Uncovering the underlying pathologies in capacity planning : matching load to capacity -

Abstract

The need for proper and reliable planning is essential for project success. Capacity planning, which is the allocation of activities to available resources, has received good attention in the construction community but few metrics exist to assess its performance. Since it is impossible to improve what cannot be measured, the goal of this thesis is to firstly introduce new capacity planning metrics that will help visualize and understand the current state of capacity planning on construction projects. Is there an overloading or under loading of resources? Secondly, to use these metrics on real data to check whether the metrics paint a proper picture of planning patterns and reliability of planning. The new metrics developed in this research, will attempt to help in assessing the state of equilibrium in choosing the weekly load of tasks to match the existing capacity, or at least, to minimize the gap between the two as much as possible. These new metrics, in theory, will achieve the goal of informing planners and last planners about the status of load vs. capacity, the matching between the two, and the reliability of capacity planning on a project. Furthermore, the metrics were applied to real data from two on-going projects in the US. The two projects were analyzed individually, and then compared. The metrics showed the performance level each project and proved that (1) there is a mismatch problem between load and capacity, (2) teams on projects are not carrying out proper capacity planning techniques, (3) one cannot look at performance metrics such as the Percent Planned Complete (PPC) alone to assess the performance of projects and the allocation of resources, (4) one cannot look at a single metric to analyze performance and reliability since no metric is a standalone metric, (5) a time-series analysis showed that most teams do not learn from previous mistakes, and make decisions independent of previous ones, (6) some teams focus on one aspect of performance such as allocation of resources or matching load to c