Hybrid intelligence approach based on LS-SVM and Differential Evolution for construction cost index estimation: A Taiwan case study

Abstract In construction industry, cost escalation is a major cause of project failure. The reason is that construction prices fluctuate over time and they are influenced by numerous factors. This study establishes a hybrid intelligence system, named as ELSVM, for modeling construction price variations quantified by the construction cost index (CCI). The proposed method is developed based on the fusion of Least Squares Support Vector Machine (LS-SVM) and Differential Evolution (DE). LS-SVM is used for mining the mapping function between CCI and its causative input factors. Meanwhile, DE is employed to optimize LS-SVM tuning parameters. In this research, a database consisting of 122 historical cases was collected to establish the intelligence system. Based on experimental results, ELSVM can successfully model CCI fluctuations since it has achieved a comparatively low MAPE which is less than 1%. Thus, the proposed system can be a promising tool to assist decision-makers in construction management.

[1]  Akintola Akintoye,et al.  Macro-economic leading indicators of construction contract prices , 1998 .

[2]  R. Fellows.  Escalation management: Forecasting the effects of inflation on building projects , 1991 .

[3]  Ka Chi Lam,et al.  Prediction of tender price index directional changes , 2000 .

[4]  Ali Touran,et al.  Modeling Cost Escalation in Large Infrastructure Projects , 2006 .

[5]  Sandhya Samarasinghe,et al.  Neural Networks for Applied Sciences and Engineering , 2006 .

[6]  Trefor P. Williams,et al.  Predicting Changes in Construction Cost Indexes Using Neural Networks , 1994 .

[7]  Seokyon Hwang,et al.  Dynamic Regression Models for Prediction of Construction Costs , 2009 .

[8]  R. Storn,et al.  Differential Evolution: A Practical Approach to Global Optimization (Natural Computing Series) , 2005 .

[9]  Rainer Storn,et al.  Differential Evolution – A Simple and Efficient Heuristic for global Optimization over Continuous Spaces , 1997, J. Glob. Optim..

[10]  Haifeng Wang,et al.  Comparison of SVM and LS-SVM for Regression , 2005, 2005 International Conference on Neural Networks and Brain.

[11]  Min-Yuan Cheng,et al.  Artificial intelligence approaches to achieve strategic control over project cash flows , 2009 .

[12]  Jian Lu,et al.  Time Series Analysis of ENR Construction Cost Index , 2010 .

[13]  Jui-Sheng Chou,et al.  Optimizing the Prediction Accuracy of Concrete Compressive Strength Based on a Comparison of Data-Mining Techniques , 2011, J. Comput. Civ. Eng..

[14]  Min-Yuan Cheng,et al.  Evolutionary support vector machine inference system for construction management , 2009 .

[15]  Nasser M. Nasrabadi,et al.  Pattern Recognition and Machine Learning , 2006, Technometrics.

[16]  S. Thomas Ng,et al.  Forecasting construction tender price index in Hong Kong using vector error correction model , 2010 .

[17]  Ruhaidah Samsudin,et al.  A hybrid model of self-organizing maps (SOM) and least square support vector machine (LSSVM) for time-series forecasting , 2011, Expert Syst. Appl..

[18]  Raja R. A. Issa.  Application of Artificial Neural Networks to Predicting Construction Material Prices , 2000 .

[19]  Seung Heon Han,et al.  Time Series Analysis of Construction Cost Index Using Wavelet Transformation and a Neural Network , 2007 .

[20]  Sungwoo Moon,et al.  Stochastic Forecast of Construction Cost Index Using a Cointegrated Vector Autoregression Model , 2013 .

[21]  Chunlu Liu,et al.  An Economic Analysis of Construction Industries Producer Prices in Australia , 2010 .

[22]  Dilip Kumar Pratihar,et al.  Tuning of neural networks using particle swarm optimization to model MIG welding process , 2011, Swarm Evol. Comput..

[23]  Sai On Cheung,et al.  An integrated regression analysis and time series model for construction tender price index forecasting , 2004 .

[24]  Seokyon Hwang,et al.  Time Series Models for Forecasting Construction Costs Using Time Series Indexes , 2011 .

[25]  Min-Yuan Cheng,et al.  A novel time-depended evolutionary fuzzy SVM inference model for estimating construction project at completion , 2012, Eng. Appl. Artif. Intell..

[26]  Huanhuan Chen,et al.  Evolving Least Squares Support Vector Machines for Stock Market Trend Mining , 2009, IEEE Trans. Evol. Comput..

[27]  Johan A. K. Suykens,et al.  Least Squares Support Vector Machine Classifiers , 1999, Neural Processing Letters.

[28]  Awad S. Hanna,et al.  Computerized Approach for Forecasting the Rate of Cost Escalation , 1993 .

[29]  Johan A. K. Suykens,et al.  LS-SVMlab Toolbox User's Guide version 1.7 , 2003 .

[30]  Min-Yuan Cheng,et al.  Evolutionary fuzzy decision model for cash flow prediction using time-dependent support vector machines , 2011 .

[31]  Baabak Ashuri,et al.  Is the Information Available from Historical Time Series Data on Economic, Energy, and Construction Market Variables Useful to Explain Variations in ENR Construction Cost Index? , 2012 .

[32]  Min-Yuan Cheng,et al.  Hybrid use of AI techniques in developing construction management tools , 2003 .