Research On R&D Program Management Based On Knowledge Perspective

Nowadays,with the increasing diversity of user needs and technical complexity,the project management of new product development(NPD)has become more and more complex,and traditional project management method,however,do not suffice for today’s organization.When multiple projects in an organization are related to each other to achieve its goals,an program management approach is required for these multiple projects.Programs may consist of multiple projects that are related to each other,and reaching the objectives of a program would not be possible by managing the projects independently.However,the study of research and development(R&D)program management faces some new challenges: How to realize the synergy of R&D multi-project through knowledge commonality.How to scientifically set up R&D programs,so that the enterprise knowledge resources to get effective and reasonable utilization.How to improve the visual management of R&D program knowledge,etc.To address these challenges,this paper is based on knowledge perspective,from the view of interproject dependency and project similarity.It adopts dependency and structure modeling(DSM),clustering,simulation,etc.,focusing on the following four aspects of research and innovation:Firstly,a program clustering method based on knowledge dependency among projects under the concurrent execution condition is proposed.In the field of NPD,the influence of overlapping and concurrent process on inter-project knowledge transfer and program clustering is ignored.This study proposes the inter-project knowledge dependency model under the concurrent execution condition,and improves the two-stage clustering algorithm to identify programs.So that the knowledge dependency of projects within program is maximized and the unnecessary connections of projects among programs is minimized,which can use knowledge effectively and reduce management complexity.Meanwhile,this study analyzes the influence of knowledge dependency of projects inside and outside the programs on program clustering,and verifies that this research method can shorten the total duration of programs and reduce the total coordination cost of programs.Secondly,a program clustering method based on knowledge reusability and similarity is proposed.Research in the field of NPD neglects the current situation of managing multi-project from the perspective of project similarity.This study proposes a project similarity model from the perspective of knowledge reusability and similarity,and uses hierarchical clustering algorithm to identify programs.So that the projects with similar structure and large knowledge gap can be gathered in one program,which is helpful to improve the knowledge sharing level within the program.Meanwhile,this study analyzes the influence of iteration,common activities among projects and intra-project knowledge transfer relationships on program clustering,and finds that iteration not only affects project similarity,but also improves knowledge increment generated by learning.Thirdly,a clustering and knowledge transfer visualization of program method considering attribute is proposed.Since attribute characteristics are ignored in the research on project similarity coefficient based on knowledge relationships,and the existing research on the program visual management is scarce.This study constructs the project similarity model from the two aspects of relationship and attribute,and uses the hierarchical clustering algorithm to identify the programs.On this basis,it analyzes the influence of the project structure and attribute similarity coefficient on program clustering,and proposes the program knowledge transfer visualization method.The results show that intra-project indirect knowledge transfer relationships can improve the degree of similarity among projects,and project attribute similarity has a significant impact on the program clustering results,both of which cannot be ignored when measuring the project similarity.Fourthly,an evaluating the knowledge innovation performance of program method based on hypernetwork model is proposed.Since the existing studies on program management ignore the dynamic knowledge transfer relationship among projects,and the method of evaluating knowledge innovation performance has not yet formed a system.This study constructs the “project-knowledge-product”program dynamic knowledge hypernetwork model,and proposes a method to measure the innovative combination of functional components of products,the knowledge level of programs and the success probability of knowledge innovation.Thus,analyzing the influence of knowledge transfer on program product innovation portfolio and knowledge innovation performance.The Netlogo simulation results show that knowledge transfer within programs can not only improve the product innovation ability and knowledge level of programs,but also make effective use of enterprise’s internal knowledge resources.In short,the contribution of this study is to introduce knowledge management theory into the field of program management.This study proposes reasonable methods to identify R&D programs from three aspects: considering the knowledge dependency among overlapping projects,knowledge reusability and similarity,and project knowledge similarity considering the attribute.It presents a dynamic knowledge hypernetwork model of program to analyze the impact of dynamic knowledge transfer within programs on product innovation portfolio and project knowledge innovation performance.What’s more,this thesis proposes the method to enhance the program knowledge visual management.The conclusion provides an effective decision basis for project managers to use programs to manage multiproject based on knowledge perspective.