Bridge design efforts and cost estimation models for PPCB bridge projects
The estimation of design effort and cost plays a vital role in authorizing funds and controlling budget during the project development process. Typically, the design phase consists of various engineering activities that require substantial efforts in delivering final construction documents for bid preparation. Estimating these efforts accurately and efficiently is critical for transportation agencies to properly allocate funds and assign appropriate time and resources.
Previous studies have reported several problems associated with the estimation of design effort such as lack of predictive tools, inaccurate forecasts and misallocation of efforts. Thus, there is a need for a proactive scheme to estimate more accurate and reliable design efforts and costs in order improve the confidence of the design office at the negotiation table with consulting firms and finally enhance the accountability and transparencies of funding decisions.
This study develops advanced design effort and cost estimation models using multivariate linear regression (MLR), multivariate polynomial regression (MPR) of second degree and Artificial Neural Network (ANN) methods. First, the study develops a master database that consolidates various data points of historical pretensioned prestressed concrete beam (PPCB) bridge projects designed by external consultants. The master database includes data attributes such as bridge design attributes, various physical attributes of bridges, consultant’s proposed fees and workhours, Iowa Department of Transportation (DOTs) proposed fees and workhours, contracted fee and work hours and actual amount of fee paid after the completion of design. Seven MLR, MPR and ANN models have been developed to estimate design fees and work hours at different negotiation and design stages. Two approaches are used to predict cost and workhours, the first one utilized the data from bridge design attributes and the other one utilizes bridge design attributes along with number of various kinks of design sheets. The MLR, MPR and ANN models are developed using commercial prediction analytics software JMP pro.
The performances of all the models are evaluated by comparing the MAPE (Mean absolute percentage error). The MLR models did not perform well because of their inability to recognize non-linear patterns whereas MPR models performed slightly better due to the ability of recognizing some non-linearity in the data. However, ANN models are able to detect non-linear patterns along with interactions in the training data. The results reveal that ANN models perform significantly better with MAPE range of 5-13% with only bridge data attributes and MAPE range of 4-12% with bridge data attributes and different design sheets as inputs.