A Study on the Project Management Model with Probability of Completion：Based on Earned Value Management and Critical Chain

• Main Authors: Wu, Guo-Heng, 吳國亨
• Other Authors: Shao, Min-Hua
• Format: Others
• Language: zh-TW
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/mp3q9w

碩士 === 國立屏東科技大學 === 資訊管理系所 === 106 === For enterprises, project management has always been a very important issue because the pre-operations for any project, project planning, policy, allocation of human resources and marketing strategy all need to be done through project management to set the subprojects to be implemented in the project, the time required as well as the sequence of the process. The program evaluation and review technique (PERT) and other scientific management methods can be used to calculate the earliest and latest time of completion as well as the earliest and latest starting time, so as to find out the critical path to implement the project preferentially without any delay and further calculate the probability to achieve the project; then the decision is made according to the results to complete the project most efficiently with less time to achieve the final purpose. Based on the multiple researches into traditional project management methods, only 44% projects can be completed according to the time scheduling, quality and cost specified. Lee et al. (2010) proposed an idea that the cost and value are used as the measurement indicators to adjust buffer time, but it is too simple and not rigorous enough because they only consider the cost and value. Therefore, this research added the probability of completion to add the buffer time for the activities with lower probability of completion. Thus, the activities with higher probability of completion will reduce the buffer time. Meanwhile, the net value was added, so that the buffer time of the activities with the same value will not be determined by the higher cost. This research applied the earned value management, the probability of completion of PERT as well as the combination of net value and critical chain to control and manage the projects in a more accurate way, so as to assist the project managers to improve the possibility that the software projects can be completed within the scope of specified time scheduling, quality and cost. This research firstly analyzed each activity of the project and transformed them into flowchart of critical chain and then the longest path in the dependent operations was used to find out the critical chain; next, the root-mean-square error method was adopted to calculate the buffer time of each activity and then probability of completion of PERT was used to adjust the measurement indicators of the subsequent earned values and the net value of the new measurement indicators as the weight of the buffer time allocated to each activity, so that the buffer time of project scheduling in the entire critical chain will be allocated more reasonably. According to the results of this research, it can be known that when an activity has the lower probability of completion, it will obtain much more buffer time than that before the adjustment. Conversely, when an activity has the higher probability of completion, it will obtain less buffer time than that before the adjustment.