Labor productivity plays an important role
in the successful delivery of construction projects. It is
substantially important to understand the nature and extent
to which certain factors affect productivity. A field study
conducted to examine a set of parameters that impact labor
productivity on building construction jobsites is presented.
These parameters include temperature, relative humidity,
wind speed, precipitation, gang size, crew composition,
height of work, type of work and the method employed.
Techniques from inferential statistics and artificial
intelligence such as Fuzzy subtractive clustering, neural
network modeling and stepwise variable selection procedure
are used to analyze and determine the relative importance
and contribution of each parameter towards productivity
estimates. Required data was collected over a period of ten
months directly from construction jobsites. For further
insight on the impact of these factors on productivity, a
set of neural network models were developed. Productivity is
expressed as a function of one parameter at a time. The
trends obtained in this study are compared with related
material in literature and the findings are
reported.