|M.Sc Student||Ben-Alon Lola|
|Subject||Simulating and Visualizing the Flow of Trade Crews using|
Agents and Building Information Models
|Department||Department of Civil and Environmental Engineering||Supervisor||Professor Rafael Sacks|
|Full Thesis text|
Simulation is particularly useful for testing different production control and information flow methods in construction, because field experiments suffer difficulties with isolating cause and effect. Existing methods such as Discrete Event Simulation are limited in their ability to model the behavior of crews and of individuals who have distinct behavior, context and knowledge. Agent-Based Simulation may offer a better solution.
The aim of this work was to build an experimental tool capable of reflecting the emergent nature of production in construction.
This required capturing trade crew behaviors through interviews and encapsulating the behavior in software agents. The system models trades' decision making and knowledge while using BIM models to define the physical and the process environment for the simulation.
The simulation exhibits the interdependence of individual workers and crews as they interact with each other and share resources. By varying parameters of material supply stability, arrival of design information, and agent behavior (which may be varied from agent to agent), the simulation is able to generate different aggregate system performances. The developed simulation tool was initially validated by testing predictable "what-if" scenarios. The resulting patterns for known conditions proved similar to those found in an actual building context in the construction site. Finally, the tool was used to generate emergent outcomes of more complex scenarios, which could not be predicted.
The simulation platform developed in this work is uniquely capable of testing the impact of different production organization methods. Significant among them are lean construction approaches, such as Last Planner System (LPS), because the model simulates the different production behaviors that result from the interaction of the myriad subcontracting teams and suppliers that perform construction work on and off site. In particular, the influence of each participants' knowledge, context, and motivations on their day to day decisions about resource allocation and work sequence can be modeled in the ABS, whereas they could not be modeled using DES. Furthermore, unlike the few existing research models, the simulation was situated in a realistic virtual environment modeled using BIM, allowing future experimental setups that can incorporate human subjects and real buildings.