Content area
Full Text
Nomemclature
AI=Artificial intelligence
ANN=Artificial neural network
BLDG.=Building
BPNN=Back propagation neural network
DP=Daily productivity
F =Floor level
G =Gang size
kms/hr=Kilometers per hour
L =Percent labour
M =Work method
m2 /man-hr=Meter square per man-hour
MAE=Mean absolute error
MSE=Mean square error
n =No of data points or input patterns
NN=Neural networks
P =Precipitation
Ppt.=Precipitation
R2 =Coefficient of multiple determination
T =Temperature
TW=Work type
W =Wind speed
Introduction
The importance of labour productivity in construction arises from its impact on completing projects within their targeted time and cost. For instance, contractors have often focused on labour productivity rates as the primary source of the overall success or failure of a project ([13] Missbauer and Hauber, 2006). Contractors at the bidding stage of a project are interested in knowing site labour productivity figures, to estimate the labour cost component of the project. Thereafter, if the contract is awarded to the contractor, the company needs to ensure that the estimated level of productivity is achieved or bettered. It was widely believed, in the construction industry, that productivity exhibited decline over the past few decades ([3] Brisco, 1988, [4] Christian and Hachey, 1995). Later, however, some microeconomic studies ([2] Allmon et al. , 2000), reported the contrary. Irrespective of productivity trends, a relatively recent research cautioned against unreliability in labour productivity raw data and the uncertainties inherent in the methods used for its estimation and interpretation ([8] Eddy and Peerapong, 2003).
Though extensive research has been done on parameters influencing productivity, most focused on determinants that have mid- or long-term impacts on labour productivity ([1] AbouRizk et al. , 2001; [6] Crawford and Vogl, 2006; [12] Lam et al. , 2001; [16] Moselhi et al. , 1997; [20] Thomas, 1992; [22] Thomas and Zavrski, 1999). Considerably less number of studies have targeted individual; not multiple factors, that cause daily or short-term variations in productivity ([9] Hancher and Abd-Elkhalek, 1998; [11] Koehn and Brown, 1985; [21] Thomas and Sakarcan, 1994; [19] Sanders and Thomas, 1993; [14] Mohamed, 2005). The aim of this paper is to study labour productivity in building construction with a focus on formwork operations and to quantify its variations due to daily changes of factors considered to have direct impact on labour productivity. This paper presents...