Identifying high performance construction projects

Learning from high performance projects is crucial for construction improvement. Therefore, we need to identify outstanding projects or role models. A minimum prerequisite for identifying such projects is the ability to measure the performance. Unfortunately, two issues complicate the measuring task: i) diseconomies or economies of scale and ii) multidimensional inputs and outputs. We propose to use Data Envelopment Analysis (DEA) to measure the productivity of building projects. DEA fulfils the two requirements stated above, and to our knowledge, it is the only method complying with these two vital requirements. The presentation emphasises the strengths as well as the limitations of DEA, comparing it with regression analysis. The results from this empirical study of 58 projects extracted from a database in a large Norwegian construction company suggest that there is a 50% potential for productivity improvement by learning from and copying the role models that were identified. Also, we discuss economies and diseconomies of scale in construction projects. We recommend DEA as an appropriate method for identifying role models and for benchmarking projects. Used together with methods for hypothesis testing, DEA is a useful technique for assessing the effect of alleged process improvements.