Dealing with construction cost overruns using data mining

Construction Management and Economics (Impact Factor: 0.8). 06/2014; 32(7-8):682-694. DOI: 10.1080/01446193.2014.933854

ABSTRACT One of the main aims of any construction client is to procure a project within the limits of a predefined budget. However, most construction projects routinely overrun their cost estimates. Existing theories on construction cost overrun suggest a number of causes ranging from technical difficulties, optimism bias, managerial incompetence and strategic misrepresentation. However, much of the budgetary decision-making process in the early stages of a project is carried out in an environment of high uncertainty with little available information for accurate estimation. Using non-parametric bootstrapping and ensemble modelling in artificial neural networks, final project cost-forecasting models were developed with 1,600 completed projects in this experimental research. This helped to extract information embedded in data on completed construction projects, in an attempt to address the problem of dearth of information in the early stages of a project. 92% of the 100 validation predictions were within ±10% of the actual final cost of the project whiles 77% were within ±5% of actual final cost. This indicates the model's ability to generalise satisfactorily when validated with new data. The models are being deployed within the operations of the industry partner involved in this research to help increase the reliability and accuracy of initial cost estimates.

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Producing reasonably accurate cost estimates at the planning stage of a project important for the subsequent success of the project. The estimator has to be able to make judgement on the cost influence of a number of factors including site conditions, procurement, risks, price changes, likely scope changes or type of contract. This can shroud the estimation process in uncertainty, which has often resulted in project cost overruns. The knowledge acquisition, generalisation and forecasting capabilities of Artificial Neural Networks (ANN) are explored in this pilot study to build final cost estimation models that incorporate the cost effect of some of the factors mentioned above. Data was collected on ninety-eight water-related construction projects completed in Scotland between 2007-2011. Separate cost models were developed for normalised target cost and log of target costs. Variable transformation and weight decay regularisation were then explored to improve the final model’s performance. As a prototype of a wider research, the final model’s performance was very satisfactory, demonstrating ANN ability to capture the interactions between the predictor variables and final cost. Ten input variables, all readily available or measurable at the planning stages for the project, were used within a Multilayer Perceptron Architecture and a Quasi-Newton training algorithm.
    28th ARCOM Conference, Edinburgh, UK; 09/2012
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Purpose - Drawing on mainstream arguments in the literature, the paper presents a coherent and holistic view on the causes of cost overruns, and the dynamics between cognitive dispositions, learning and estimation. A cost prediction model has also been developed using data mining for estimating final cost of projects. Design/methodology/approach - A mixed-method approach was adopted: a qualitative exploration of the causes of cost overrun followed by an empirical development of a final cost model using Artificial Neural Networks (ANN). Findings - A conceptual model to distinguish between the often conflated causes of underestimation and cost overruns on large publicly funded projects. The empirical model developed in this paper achieved an average absolute percentage error of 3.67% with 87% of the model predictions within a range of ±5% of the actual final cost. Practical implications - The model developed can be converted to a desktop package for quick cost predictions and the generation of various alternative solutions for a construction project in a sort of what-if analysis for the purposes of comparison. The use of the model could also greatly reduce the time and resources spent on estimation. Originality/value - A thorough discussion on the dynamics between cognitive dispositions, learning and cost estimation has been presented. It also presents a conceptual model for understanding two often conflated issues of cost overrun and under-estimation.
    Journal of Financial Management of Property and Construction 02/2014; 19(1). DOI:10.1108/JFMPC-06-2013-0027
  • [Show abstract] [Hide abstract]
    ABSTRACT: Although extensive research has been undertaken on factors influencing the decision to tender and mark-up and tender price determination for construction projects, very little of this research contains information appropriate to the factors involved in costing construction projects. The object of this study was to gain an understanding of the factors influencing contractors' cost estimating practice. This was achieved through a comparative study of eighty-four UK contractors classified into four categories, namely, very small, small, medium and large firms. The initial analysis of the 24 factors considered in the study shows that the main factors relevant to cost estimating practice are complexity of the project, scale and scope of construction, market conditions, method of construction, site constraints, client's financial position, buildability and location of the project. Analysis of variance, which tests the null hypothesis that the opinions of the four categories of companies are not significantly different, shows that except for the procurement route and contractual arrangement factor there is no difference of opinion, at the 5% significance level, on the factors influencing cost estimating. Further analysis, based on a factor analysis technique, shows that the variables could be grouped into seven factors; the most important factor grouping being project complexity followed by technological requirements, project information, project team requirement, contract requirement, project duration and, finally, market requirement.
    Construction Management and Economics 02/2000; 18(1):77-89. DOI:10.1080/014461900370979 · 0.80 Impact Factor


Available from
Jul 14, 2014