A Knowledge-Based Framework for Quantity Takeoff and Cost Estimation in the AEC Industry Using BIM S. Aram a , C. Eastman a and R. Sacks b a College of Architecture, Georgia Institute of Technology, USA b Civil and Environmental Engineering, Technion - Israel Institute of Technology, Israel E-mail: shiva_aram@gatech.edu, charles.eastman@coa.gatech.edu, cvsacks@techunix.technion.ac.il Abstract An important set of information provided through Building Information Modeling (BIM) platforms are quantitative properties of design elements and assemblies. The capability to extract or deduce such quantitative properties from explicit and implicit model information is essential for bidding, procurement, production planning, and cost control activities in the AEC projects. Current solutions for quantity take off (QTO) and cost estimation (CE) are developed based on the assumptions that the design models are suitable, contain adequate information to perform these tasks efficiently and accurately. In practice often these criteria do not exist in the models that cost estimators receive. Many estimators, engineers and managers distrust BIM operations as a result or find it difficult to adopt a BIM-based preconstruction process. This leads to a cumbersome, manual and error-prone QT and CE process currently used by most construction companies. In order to overcome these shortcomings, we have developed a framework for a knowledge-based system to perform model based QTO and CE. This framework includes domain, reasoning, task and interface layers. This paper reports on the progress on an ongoing research effort which so far mostly focused on developing a domain layer and rule libraries for the reasoning layer. The domain layer contains a knowledge base which along with rule libraries were developed by acquiring and representing domain experts’ knowledge. The rule libraries include modules of rules to infer knowledge about different product features. The inferred knowledge will enable providing and representing model information in a compatible format for QTO and CE tasks. It facilitates filtering, grouping and representing feature information provided in design models based on criteria that determines their true cost behavior. Finally, this knowledge will enable forecasting the properties of product features absent from design models. Examples are drawn from various fields inside and outside of the AEC industry, with a focus on the precast concrete industry. Keywords Knowledge based systems, knowledge inference, quantity take off, cost estimation, precast concrete 1 Introduction Efficient and accurate quantity take off (QTO) and cost estimation (CE) are pivotal to a project’s success. They are knowledge-intensive [1]; they are the prerequisites to many other activities in a project from budgeting, bidding and contracting to value based design, production planning and budget control; they require extracting information based on the knowledge of domain experts about the rules and processes throughout the products and projects lifecycle. There are commercial software products available that attempt to semi-automate these tasks through augmenting the quantitative information elicited from design models, creating pre-structured yet customizable cost databases and reducing repetitive aspects of these tasks [2]. Based on our study, QTO software products need to maintain three conditions for their successful performance (i) architectural and structural design models to be readily suitable for quantity takeoff and cost estimation; (ii) all the needed information to be quantitative in nature; (iii) designers’ models to contain complete information needed for these tasks. In practice these conditions are rarely met. The focus here is not on users’ modeling practices and their use of correct The 31st International Symposium on Automation and Robotics in Construction and Mining (ISARC 2014)