摘要：This study analyzes construction productivity based on the construction duration per floor and per gross area over 20 years (1996–2015) and compares the results among the United States, United Kingdom, South Korea, and Japan, which have similar sizes of total construction investment and market risk. Although construction labor productivity is widely used to analyze and compare construction productivity among countries, it does not consider the changed construction duration caused by levels of investment and technology. Therefore, construction duration per floor and gross area was selected analyze and compare construction productivity in this paper. Regular and non-modular buildings with a total of five or more floors and a basement are collected during the analysis period (1996–2015). The total number of collected buildings is 800 and it includes buildings in the United States (194), the United Kingdom (186), South Korea (322) and Japan (98). Construction duration, increase rate and standard deviation are then compared between each country. Finally, factors that influence construction duration are derived and additionally considered to explain and adjust the trends and changes of construction productivity related to construction duration in the four countries. The productivity of the United States is the highest, but the difference between it and other countries decreases steadily because the increase rate of the construction duration in the United stated is larger than those of other countries. Then, the factors influencing the construction duration are derived as a learning effect by the number of ground floors and gross area, as well as the rate of constructed buildings with a first basement floor for efficient productivity management. The rate of the first basement floor influences both the construction duration per floor and per gross area. This study contributes to the field by explaining the productivity change based on the construction duration and proposing the key management point of the productivity by deriving the influence factors.
