Vol.5 No.1 2012
13/82

Research paper : Construction of a traceability matrix for high quality project management (A. Sakaedani et al.)−10−Synthesiology - English edition Vol.5 No.1 (2012) where the unit matrix is subtracted from system matrix s, all diagonal components will be -1. Therefore, the norm will be √N, and the complexity will converge to the following value (Equation 2).Fig. 10 Example of sensitivity analysis for project complexityFig. 11 Management unit and frequency complexity as the managing index. If the complexity is decreasing, it shows that the development is progressing smoothly, whereas if the complexity is increasing, it shows that some problem may be occurring in the project (Fig. 9).The other perspective is managing the individual elements of the project (seeing the trees), and progress is monitored by using interdependency as the managing index. For example, the element allotted to each team is determined, and the interdependency of the elements is managed. One proposal is to use the number of interdependencies and the rate of change as indices of progress.5.4 ACTFrom the perspective “seeing the trees” of the project, if there is a positive rate of change in complexity, an overview of the problems is gained from the model and the issues are uncovered. If the problem can be solved directly and the difficulty and interdependency can be reduced, measures are implemented. If the measures cannot be taken directly, simulation is done for other elements and a sensitivity analysis is done (Fig. 10). In this example, it can be seen that element a is more effective for reducing the complexity than element b.Fig. 9 Successive change of complexity and project statusOn change trend in progress and complexity of the projectTrend of successful projectTrend of failed projectDeadlineDevelopment periodComplexityN × SnElement bElement aSensitivity analysis of project elementsChange difficultyInitial value of elements a and bDifficultyComplexityManagement unitPDCAcycleShort termLong termSmall granularityLarge granularitySubgroup ④Subgroup ③Subgroup ②Subgroup ①Group BGroup AWhole projectComplexity = N ×Sn ・・・(Equation2)A system matrix that is perfectly diagonalized is considered a singularity, and is not subject to Equation 2. A perfectly diagonalized system matrix is an ideal project according to Fig. 2, and is not a subject of discussion in regard to complexity.Therefore, two perspectives can be considered as ways to manage the progress status of the whole project.One perspective is gaining an overview of the whole project (seeing the forest), and progress is monitored by using the

元のページ 

page 13

※このページを正しく表示するにはFlashPlayer10.2以上が必要です