Publications / 2018 Proceedings of the 35th ISARC, Berlin, Germany

Holonic Execution System for Construction Management

Berardo Naticchia, Massimo Vaccarini and Alessandro Carbonari
Pages 501-508 (2018 Proceedings of the 35th ISARC, Berlin, Germany, ISBN 978-3-00-060855-1, ISSN 2413-5844)
Abstract:

Ensuring high efficiency in construction through advanced planning techniques must be supported by the simulation of construction processes. In previous research, the potential for multi-agent based simulation to support the planning phase were successfully shown with regard to some typical construction tasks. Among the main benefits determined by this approach the following ones were reported: capability of simulating real-world processes and interaction among resources; powerful 3D visualization; optimal sizing of resources. However, those models are not easy to create, because of the absence of built-in blocks and the many relationships among single agents and between agents and context. In this paper we propose the application of a novel holonic system for construction management, which is a sub-category of multi-agent systems. It combines the high and predictable performance promised by hierarchical systems with the robustness against disturbances and the agility of heterarchical systems. Holons are self-contained, stable, intermediate forms, that can be used to create a bigger whole. By using holons, designers aim to create complex systems that can be reused and redesigned much more rapidly than standard multi-agent systems. The approach was applied to the optimal scheduling of bored piles construction. Hence, a holonic construction execution system for bored piles tasks was developed. The holonic architecture was designed so as to integrate the DMAS pattern for scheduling optimization through stigmergy. The results of this paper showed the easy reusability of our simulation sub-systems and the potentials in terms of automatic optimal planning of construction works.

Keywords: Construction management, multi-agent systems, VRP