اعتبار تجربی از عملکرد محدودیت تحمل کنترل پروژه
Abstract
1- Introduction
2- Literature review
3- Empirical experiment
4- Results and discussion
5- Conclusion
References
Abstract
The goal of project control is monitoring the project progress during project execution to detect potential problems and taking corrective actions when necessary. Tolerance limits are a tool to assess whether the project progress is acceptable or not, and generate warnings signals that act as triggers for corrective action to the project manager. In this paper, three distinct types of tolerance limits that have been proposed in literature are validated on a large and diverse set of real-life projects mainly situated in the construction sector. Moreover, a novel approach to construct tolerance limits that integrate the project risk information into the monitoring process is introduced. The results of the empirical experiment have shown that integrating project-specific information into the construction of the tolerance limits results in a higher efficiency of the monitoring process. More specifically, while including cost information increases the efficiency only marginally, incorporating the available resource information substantially improves the efficiency of the monitoring process. Furthermore, when projects are not restricted by scarce resources, the efficiency can be enhanced by integrating the available project risk information.
Introduction
An important factor of project success is the timely completion of projects. In this paper, three control methodologies proposed in recent literature to control the schedule progress of projects are empirically compared and validated on the large and diverse real-life project database of Batselier and Vanhoucke [1]. From this database, 93 projects have been selected, of which 71 are situated in the broad construction sector. More specifically, the tolerance limits have been evaluated for commercial, residential and institutional building projects and for civil and industrial construction projects. Moreover, a novel control methodology that integrates the activity risk information into the project control phase is introduced. Project control is, together with baseline scheduling and risk analysis, one of the three major components of Integrated Project Management and Control [2]. While the scheduling and risk analysis phases are performed before the project execution is started, the project control phase is conducted during project execution. The goal of this phase is to identify potential problems or opportunities during project execution, and to take corrective actions to get the project back on track if necessary. During project execution, the actual project progress is monitored and evaluated by comparing it to the baseline schedule. A well-known methodology to monitor the project progress is Earned Value Management (EVM), which originated in the 1960s at the US Department of Defense [3]. While EVM provides simple metrics to measure the current performance of a project, they should be used in conjunction with tolerance limits to assess this performance. These tolerance limits for project control have been established as a tool to support the project manager in deciding whether corrective actions should be taken to get the project back on track. Hence, the goal of these tolerance limits is generating warning signals when the monitored project progress is below a certain threshold, indicating that it is likely that the project will exceed its deadline. These warning signals thus act as a trigger for corrective action for the project manager. The control methodologies validated in this paper are analytical tolerance limits for schedule control using EVM metrics. This type of tolerance limits sets threshold values for the schedule progress at each project phase based on project-specific characteristics. Moreover, each of the tolerance limits evaluated in this paper are constructed for projects with a project buffer. Hence, these tolerance limits generate warning signals when it is expected that the project buffer will be consumed entirely before the project is completed, resulting in a project exceeding its deadline.