Predicting final cost for competitively bid construction projects using regression models

This paper studies the relationship between the low bid and completed cost for competitively bid highway projects. Data are analyzed from several agencies managing highway and dredging projects. A natural log transformation of the low bid and final cost was found to produce regression models for each agency that had high R and R² values. These models can predict completed project cost using only the project low bid as input. A separate regression model was determined for each agency. Based on the form of the calculated models it appears that highway agencies construct projects where the final cost increases as an increasing percentage of the low bid price as the project magnitude increases. The dredging projects appear to follow different mechanisms of cost increase than the highway projects and were not predicted as well by the regression models. The regression models can also be used for budgeting purposes by submitting the sum of the low bids for a group of projects as input. The regression models output a prediction of the cost of the group of projects that was found to be highly accurate.     

Predicting construction cost overruns using text mining,numerical data and ensemble classifiers

This paper discusses how text describing a construction project can be combined with numerical data to produce a prediction of the level of cost overrun using data mining classification algorithms. Modeling results found that a stacking model that combined the results from several classifiers produced the best results. The stacking ensemble model had an average accuracy of 43.72% for five model runs. The model performed best in predicting projects completed with large cost overruns and projects near the original low bid amount. It was found that a stacking model that used only numerical data produced predictions with lower precision and recall. A potential application of this research is as an aid in budgeting sufficient funds to complete a construction project. Additionally, during the planning stages of a project the research can be used to identify a project that requires increased scrutiny during construction to avoid cost overruns.     

Risk‐minimizing approach to bid‐cutting limit determination

Intense price competition is quite common in the construction industry. In many markets, contractors have to cut their bids to compete, giving priority to winning enough contracts to sustain normal operation, and it is common to see a winning bid close to the expected project cost. While cutting bids not only gives up profits but also undoubtedly increases the risk of making a loss, the behaviour of contractors in intense competition is difficult to explain by existing academic bidding models. An approach to determining the lower limit of the bid for a project is proposed based on minimization of the overall loss risk defined by a probabilistic model. The approach can be used to prevent arbitrary over‐cuts in final bid decision where price competition is intense. Factors influencing the suggested bid‐cutting limit for a project are analysed. An illustrative example using real case data is provided.     

The accuracy of preliminary cost estimates in Public Works Department (PWD) of Peninsular Malaysia

This paper presents the accuracy of preliminary cost estimates prepared by Public Works Department (PWD), which is the largest public construction management in Malaysia. It attempts to understand the Quantity Surveyors' (QS) estimation accuracy in relation to public projects. This study analyses 83 projects of estimates and tender bids. The analysis includes three (3) estimating targets i.e. lowest bid, accepted bid and mean of the bids. To broaden the study, 344 QS involved in the procurement answered the questionnaires. Linear multiple regression analysis on project characteristics shows that project size, number of bidders, location and type of schools affect the bias. Contract period affects the consistency. The use of mean of the bids is the best-fit target to explain the bias in terms of adjusted R². The accuracy may improve if sufficient design information is available, proper cost planning and improving the application of historical cost data. As an alternative, the phenomenon of overestimation is resulted from government directive instruction, which could challenge the rational of accurate estimate.     

Data modelling and the application of a neural network approach to the prediction of total construction costs

Neural network cost models have been developed using data collected from nearly 300 building projects. Data were collected from predominantly primary sources using real-life data contained in project files, with some data obtained from the Building Cost Information Service, supplemented with further information, and some from a questionnaire distributed nationwide. The data collected included final account sums and, so that the model could evaluate the total cost to the client, clients' external and internal costs, in addition to construction costs. Models based on linear regression techniques have been used as a benchmark for evaluation of the neural network models. The results showed that the major benefit of the neural network approach was the ability of neural networks to model the nonlinearity in the data. The ‘best’ model obtained so far gives a mean absolute percentage error (MAPE) of 16.6%, which includes a percentage (unknown) for client changes. This compares favourably with traditional estimating where values of MAPE between 20.8% and 27.9% have been reported. However, it is anticipated that further analyses will result in the development of even more reliable models.     

Bidding strategy to support decision-making by integrating fuzzy AHP and regression-based simulation

This study proposes a new bidding strategy to support decision-making that is based on a combined framework of the Fuzzy Analytical Hierarchy Process (FAHP) and regression-based simulation. First, the FAHP method integrates the AHP with fuzzy set theory to determine the weights of factors that influence the cost of a project. Second, the integration of the cumulative distribution functions that are generated by the Monte Carlo simulation with a regression model yields bid amounts that correspond to various confidence levels. The proposed approach is used to analyze data on bridge construction projects that are taken from a database of the Taiwan Public Construction Commission. The systematic bid assessment model and the cost-probability curve can be used as strategic tools for quantifying project risks and calculating bids for construction projects.     

