A Process for the Development and Evaluation of Preliminary Construction Material Quantity Estimation Models Using Backward Elimination Regression and Neural Networks
Journal of Cost Analysis and Parametrics
During the early stages of a project, it is beneﬁcial to have an accurate preliminary estimate of its cost. One way to make those estimates is by determining the amount of construction material quantities that are required and then multiplying the estimated construction material quantities by the corresponding unit cost. One advantage of making estimates in this way is that it allows for the segregation of quantities and costs. This way they can be updated separately as new information becomes available. They can also be tracked separately allowing decision makers to make better decisions about the project during its conceptual phase. There are several techniques that can be used to develop estimation models. The most common include regression analysis and artiﬁcial intelligence, such as neural networks. Work has been done, however, in a non-standardized way, leaving practitioners without guidance as to how to develop and evaluate models for their speciﬁc purposes. This can be seen in particular in the many different types of metrics used for the evaluation of models. The goal of the work presented in this article was to create a process to (1) develop models to be used to prepare preliminary estimates of construction material quantities taking into consideration the available data during the early stages of a project, and (2) evaluate the developed models using the Akaike information criterion. The proposed process is illustrated with an example in which data from 58 storage buildings was used to develop models to estimate the amount of concrete and reinforcement required using backward elimination regression analysis and neural network techniques. The developed models were then evaluated using a second-order correction Akaike information criterion (AICc) to select the most accurate model for each construction material quantity. The proposed process is useful for practitioners in need of developing robust estimation models in a consistent and systematic way, and the AICc metric proved to be effective for selecting the most accurate models from a set.
Borja García de Soto received the degree of Bachelor of Science in Civil Engineering from Florida International University (FIU) in 2000, the degree of Master of Science in Civil Engineering in the area of Structural Design from FIU in 2001 and the degree of Master of Science in Engineering in the area of Engineering and Project Management from the University of California at Berkeley (UC-Berkeley) in 2004. He is a registered Civil Engineer (licensed in California and Florida) with international experience in multiple aspects of project management, including risk management and control, delay analysis, forensic engineering, and project cost estimation. He worked as a Project Engineer in charge of the structural design of steel, wood, and reinforced concrete structures at FC Consulting Engineers, Inc. in Miami, Florida (2000-2002). In 2003 he was a Project Manager at Cyopsa-Sisocia, S.A. (Madrid) for the Northern Spain Region. After completing his MSc at UC-Berkeley in 2004, he joined JKA Inc., a construction consulting ﬁrm in the San Francisco Bay Area, California (2004-2010). As a Senior Consultant and licensed Professional Engineer at JKA he was responsible for a large variety of fast-paced projects covering multiple aspects of the construction industry. In July 2010 he founded the BGSL Consulting Group. Until July 2011 he led a large variety of fast-paced projects in the con-struction industry. In April 2011 he became part of the Infrastructure Management Group (IMG) in the Institute of Construction and Infrastructure Management (IBI) at the Swiss Federal Institute of Technology in Zürich (ETHZ).
Bryan T. Adey obtained a Bachelor’s of Civil Engineering from Dalhousie University in 1995, a Master’s of Science in Structural Engineering from the University of Alberta in 1997 and a Ph.D. in the area of Infrastructure Management from the Swiss Federal Institute of Technology in Lausanne (EPFL), Switzerland in 2002. His Ph.D. focused on the integration of the consideration of natural hazards into bridge management systems. After completing his Ph.D. he was employed in the Division of Maintenance and Safety at the EPFL (2002-2003). In 2003, Dr. Adey co-founded the consultancy Infrastructure Management Consultants Ltd. (IMC) and assumed the role as vice-president. From 2003-2009 he evaluated, developed and improved business models and business processes related to the maintenance and operation of infrastructure, developed and implemented infrastructure management systems, and analyzed infrastructure with respect to physical condition, risk and economics. This worked focused principally on road and rail infrastructure, but also included water distribution and water transportation networks. During this time he also continued research in the area of infrastructure management, in particular focusing on the methodologies to be used to estimate the total costs of infrastructure interventions, to evaluate the risk associated with road networks, and to assess risk reducing intervention strategies for road networks. In 2010, Dr. Adey left IMC to become the head of the Infrastructure Management Group (IMG) in the Institute of Construction and Infrastructure Management (IBI) at the Swiss Federal Institute of Technology in Zürich (ETHZ). His vision for the IMG is to be a world leader in the provision of cutting edge frameworks, methodologies, models and tools to improve the management of infrastructure.
Dilum Fernando obtained a Bachelor’s of Civil Engineering from Monash University in 2005. During the last semester of his bachelor studies, he joined Connell Wagner (pty) Ltd., as a part time design engineer. After completing his bachelor’s degree in May 2005, he joined Connell Wagner (pty) Ltd. as a full time design Engineer and worked there until April 2006. After that Dr. Fernando started his PhD studies at The Hong Kong Polytechnic University in the area of Carbon Fibre Reinforced Polymer (CFRP) strengthening of metallic structures. His main research topics involved bond behavior between CFRP and steel, analytical and numerical modelling of debonding behavior of CFRP-steel bond joints and numerical modelling of the behavior of various CFRP strengthened metallic struc-tures. After completing his PhD in 2010, he joined the Infrastructure Management Group (IMG) in the Institute of Construction and Infrastructure Management (IBI) at the Swiss Federal Institute of Technology in Zürich (ETHZ), as a Post-Doctoral fellow. The main research topics during his post-doctoral work at ETHZ involved, modelling the behavior of transportation infrastructure objects and networks under hazard situations, sustainable design of transportation infrastructure systems, which consider variations of design param-eters, novel materials and techniques, and the development of administrative tools that facilitate the improved design and management of infrastructure assets. Dilum Fernando joined The University of Queensland in August 2013 as a lecturer in the School of Civil Engineering. His recent research topics include, structural rehabilitation, innovative appli-cations of emerging materials in new structures, advanced numerical modeling, sustainable design and management of infrastructure assets.