CAJ | 학술논문

能力约束缓冲的设置问题是关键链多项目管理法的核心问题,对成功执行关键链多项目调度非常重要。本文考虑了使用＂鼓资源活动之间存在间隙的特点,分析了集中设置能力约束缓冲方法的不足之处,提出了分散式能力约束缓冲设置法。该法阐述了能力约束缓冲在＂鼓＂计划中的配置位置和大小。然后通过数据实验将该法与集中式能力约束缓冲设置法进行了比较,结果表明：集中式能力约束缓冲会产生过大或者过小的缓冲,而分散式能力约束缓冲设置法能针对不同的＂鼓＂计划设置合适的缓冲大小。特别当＂鼓＂活动间隙靠近最后一个＂鼓＂活动时,分散式设置法能提供更短的项目群计划完工期,同时得到与集中式设置法一样的按时完工率,以及较低的缓冲超出概率。
능력약속완충적설치문제시관건련다항목관리법적핵심문제,대성공집행관건련다항목조도비상중요。본문고필료사용＂고자원활동지간존재간극적특점,분석료집중설치능력약속완충방법적불족지처,제출료분산식능력약속완충설치법。해법천술료능력약속완충재＂고＂계화중적배치위치화대소。연후통과수거실험장해법여집중식능력약속완충설치법진행료비교,결과표명：집중식능력약속완충회산생과대혹자과소적완충,이분산식능력약속완충설치법능침대불동적＂고＂계화설치합괄적완충대소。특별당＂고＂활동간극고근최후일개＂고＂활동시,분산식설치법능제공경단적항목군계화완공기,동시득도여집중식설치법일양적안시완공솔,이급교저적완충초출개솔。
The Critical Chain Project Management （ CCPM）, based on Goldratt＇ s Theory of Constraints （TOC）, has been a popular and effective approach to understanding project management. The mechanism of this theory is based on the risk-sharing concept. CCPM tries to eliminate the impact of student syndrome and Parkinson＇ s Law, protect team members from interruption and multi-tasking, shift task measurements from on time task completion to on-time project completion, and reduce the number of works in process. The procedure of critical chainmulti-project scheduling is provided to solve the negative effect of multi-tasking and uncertain factors in a multi-project environment. Capacity constraint buffer （CCB） is a time interval between two projects in a sequence to prevent any negative variability in accomplishing the drum tasks in one project from impacting the start of drum tasks in another project. How to set CCB and how much CCB to set are two remaining issues. The current literature suggests that CCB should be placed before the first use of the drum resource in the latter project and its size is equal to the sum of the safety of all drum activities, which neglects the effect of drum activities space and its location in drum schedule on the buffer size. The centralized capacity constraint buffer allocation method provided by Bie and Cui （2011） considers the effect of the sizes of drum activities space without considering the effect of their locations in drum schedule. Several project examples are provided to show that even though the sum of drum activities space in drum schedule is fixed they may produce different protection effect because of the difference of their location in drum schedule. This centralized buffer allocation method can＇ t generate the adaptive buffer size. Based on the characteristics of drum activities space and the defect of the centralized capacity constraint buffer allocation methods used in some projects, we propose a method of setting distributed capacity constraint buffer. This approach places a CCB behind each drum activity that has a space. The size of CCB is set to equal to the minor value from the drum activity safety and the time interval between this drum activity and the next drum activity. In particular, the size of the CCB behind the last drum activity is set to equal to its safe time. A simulation experiment is carried out to demonstrate the application of the proposed method. We compare the distributed capacity constraint buffer allocation method with the centralized capacity constraint buffer allocation method. The result shows that the centralized buffer allocation method will overestimate the buffer size in some situations while it will underestimate the buffer size in other situations. The distributed buffer allocation method is able to provide the proper buffer size so that the characteristics of drum activities space can be taken into account. The distributed buffer allocation method performs better than the centralized buffer allocation method in total projects time and on-time delivery when drum activities spaces are near the last drum activity in the former two projects. Future research is needed to achieve the shortest total project duration by adjusting the proportion of safety to drum activities and keeping the same on-time project completion rate. Additional research is also needed to focus on the validation of the proposed method in more extensive problem sets.