الفهرس | Only 14 pages are availabe for public view |
Abstract Optimization problem involving both continuous and discrete variables are able to describe many real world problems. A mixed variable pro- gramming problem is an optimization problem refers to mathematical programming with continuous variables and discrete variables (zero- one variables, integer variables or discrete variables). A particular class of such problems is called mixed variable programming which is important but di cult to solve. Di erent type of problems in eco- nomics, science, engineering, tra c and medicine can be reformulated as a mixed variable programming problem. So, in this work, we pay a great attention to solve this problem. Global optimization problems represent a main category of such problems. Global optimization refers to finding the extreme value of a given non-convex function in a certain feasible region. Such problems are classified in two classes; unconstrained and constrained problems. In this study, both global optimization problem classes; uncon- strained and constrained problems are considered. New hybrid ver- sions of genetic algorithm are proposed as promising solver for the considered problems. The proposed methods aim to overcome the drawbacks of slow convergence and random constructions of genetic Approval algorithm. In this hybrid metho ds, local search strategies are laid in- side genetic algorithm in order to guide them, especially, in the vicinity of local minima, and overcome their slow convergence, especially, in the final stage of the search. Data clustering is related to many disciplines and plays an important role in a wide range of applications. The applications of data clustering usually deal with large datasets and data with many attributes. Data clustering has found many applications, including do cument extrac-tion, image segmentation, market research, social network analysis, etc. Data clustering problem can be reformulated as a mixed variable programming. In order to examine the designed methodologies in this study, they are applied to solve the data clustering problem. |