Owing to the advances in information technology and heightened awareness regarding disaster response, many evacuation simulations have been performed by researchers in recent years. It is necessary to develop suitable disaster prevention plans or evacuation plans using data generated by such simulations. However, it is difficult to understand the simulation results in their original form because of the detailed and voluminous data generated. In this study, we focus on tensor decomposition, which is employed for analyzing multi-dimensional data, in order to analyze the evacuation simulation data, which often consists of multiple dimensions such as time and space. Tensor decomposition is applied to the movement trajectory data generated in the evacuation simulation with the objective of acquiring important disaster or evacuation patterns.

1. [1] T. Oki and T. Osaragi, “Modeling Human Behavior of Local Residents in the Aftermath of a Large Earthquake – Wide-Area Evacuation, Rescue and Firefighting in Densely Built-Up Wooden Residential Areas,” J. Disaster Res., Vol.11, No.2, pp. 188-197, 2016.
2. [2] T. Osaragi and T. Oki, “Wide-Area Evacuation Simulation Incorporating Rescue and Firefighting by Local Residents,” J. Disaster Res., Vol.12, No.2, pp. 296-310, 2017.
3. [3] Ministry of Land, Infrastructure, Transport and Tourism, “Guidelines on Planning for Urban Development Preparing for Tsunami,” Jun. 2013 (in Japanese).
4. [4] Y. Kawai, J. Zhao, K. Sugiura, Y. Ishikawa, and Y. Wakita, “An Analysis Technique of Evacuation Simulation Using an Array DBMS,” J. Disaster Res., Vol.13, No.2, pp. 338-346, 2018.
5. [5] J. Zhao, Y. Ishikawa, Y. Wakita, , and K. Sugiura, “Difference Operators in Simulation Data Warehouses,” J. Disaster Res., Vol.12, No.2, pp. 347-354, 2017.
6. [6] J. Zhao, K. Sugiura, Y. Wang, , and Y. Ishikawa, “Simulation Data Warehouse for Integration and Analysis of Disaster Information,” J. Disaster Res., Vol.11, No.2, pp. 255-264, 2016.
7. [7] H. Lustosa, F. Porto, P. Valduriez, and P. Blanco, “Database System Support of Simulation Data,” Proc. VLDB Endow., Vol.9, No.13, pp. 1329-1340, 2016.
8. [8] A. Shashua and T. Hazan, “Non-negative Tensor Factorization with Applications to Statistics and Computer Vision,” Proc. Int’l Conf. on Machine Learning (ICML 2005), pp. 792-799, 2005.
9. [9] A. Ozerov, C. Févotte, R. Blouet, and J. Durrieu, “Multichannel nonnegative tensor factorization with structured constraints for user-guided audio source separation,” In Proc. Int’l Conf. on Acoustics, Speech and Signal Processing (ICASSP 2011), pp. 257-260, 2011.
10. [10] T. Matsubayashi, M. Kohjima, A. Hayashi, and H. Sawada, “Brand-choice analysis using non-negative tensor factorization,” Transactions of the Japanese Society of Artificial Intelligence, