Predicting Penetration of the Project Buffer Time of Critical Chain Project Management (CCPM) Using a Linear Programming Approach

Abstract

Listen

Translate article

Managing the disruptions in projects is a challenging task for project managers. In this respect, critical chain project management (CCPM) has been considered as a promising methodology in expediting projects. However, the effectiveness of this methodology is often lost in search of optimal use of project buffers to hedge against delays occurring in the critical chain. The more critical chain activity gets delayed, the greater is the likelihood of penetration of project buffer time. The resource constraints of critical chains have been considered as a major determinant of project buffer penetration. Although a project buffer is provided to protect the critical chain, to keep the CCPM project schedule competitive, it is the priority of the project manager to minimize the blatant consumption of project buffer time by multiple critical chain activities. Historically, resource constraints within the critical chain have been associated with penetration of project buffer time. The literature suggests that the productivity of resources deployed in the critical chain can predict the penetration of project buffer time. Based on the premise that delays are omnipresent and unavoidable but predictable, this research aims to consider the post-facto measures of delays instead of pre-facto measures. Pre-facto measures are resources productivity and resources availability while post-facto measures are the compensation paid by the client and the cost of liquidity damages incurred by the contractor. It is assumed that pre-facto measures are convertible into post-facto measures. A linear programming model is formulated and tested using the case of Botanical Garden Construction Project in Pune city. Sensitivity analysis revealed that the propensity of project buffer penetration on a critical chain varies with the compensation matrix. Varieties of scenario are developed and the optimal solution is validated with the case study.