Improvement to the forecast precision of the factor model for masonry productivity

Several techniques exist for forecasting productivity of labor-intensive construction activities. The Factor Model is one approach. This method theorizes that productivity is influenced by factors in the work and in the environment. The factors in the work can be identified and quantified. A regression model is created from an analysis of these factors and the model equation is used to predict productivity. The Factor Model for masonry has been demonstrated as a reliable and accurate forecasting method on average; however, precision can be off for any single project. Improvement to the forecast precision of the Factor Model is the focus of the research. More than 900 days of daily productivity observations from 25 projects were analyzed. Analysis of variance, correlation and multivariate regression were used in the study. The study defines a new factor and establishes a revised model structure for predicting masonry productivity. A complexity variable is quantified for exterior and foundation work and incorporated into the revised model structure. The new model is called the Complexity Model. Forecasts of productivity using the Complexity Model and the Factor Model are compared for eight projects at the 5% to 25% completion stage. At 5% project completion, the Complexity Model forecasts are 6% more precise than the Factor Model forecasts. Accuracy decreases by approximately the same percentage. For one group, consisting of four projects with small first-day variation, the Complexity Model showed a 40% improvement in both forecast accuracy and precision. The findings provide an understanding in explaining, predicting and forecasting labor-intensive resources for construction operations. Further research is indicated.