摘要
十一五期间,中国汽车欧亿·体育（中国）有限公司迅速发展,从2005到2010的5年间平均增速达到
25%。当代汽车欧亿·体育（中国）有限公司的产品开发,已经全部依靠项目管理的方式在进行实现,风险评
估作为当代项目管理中的一个重要部分,在国外已经经历了长期的发展,可是对于
多数国内企业这个领域的认知还不是很深。这些对于风险的识别及评估方法的不足,
将严重影响本土零部件供应商水平和能力的提升。本文将对S公司新产品开发项目
的风险评估进行探索和研究,主要内容如下:
基于S公司的经验,请相关专家组进行头脑风暴,得出所有的相关二级指标。通
过主观评分的,式进行风险因素的评分,并通过应用层次分析法,计算出各风险因
素的权重指标,由于因素指标较多,为了简化分析过程,依据8020原则,仅取其中
最有影响的80%的因素作为主要调查分析指标,构建出风险指标体系并评分。
为了得到S公司新项目各风险因素与最终项目总风险的关系,本文结合神经网络
的相关理论构建了一个风险评估模型,并将之前的调查结果输入,通过广义回归神
经网络,将S公司的项目通过此模型进行计算,并且检验以及评估。以该项目案例
为基础配合以上得到的风险因素及评估模型,分析在实际工作中的应用方法,使之
成为能处理一些实际问题的有效工具。
关键词:项目管理;汽车零部件;风险评估
Abstract
During the 11th five-year plan of China, Automotive industry made quite noticeable
progress. From 2005 to 2010,average increase rate around 25% in these five years.
Automotive parts development project is a systemic project. Automotive industry project
are managed by PM in these days. Risk assessment, one of the most important part of PM,
already developed for a long time, but most of local suppliers are still not aware of this.
So lack of risk identification and assessment methods,which will be negative on local
supplier product development. This research focus on risk assessment.
Firstly with brain storm, list risk factors as possible. According to Delphi analysis,
evaluate risk factors by the committee of experts from S company. By AHP, calculated
coefficient of risk factors, according to 8020 principle, select most important 80% factors
as main factors, and finished building risk factors system.
For clarify the influence of factors, a risk assessment model was built up based on
Neural Network with self learning function. According to input data, training the network
and apply the trained network to test and forecast.
An application case will be explained according to the Model of risk assessment,
describe detail in applied method to show how it is an effective tool.
Keywords: Project Management; Automotive Products; Risk Assessment