摘要
复杂产品开发企业往往拥有多个产品线开发项目和公用技术平台开发项目.由于资源的稀缺性,企业决策者不得不面临着R&D项目组合选择的难题.针对复杂产品开发过程中存在的多项目间混合依赖特性,采用关联性分配矩阵定量化描述三个及以上项目共同作用时的相互依赖程度;综合考虑项目间的收益依赖、资源依赖和技术依赖等关联特征,构建一种新颖的面向研发项目组合选择问题的非线性规划模型;并设计和实现了基于模拟退火的相关求解算法.最后将本文提出的方法应用于国内某汽车公司的研发项目组合管理实践中,实验结果证明了该方法的合理性与有效性.
Complex product development companies often have multiple product lines development projects and public technology platform development projects. Due to resource scarcity, decision makers have to be faced with the R&D project portfolio selection problems. In this paper, a conception of dependence assignment matrix is formulated to quantify the interaction in a portfolio based on multi-interdependences among projects in the process of complex product development. An improved non-linear planning model for R&D project portfolio selection is developed in consideration of combined benefit, resource and technology interaction. A solving method to find a satisfying portfolio is designed and implemented based on simulated annealing algorithm. Finally, the proposed method is applied to R&D project portfolio management practice in a domestic auto company. Experimental results show that the method is reasonable and effective,
出处
《系统工程学报》
CSCD
北大核心
2011年第6期834-841,共8页
Journal of Systems Engineering
基金
国家自然科学基金资助项目(70971035)
安徽省自然科学基金资助项目(11040606Q28)
合肥工业大学基金资助项目(2010HGXJ0010)
关键词
项目组合选择
R&D项目
多依赖性
模拟退火算法
project portfolio selection
R&D project
multi-interdeperidences
simulation annealing algorithm