Error Propagation Model for Analyzing Project Labor Cost Budget Risks in Industrial Construction

Abstract

Industrial construction employs various trades in large-scale prefabrication operations to produce modules and structural components at an offsite facility that will be shipped to the field for rapid installation. Developing an analytical methodology for characterizing the effect of variability in productivity on labor cost budgeting is vital to this particular construction type. Integrating current practices of estimating, scheduling, and budgeting in industrial construction, this paper describes an error propagation model for calculating the standard deviation of the cumulative labor hours at particular time points of the project duration and establishing a confidence interval around the average value. Analogous to plotting an S-curve, the lower bound and upper bound of the interval for cumulative labor hours budgeted at control points along the project duration can be articulated to form the S-stripe, which visually portrays the risk of labor cost budget due to risks inherent in labor productivity. The application and verification of the proposed analytical methodology are illustrated with a steel fabrication project case. Monte Carlo simulation is applied to the same project data in the case study, resulting in a near-perfect correlation between the two sets of results. In the simulation experiment design, determining the minimum number of simulation runs that are deemed sufficient to obtain reliable sampling results entails trial and error, and the obtained result is case-specific. In contrast, the proposed analytical method circumvents this barrier by analytically deriving the project labor cost budget in the form of an S-stripe.