CONTINUOUS ASSESSMENT OF PROJECT PERFORMANCE

Project managers for owners, designers, and contractors need real-time information to assist them in managing projects. This paper describes a process whereby owner, engineer, and construction contractor organi zations can use continuous or time-dependent variables (e.g., owner ex penditures, construction effort hours expended) to predict project out comes from start of detailed design through construction completion. C ontinuous variable data were collected on 54 construction projects. S- curves were developed for two project outcome categories: (1) ''succes sful'' (meeting or exceeding budget and schedule expectations); and (2 ) ''less-than-successful'' (not meeting budget and/or schedule expecta tions of the owner). Statistical analysis was performed to identify th ose variables showing a statistically significant difference between t he two project outcome categories. Variables exhibiting a significant difference between the S-curves for ''successful'' and ''less-than-suc cessful'' projects can be used as predictors of project outcome. Resul ts show that different variables were predictors of success at differe nt points of time during the project life cycle. Practical application s of these results along with limitations and future research are desc ribed.