The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLII-4/W5
https://doi.org/10.5194/isprs-archives-XLII-4-W5-45-2017
https://doi.org/10.5194/isprs-archives-XLII-4-W5-45-2017
05 Oct 2017
 | 05 Oct 2017

PERCEPTION MODELLING OF VISITORS IN VARGAS MUSEUM USING AGENT-BASED SIMULATION AND VISIBILITY ANALYSIS

B. G. Carcellar III

Keywords: GIS, Perception Modelling, Agent-Based Simulation, Museum Management, GAMA

Abstract. Museum exhibit management is one of the usual undertakings of museum facilitators. Art works must be strategically placed to achieve maximum viewing from the visitors. The positioning of the artworks also highly influences the quality of experience of the visitors. One solution in such problems is to utilize GIS and Agent-Based Modelling (ABM). In ABM, persistent interacting objects are modelled as agents. These agents are given attributes and behaviors that describe their properties as well as their motion. In this study, ABM approach that incorporates GIS is utilized to perform analyticcal assessment on the placement of the artworks in the Vargas Museum. GIS serves as the backbone for the spatial aspect of the simulation such as the placement of the artwork exhibits, as well as possible obstructions to perception such as the columns, walls, and panel boards. Visibility Analysis is also done to the model in GIS to assess the overall visibility of the artworks. The ABM is done using the initial GIS outputs and GAMA, an open source ABM software. Visitors are modelled as agents, moving inside the museum following a specific decision tree. The simulation is done in three use cases: the 10 %, 20 %, and 30 % chance of having a visitor in the next minute. For the case of the said museum, the 10 % chance is determined to be the closest simulation case to the actual and the recommended minimum time to achieve a maximum artwork perception is 1 hour and 40 minutes. Initial assessment of the results shows that even after 3 hours of simulation, small parts of the exhibit show lack of viewers, due to its distance from the entrance. A more detailed decision tree for the visitor agents can be incorporated to have a more realistic simulation.