Management-Reserve Estimation for International Construction Projects Based on Risk-Informed k-NN

Abstract

International construction projects normally involve high levels of risk. To reduce such risk, contractors typically include a reserve to mitigate cost overruns and schedule delays. This reserve comprises two categories: (1) a contingency reserve covering potentially required changes, and (2) a management reserve (MR) covering unplanned changes. Since estimating an MR is difficult because the inherent risks are unpredictable, the traditional percentage method is still used for this purpose. To address this shortcoming, this study proposes an MR estimation method based on the cost and schedule performance ratios of international construction projects. K-nearest neighbor (k-NN) is applied to retrieve similar projects and a genetic algorithm is adopted to optimize the retrieved cases. Finally, case studies are conducted to validate the estimated ratios and test the applicability of the proposed MR estimation method. The case study results show that the proposed method can infer the cost and schedule performance ratios within a 5% error rate. As a result, it can estimate the MR more accurately than the traditional percentage method.