Duration determination for rural roads using the principal component analysis and artificial neural network

Purpose Determining the duration for road construction projects represents a problem for construction professionals in Ghana. The purpose of this paper is to develop an artificial neural network (ANN) model for determining the duration for rural bituminous surfaced road projects. Design/methodology/approach Data for 22 completed bituminous surfaced road projects from the Department of Feeder Roads (rural road agency) were collected and analyzed using the principal component analysis (PCA) and ANN techniques. The data collected were final payment certificates which contained payment bill of quantities (BOQ) of work items executed for the selected completed road projects. The executed quantities in the BOQ were the total quantities of work items for site clearance, earthworks, in-situ concrete, reinforcement, formwork, gravel sub-base/base, bitumen, road line markings and furniture, length of road and actual durations for each of the completed projects. The PCA was first employed to reduce the data in order to identify a smaller number of variables (or significant quantities) that constitute 81.58 percent of the total variance of the collected data. The ANN was then used to develop the network using the identified significant quantities as input variables and the actual durations as output variables. Findings The coefficient of correlation (R) and determination (R2) as well as the mean absolute percentage error (MAPE) obtained show that construction professionals can use the developed ANN model for determining duration. The study shows that the best neural network is the multi-layer perceptron with a structure 3-38-1 based on a back propagation feed forward algorithm. The developed network produces good results with an MAPE of 17.56 percent or an average accuracy of 82.44 percent. Research limitations/implications Apart from the fact that the sample size was small, the developed model does not incorporate the implications of other likely factors that may affect contract duration. Practical implications The outcome of this study is to help construction professionals to fix realistic contract duration for road construction projects before signing a contract. Such realistic contract duration would help reduce time overruns as well as the payment of liquidated and ascertained damages by contractors for late completion. Originality/value This paper proposes an alternative way of determining the duration for road construction projects using the total quantities of work items in a final payment BOQ. The approach is based on the PCA and ANN model of quantities of work items of completed road projects.

Download Full-text

Predicting the Amount of Municipal Solid Waste via Hybrid Principal Component Analysis-Artificial Neural Network Approach

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.768.722 ◽

2015 ◽

Vol 768 ◽

pp. 722-727 ◽

Cited By ~ 1

Author(s):

Salman Safavi ◽

S. Mohyeddin Bateni ◽

Tong Ren Xu

Keyword(s):

Neural Network ◽

Principal Component Analysis ◽

Artificial Neural Network ◽

Municipal Solid Waste ◽

Solid Waste ◽

Principal Component ◽

Component Analysis ◽

Ann Model ◽

Input Variables ◽

Overfitting Problem

Accurate prediction of the amount of municipal solid waste (MSW) is crucial for designing and programming MSW management systems. Reliable estimation of MSW is difficult since many variables such as socio-economic characteristics, climatic factors and standard of living affect it. A number of studies used artificial neural network (ANN) to predict MSW. However, due to the large number of input variables to the ANN, it could not not perform well and generally encountered overfitting. This study takes advantage of the principal component analysis (PCA) technique to reduce the number of input variables to the ANN model in order to overcome the overfitting problem. The proposed PCA-ANN approach is used to predict the weight of MSW in the province of Mashhad, Iran. The utilized experimental data in this study are obtained from the Recycling Organization of Mashhad Municipality archive (http://www.wmo.mashhad.ir). It is found that the PCA approach can successfully decrease the number of input variables from thirteen to eight. The PCA-ANN model (with eight input variables) outperforms ANN (with thirteen input variables) and provides more accurate estimates of MSW as it mitigates the overfitting problem associated with ANN. The root-mean-square-error (RMSE) of MSW estimates reduces from 499000 Kg to 448000 Kg by using the PCA-ANN model instead of ANN.

Download Full-text