区间数网络计划求解时间的影响因素分析

Research of influence factors on computing time in interval project scheduling problems

通过对区间数网络计划问题的算法进行总结,提出了一种结合了多项式算法和枚举算法的新算法,用该算法对一系列项目案例数据进行了求解分析,并通过统计分析得出结论:项目活动时长变动幅度和网络顺序强度对区间数网络计划的求解时间都有显著影响,活动时长变动幅度越大,网络顺序强度越大,求解时间越长。在实际项目计划过程中,应该通过控制项目活动时长变动幅度和网络串行程度来减少项目网络计划问题的复杂度,便于项目进度管理工作的开展。

In this paper, the model of interval project scheduling problem is described, and a new algorithm to assert activity criticality is introduced, in which criticality of activities is asserted by the combination of polynomial algorithms and path enumeration. With the adoption of this algorithm, a series of hypotheses concerning the relationships among network order strength, activity duration change and other parameters, as well as their integrated impact on computation time are proposed and tested by a series of experiments. The conclusion is that activity duration change and order strength of net- work have a positive impact on the computation time. In practice project planning, planners must limit the order strength of project network and activity duration change as strictly as possible to finish project planning smoothly.