| In recent years,Virtual Software Research and Development Organizations(VSR&DO)get more and more attentions,and the virtual enterprise on different views has been studied.As far as we know,the research about VSR&DO has not been reported.This paper first proposes the concept of VSR&DO,compares it with the traditional software development organizations,and further describes the results of the virtual enterprise partner's choice at home and abroad,finally introduces the comprehensive evaluation methods and genetic algorithms.In others,based on the research,partner selection of VSR&DO is a multi-objective optimization problem,this paper filters the defending deceives,and resolves the problem using the selection genetic algorithm of the multi-objective elitist. The main idea of genetic algorithm is building a multi-objective essential combination optimize problem of a multi-condition restriction based on a series and parallel connection reduction regulation,separating a multi-objective problem to some single objective optimize problems,and parallel searching use genetic algorithm in every single objective population.In the process of descendiblity,each of populations can select its probability of the crossover and mutation automatically according to its fitness.With every generation in evolution,by an elitist selection strategy the parallel search speeds can improve rapidly,finally all of the Pareto optimal solutions of elitist population are the solutions of the multi-objective problem.This paper presents a reachability model of VSR&DO based on network,and proposes a model to estimate the reachability of VSR&DO,which based on the analysis of the process of VSR&DO model.At the same time,this paper also studies the reachability of production software,provides four basic structures and a computing model of VSR&DO after researching the characteristic and relations among members,and discusses the whole reacha-bility model of VSR&DO. |
PDF Full Text Request