现有工程招标评标的不规范现象及对策

工程项目评标是招标工作的主要环节,评标决策方法的科学、客观与否,直接关系到招标工作的成功,保证公开、公平、公正地选择中标人,是建筑市场环境的综合反映。现有工程招标评标的不规范现象,其根源就是标底的存在。随着市场经济的发展,采用“最低价中标”的方法是大势所趋。但“最低价中标”原则却遭到了许多人的反对。投标人为了不使总报价过高,对在施工过程中无望增加工程量或可能减少工程量的项目压低报价。同时,有些投标人不惜一切手段了解专家名单,长期拉拢某些专家,影响评标的科学性和公正性。规范招投标行为,提高招投标透明度,实质是通过公开、公平、公正的招投标活动,将建立高效、规范的管理体系的过程法律化。     

Fuzzy Logic Model for Determining Minimum Bid Markup

Abstract:   Many construction markets exhibit severe price competition where contractors have to cut their bids to compete, giving priority to winning enough contracts to sustain normal operation, and it is common to see a winning bid close to the expected project cost. While cutting bids not only gives up profits but also undoubtedly increases the risk of making a loss, the behavior of contractors in intense competition is difficult to explain by existing models. A fuzzy-logic-based model is proposed for determining the minimum bid markup with assessments of chance of winning and loss risk. The model incorporates the position of a decision maker in the fuzzy rules according to his/her attitude toward risk and degree of need for the job. Two illustrative examples, one hypothetical and one real, are provided, in which differences in priorities are simulated by four sets of fuzzy rules for a comparison of the effects. The results show that the model is sensitive enough to differentiate a decision maker's position on bidding and suggest bid-cutting limits consistently, thereby remedying some shortcomings of existing models.     

Parameter prediction for cash flow forecasting models

The paper describes the application of the DHSS formula to 27 completed construction projects comprising four types - steel-framed low rise buildings, new build housing developments, housing refurbishment projects, and multi-house ‘pre-paint’ maintenance contracts. Application of the formula to individual projects indicates that the ‘best’ parameter values offer a ten-fold improvement over the published values based on project size. Similar results occur when using the best parameter values for other two parameter models.

Various approaches are considered in attempting to predict the best parameter values of the models based on known characteristics of the project. A multiple linear regression with project value, duration, and type independent variables is shown not to produce any significant improvement on standard DHSS formula predictions. However, a reduction in the number of independent variables by cross validation produces an approximately 25% improvement on standard DHSS formula forecasts outside the data base. Examination of the models derived from this analysis indicate the type of project to be of major importance.     

A study of preproject planning and project success using ANNs and regression models

It has long been recognized by the industry practitioners that how well preproject planning is conducted has a great impact on project outcome. Through industry project data collection and model analysis, this research intends to investigate the relationship between preproject planning and project success. Preproject planning and project performance information from 62 industrial projects and 78 building projects, representing approximately $5 billion U.S.D. in total construction cost, is collected and used for this research analysis. Based on the information obtained, preproject planning is identified as having direct impact on the project success (cost and schedule performance). Two techniques were then used to develop models for predicting cost and schedule performance: statistical regression analysis, and artificial neural networks (ANNs). The research results provide a valuable source of information that supports better planning in the early stage of the project life cycle and have positive impact on the final project outcome.     

Estimating moment capacity of ferrocement members using self-evolving network

In this paper, an empirical model based on self-evolving neural network is proposed for predicting the flexural behavior of ferrocement elements. The model is meant to serve as a simple but reliable tool for estimating the moment capacity of ferrocement members. The proposed model is trained and validated using experimental data obtained from the literature. The data consists of information regarding flexural tests on ferrocement specimens which include moment capacity and cross-sectional dimensions of specimens, concrete cube compressive strength, tensile strength and volume fraction of wire mesh. Comparisons of predictions of the proposed models with experimental data indicated that the models are capable of accurately estimating the moment capacity of ferrocement members. The proposed models also make better predictions compared to methods such as the plastic analysis method and the mechanism approach. Further comparisons with other data mining techniques including the back-propagation network, the adaptive spline, and the Kriging regression models indicated that the proposed models are superior in terms prediction accuracy despite being much simpler models. The performance of the proposed models was also found to be comparable to the GEP-based surrogate model.     

Detection and prevention of unbalanced bids

Unbalanced bidding is a serious problem for the construction owner because it may increase the cost of construction The most common way to mathematically unbalance a bid is frontloading where a bidder overstates the unit price of line items scheduled to be performed early in the project and understates the unit price of line items performed later. A bidder can also overstate the unit price of a line item whose quantity was somehow underrated by the engineer. If the owner proves that a mathematically unbalanced bid costs more to perform, the bid is said to be materially unbalanced, in which case the owner can reject the bid. A model is presented that formalizes and automates the process of detecting mathematically and materially unbalanced bids by comparing line item prices with the engineer’s estimates and the average prices offered by the bidders. This model allows owners to detect and reject unbalanced bids, and deters bidders from unbalancing their bid.     

