A framework proposal for construction site agent-based simulation

Construction sites are complex and dynamic environments where the continuous mutual interactions among a sheer number of entities (e.g. workers, machines, equipment) and activities determine the frequent occurrence of unpredictable hazardous events that makes risk estimation one of the most challenging tasks of Health & Safety (HS) management. In recent years, the growing adoption of Building Information Modeling (BIM) and Internet-of-Things (IoT) supported HS Managers in the improvement of construction site planning and monitoring and hence increased workers¶ safety. Nonetheless, probabilistic estimation of hazardous events and related risk estimation still heavily relies on HS Managers¶ field experience. This contribution proposes a novel framework for the adoption of an Agent-Based Modeling and Simulation approach integrated with BIM in a game engine environment for the probabilistic estimation of collision events in construction sites, between workers and heavy vehicles. The paper provides the definition of the hazardous event occurrence and discusses in detail the main aspects of the framework, including: the use of BIM data to create the simulation scenario, the agents and the rules that drive their behaviors. The presented framework is the first part of an ongoing research that will comprise its implementation and application in case study.