: In last decades, there has been much effort on the solution and the analysis of the mixed complementarity problem (MCP) by reformulating MCP as an unconstrained minimization involving an MCP function. In this paper, we propose a new modified one-step smoothing Newton method for solving general (not necessarily ) mixed complementarity problems based on well-known Chen–Harker–Kanzow–Smale smooth function. Under suitable assumptions, global convergence and locally superlinear convergence of the algorithm are established. [Copyright &y& Elsevier]