Evaluation of Lean Improvements in Residential Construction Using Computer Simulation

Changes to apartment designs initiated by clients are an inherent part of residential construction. Competition for customers where supply exceeds demand, and increasingly sophisticated consumers, have forced construction companies to accept the phenomenon. However, client changes delivered during construction adversely affect the stability and flow of work at the construction site. To cope with this challenge, three main changes were suggested to the traditional approach of construction of high-rise residential buildings: reduce the batch size to single apartments, work with multi-skilled teams and schedule work with a pull controlled system. To evaluate these three changes, Sacks and Goldin developed a management simulation game, called ‘LEAPCON’, which simulates the construction process of an eight story building with 32 apartments using Lego pieces. Repeated executions of the live game indicated reduced cycle times, cash flow and levels of work in progress, and improved throughput. However, investigation of the separate and combined influences of each intervention could not be measured, nor could the performance indicators be monitored through time. Thus computerized discrete event simulation was needed for more complete evaluation of the three lean management changes. Twelve computerized simulations were built using the Stroboscope program. It was found that each one of the interventions improved one or more of the parameters of interest, while different combinations of them also improved some of the results. Reduced batch size improved cash flow and work in progress; multi-skilled teams improved labour utilization rates, and pull flow enabled execution of all client changes. However, application of all three interventions in unison was essential for achieving the anticipated results simultaneously.