36 th International Symposium on Automation and Robotics in Construction (ISARC 2019) Optimizing Site Layout Planning utilizing Building Information Modelling Abhishek Raj Singh a , Yash Patil b and Venkata Santosh Kumar Delhi c a PhD candidate, Department of Civil Engineering, Indian Institute of Technology Bombay, India b Undergraduate Student, Department of Civil Engineering, Indian Institute of Technology Madras, India c Assistant Professor, Department of Civil Engineering, Indian Institute of Technology Bombay, India E-mail: arsingh@iitb.ac.in, ce16b137@smail.iitm.ac.in, venkatad@iitb.ac.in Abstract Site layout planning (SLP) is categorized as a non-deterministic polynomial time (NP) hard or complete class problem. Inefficient SLP can lead to congestion, safety conflicts and productivity reductions. Significant attention to the problem is evident in the field of construction management. A number of optimization routines and mathematical models are suggested in past research to reduce costs associated with improper layouts. However, such models are seldom adopted on real-life projects, where SLP is primarily carried out based on heuristics. The two significant inhibitors to the adoption of sophisticated approaches identified in this study are; lack of realism in the mathematical models and the significant effort involved in setting up the model for each construction site. These two inhibitors tend to reflect as reluctance on the part of the project teams to adopt SLP models. In the present study, the first inhibitor is addressed by incorporating realism into the mathematical model for SLP. The SLP is formulated as an optimization problem involving the reduction of transportation cost on construction site with associated constraints. Realism to the optimization model was brought through the modelled travel distances utilizing Building Information Modelling (BIM). Genetic Algorithm (GA) was used to optimize the objective function. The combined model thus incorporates all the site constraints in terms of travel paths as captured in the BIM model thus bringing in more realism into the SLP modelling. This work is preliminary work in developing a fully automated SLP process where the second inhibitor would also be addressed. Keywords Site Layout Planning; Optimization; Building Information Modelling; Genetic Algorithm 1 Introduction Research to address the problem of layout planning also exists in sectors like manufacturing, electronics, computer science and information technology and construction. The objectives to achieve and the constraints encountered are unique to the Architecture, Engineering and Construction (AEC) industry. The layout planning for a construction project starts at the initial phase of the project. The objectives during this phase are not limited to the identification of potential locations to accommodate temporary facilities (TFs) [1], finalizing the routes for vehicular movement [2] and selection of equipment [3]. Intertwined tasks brings more complexity to be handled while planning layouts for construction sites [4]. There exist literature where researchers have tried modelling SLP as a mathematical problem. Mathematical models too possessed the complexity and were referred to NP-Hard [5] and NP- Complete [6] class problems. Despite enormous research in the domain of layout planning, the AEC industry utilizes heuristics of the experienced stakeholders on the project for the purpose of SLP. Sometime the drawbacks of layout through experiential learning come to forth in the form of site congestion, restricted access, unnecessary vehicle movement, multiple handling of material etc. The mathematical approach existing in literature provides a segmental solution to the layout problem and lacks in capturing the realism of a construction site [7], [8]. Although the studies in the domain of layout planning have focused on developing a varied approach to tackle different objectives [7] and comparing the algorithms, adopted in the domain for the solution search [6]. This research in contrast presents an effort to understand the inhibitors to the adoption of existing mathematical models and approach. An evolution in the mathematical approach is also evident to make mathematical models closer to real site scenario; from discrete space layout planning problem to continuous space optimization [3] and from