Using Last Planner Indicators to Identify Early Signs of Project Performance

Abstract

Reducing uncertainty is a very important issue in construction projects and recently some methods have been developed to deal with it. Particularly, the Last Planner System (LPS) which has been designed to improve planning reliability and project performance; however, its use does not guarantee a total success of projects. Thus, there is still an opportunity to develop new tools that can help to further reduce uncertainty and reliably predict project performance for the decision making process. This paper reports the results of a data analysis from more than thirty six Chilean projects that used LPS, where weekly information was available, in an attempt to identify indicators and patterns that could best anticipate project performance. Indicators included in this research were strategically chosen to analyze patterns regarding reliability of planning in the LPS and performance index from conventional planning methods. The indicators include: PPC (Planned Percent Complete), SPI (Schedule Performance Index), PCR (Percent Constraint Removed), and curves of actual & scheduled progress. Data and indicators were extracted from IMPERA (a software tool used based on the Last Planner methodology). The projects were classified as “Successful” or “Unsuccessful” according to their schedule performance in order to perform statistical analyses. Specifically, this research considers that projects are deemed successful when planning, organization, direction and control are developed in such a way that allows for the compliance of initially established objectives for Costs and Schedule. The results indicate that there is a close relationship between the variability of PPC, SPI and success of the projects. Additionally, different indicators trends were identified for “Successful” and “Unsuccessful” projects in different stages of projects progress. Consequently, comprehension of the relationship between the indicators and project performance could make it possible to predict whether or not a project will be “Successful” at different stages of its life cycle.