An agent based modelling approach for the office space allocation problemTools Dediu, Alexandra, Landa-Silva, Dario and Siebers, Peer-Olaf (2018) An agent based modelling approach for the office space allocation problem. In: 2018 European Modeling and Simulation Symposium (EMSS 2018), 17-19 September 2018, Budapest, Hungary. Full text not available from this repository.AbstractThis paper describes an agent based simulation model to create solutions for the office space allocation (OSA) problem. OSA is a combinatorial optimization problem concerned with the allocation of available office space to a set of entities such as people. The objective function in the OSA problem involves the minimization of space misuse and the minimization of soft constraints violations. Several exact and heuristic algorithms have been proposed to tackle this problem. This paper proposes a rather different approach by decomposing the problem into smaller goals, which are delegated to individual agents each representing an entity in the problem. Agents have an internal decision making process which guides them throughout their search process for a better allocation (room). That is, agents seek to satisfy their individual requirements in terms of room space and constraints. Computational experiments show that the agent based model exhibits competitive performance in terms of solution quality and diversity when compared to neighborhood search heuristics.
Actions (Archive Staff Only)
|
Tools
Tools
