Vol.5 No.1 2012

Research paper : Construction of a traceability matrix for high quality project management (A. Sakaedani et al.)−15−Synthesiology - English edition Vol.5 No.1 (2012) verification and efficacy check of software onboard spacecraft. Currently, Associate Professor, Keio University and Director, VSE Center, Keio University. Engages in research in systems engineering, process assessment, and design and management of space systems and ubiquitous systems. Member of INCOSE, IEEE, Information Processing Society of Japan, and others. Promotes international collaboration and interactive courses at the Graduate School of SDM. Doctor (Policy and Media). For this study, was in charge of research integration.Discussion with Reviewers1 The paper as a practice of synthesiologyComment (Motoyuki Akamatsu, Human Technology Research Institute, AIST)I understood this paper as a proposal for a traceable modeling method of how the design information for software is communicated or how the information travels in an organization. It shows how quantitative evaluation can be done for the level of complexity of a project, using this method.To create software quickly and accurately, it is important to design a project eliminating complexity, and in that sense, this is a paper related to synthesiology. However, it focuses on the explanation of the method for traceability matrix to evaluate the information communication efficiency. Although it is shown that the matrix can be used as a comparative tool of complexity, there is no example given where it is used directly, for pointing out the relevance to synthesis. Please add this point.Comment (Hideyuki Nakashima, Future University Hakodate)Please indicate specifically how the method proposed here is useful in synthesis.Answer (Akihiro Sakaedani)The point is explained in “5.1.3 Reduction of the interdependency and difficulty of individual matrix and the evaluation of complexity.” I explained how the individual matrix of the traceability matrix should be based on the properties of matrix calculation, and used specific examples to describe how to create the best synthesis using interdependency and difficulty.2 Utilization of this method in managementComment (Motoyuki Akamatsu)I can intuitively understand that the management becomes difficult as the complexity of the organization increases. However that is a hypothesis only, and I think you should describe how much management would become easier if this method is used. Since the specific management method is an important point in synthesiology, please indicate specifically how management will become easier by using this method, and how management should be done based on the indices obtained in this method.Answer (Akihiro Sakaedani)I described the PDCA cycle of management using the traceability matrix in “5. PDCA cycle of project management using the traceability matrix.” An explanation is added from the perspective of evaluating the project status using complexity as the index, and from the perspective of evaluating the status of individual elements by focusing on the difficulty and interdependency.3 PDCA cycleQuestion (Motoyuki Akamatsu) You write about the method of turning the PDCA cycle using the traceability matrix, but I imagine that calculating the complexity by rewriting the relationships and difficulties on the system matrix while turning the cycle is a rather troublesome task. I don’t think it is easy to see where the element of a certain region is used in another region, or how to determine the magnitude of the difficulty value. I think you need some maneuvering to turn the PDCA in a realistic manner. Can you please present your thoughts on this point?Answer (Akihiro Sakaedani)I added the point you indicated. An explanation is given of reviewing the unit of management and selecting the appropriate granularity of the elements, and of setting the PDCA cycle that matches the granularity. Below, I give a detailed view of the setting of component values of the system matrix below.On setting the difficultyThere are two types of difficulties in the difficulty setting. One is to understand the difficulty of an element qualitatively, and the other is to quantify that qualitative understanding.First, I shall explain the qualitative understanding of difficulty. Understanding the difficulty qualitatively also involves entering information such as the progress and risk of the project. Since progress management and risk management are done by using conventional project management technology, they can be organized, without a problem, as input information in setting the difficulty level. In fact, the project manager, architect, or team leaders have an intuitive understanding of the changing difficulty of each element. For example, in many projects, in the everyday conversion that takes place during the project, there are many discussions pertaining to the difficulty of elements that constitute the project, such as who is the key person, which tool has problems, or which activity is critical. Therefore, there should not be a particular barrier in understanding the difficulty qualitatively.However, there are issues in quantifying the risk items into difficulty levels. I have yet to verify whether independent evaluations can arrive at equivalent results based on the guidelines indicated in the paper, and this is an area targeted for future research.On setting the interdependencyIn a project without some sort of activity standard, it is reasonable to expect difficulty in organizing the interdependency. In a project without an activity standard, current process-centric management is difficult in the first place. In that sense, using the proposed model may appear cumbersome due to lack of experience, but I believe it can function effectively as a tool for understanding the status of the project. In a project that already has an activity standard, in general, it defines activities and artifacts, as well as the roles of each team, and it is possible to organize the interdependency in that phase. Once the organization is done, each project can be customized by using the template, and there should be no problem in reusing such a template.OverallAs you indicated, this proposal may, at first glance, be a very difficult management technology. In conventional project management, the main method is the management of progress status centering on process and the management of budget based on process. Therefore, the problems that arise from those perspectives were analyzed and measures were taken against the cause of the problems. However, the concept of this method is to understand the problems arising in the project through the difficulty and interdependency of individual elements based on the model, and then taking measures after gaining an overview of the whole project. This means that conventional project


page 18