The effect of client and type and size of construction work on a contractor’s bidding strategy

This paper offers a bidding strategy model for use by contractors as part of a more informed approach in selecting which contracts to bid for, and as a basis for determining the most appropriate mark-up level for various types and sizes of construction work and client types. Regression analysis is used in measuring a contractor’s competitiveness between bids (by using the lowest bid/own bid ratio) and within bids (by using the lowest bid/cost estimate ratio) according to type and size of construction work and client type. The model was tested on a large and reputable Hong Kong contractor. This particular contractor’s bidding behaviour was found to be largely unaffected by the type of construction work, but significantly affected by the client type and the size of the construction work. Three quadratic models (regressing lowest bid/cost estimate on the size of the construction work) are also successfully developed for projects from the private sector, the Hong Kong Government and the Hong Kong Housing Authority, respectively.     

价值管理在设计评标中的应用

在招标投标中,评标是一个非常重要的环节。目前我国招标法规定一般采用合理最低价中标,评标也是采用经评审的最低价中标。通过这样的招标,业主只能从参与投标的单位中择优选择设计(或施工)单位,但无法评定项目的价值。在面对实力相当的投标单位时,业主应该如何评价和选择合适的设计(或施工)单位,以实现“降低成本,提高功能”的最终目标,本文应用价值管理的原理和方法,很好地解决了这个问题。     

Hybrid ensemble soft computing approach for predicting penetration rate of tunnel boring machine in a rock environment

This study implements a hybrid ensemble machine learning method for forecasting the rate of penetration (ROP) of tunnel boring machine (TBM), which is becoming a prerequisite for reliable cost assessment and project scheduling in tunnelling and underground projects in a rock environment. For this purpose, a sum of 185 datasets was collected from the literature and used to predict the ROP of TBM. Initially, the main dataset was utilised to construct and validate four conventional soft computing (CSC) models, i.e. minimax probability machine regression, relevance vector machine, extreme learning machine, and functional network. Consequently, the estimated outputs of CSC models were united and trained using an artificial neural network (ANN) to construct a hybrid ensemble model (HENSM). The outcomes of the proposed HENSM are superior to other CSC models employed in this study. Based on the experimental results (training RMSE = 0.0283 and testing RMSE = 0.0418), the newly proposed HENSM is potential to assist engineers in predicting ROP of TBM in the design phase of tunnelling and underground projects.     

深基坑变形的神经网络预测

根据神经网络的基本原理,结合工程实例,建立了基坑变形预测的神经网络模型,并运用该模型对工程实例进行了预测,预测结果与实际变形拟合良好,表明了神经网络法在基坑变形预测中的有效性。     

Integration of Simulation-Based Cost Model and Multi-Criteria Evaluation Model for Bid Price Decisions

Abstract:   Several criteria affect bidding decisions. Current bidding models determine a markup based on a fixed project construction cost. This work presents a novel bid price determination procedure that is built by integrating a simulation-based cost model and a multi-criteria evaluation model. The cost model is used to consider cost uncertainties and generate a bid price cumulative distribution, whereas the multi-criteria evaluation model applies pairwise comparisons and fuzzy integrals to reflect bidder preferences regarding decision criteria. The relationship between the two models is based on a practical phenomenon in that a bidder has a high probability of winning when criteria evaluations favor his bid, and, consequently, the bidder would bid a low price, and vice versa. The merits of the proposed procedure are demonstrated by its application to two construction projects in Taiwan .     

Predicting construction cost and schedule success using artificial neural networks ensemble and support vector machines classification models

It is commonly perceived that how well the planning is performed during the early stage will have significant impact on final project outcome. This paper outlines the development of artificial neural networks ensemble and support vector machines classification models to predict project cost and schedule success, using status of early planning as the model inputs. Through industry survey, early planning and project performance information from a total of 92 building projects is collected. The results show that early planning status can be effectively used to predict project success and the proposed artificial intelligence models produce satisfactory prediction results.     

Approximately predicting the cost and duration of school reconstruction projects in Taiwan

Regression and neural network models have been developed to predict the cost and duration of projects for the reconstruction of schools which must be quickly rebuilt. Data for the school reconstruction projects in central Taiwan, which received the most serious damage from the Chi‐Chi Earthquake, were collected and analysed. The analytical results demonstrate that the floor area provides a good basis for estimating the cost and duration of school reconstruction projects, and suggest that the neural network model with back‐propagation learning technique is a feasible approach that yields better prediction results than the regression model for school reconstruction projects.     

A neural network based system for predicting earthmoving production

An artificial neural network based system (NN earth) is developed for construction practitioners as a simple tool for predicting earthmoving operations, which are modelled by back propagation neural networks with four expected parameters and seven affecting factors. These networks are then trained using the data patterns obtained from simulation because there are insufficient data available from industrial sources. The trained network is then incorporated as the computation engine of NN earth. To engender confidence in the results of neural computation, a validation function is implemented in NN earth to allow the user to apply the engine to historic cases prior to applying it to a new project. An equipment database is also implemented in NN earth to provide default information, such as internal cost rate, fuel cost, and operator's cost. User interfaces are developed to facilitate inputting project information and manipulating the system. The major functions and use of NN earth are illustrated in a sample application. In practice, NN earth can assist the user either in selecting a crew to minimize the unit cost of a project or in predicting the performance of a given crew.