Niemi, Elmeri (2024) Privacy-aware agent-based simulation for modeling indoor movement patterns in university campuses. Master thesis.
|
Text
niemis3783723mscthesisfinal.pdf Download (4MB) | Preview |
Abstract
This master thesis explores the development of a privacy-conscious methodology for modeling indoor movement patterns in campus environments using Agent-Based Simulation (ABS). The primary research question addresses how movement can be accurately modeled while ensuring the privacy of individuals. The study utilizes WiFi-based occupancy data and an adapted PageRank algorithm to predict movement behaviors within university buildings without individual tracking. The ABS model incorporates various factors, including agent ontologies, room attractiveness, and spatial relationships, to simulate realistic indoor movement patterns. The model is validated using real-world data from the University of Groningen, demonstrating strong correlation between simulated and observed building occupancy. The findings highlight the effectiveness of the methodology in replicating movement patterns across different areas of the building, though some areas required further refinement. The use of aggregate data ensures privacy preservation, successfully balancing the need for data-driven insights with ethical data practices. This research contributes a novel approach to the field of indoor spatial analysis, with practical applications for space management, energy efficiency, and campus planning. The flexible framework developed can be adapted to various indoor environments, offering significant potential for future research and implementation.
| Item Type: | Thesis (Master) |
|---|---|
| Degree programme: | Spatial Sciences (Research) |
| Supervisor: | Weitkamp, S.G. |
| Date Deposited: | 10 Sep 2024 13:15 |
| Last Modified: | 10 Sep 2024 13:16 |
| URI: | https://frw.studenttheses.ub.rug.nl/id/eprint/4763 |
Actions (login required)
 |
View Item |