Productivity studies using advanced ANN models

Estimating labor productivity is one of the most difficult aspects of preparing an estimate, or a control budget based on the estimate for labor-intensive activities in construction. The primary objective of research is developing artificial neural network or ANN based estimating tools to offer estimators valuable information about labor productivity in bidding new jobs.

In conjunction with a major Canadian industrial contractor, the thesis research presents case studies on the theoretical basis and practical considerations for measuring and analyzing labor productivity in industrial construction. Two important activities of process piping were investigated: pipe installation in the field and spool fabrication in the fabrication shop. Emerging computer modeling techniques such as data warehouses and ANN were researched from an academic perspective and implemented in industry to meet the challenges in productivity studies. The thesis research has addressed: (1) how to quantify labor productivity in industrial construction from a contractor's point of view; (2) how to measure actual labor productivity in industrial construction based upon on-site control practices; and (3) how to utilize ANN to analyze the variability of actual labor production rates and the sensitivity of identified influencing factors.

Using actual data, the proposed ANN models were proven to be effective in both risk analysis and sensitivity analysis of construction labor productivity. The developed data warehouses and ANN-based decision-support tools have been implemented or are in the process of implementation at the involved company. The final results of the research not only assist estimators to improve the accuracy of estimating labor production rates for studied activities in bidding new jobs, but also offer the management a precise and integrated view of corporate productivity information spanning across many business divisions. The experience and lessons learned from the successful, productive and mutually beneficial collaboration between academia and industry in the thesis research will potentially benefit other university-industry joint research projects in the future.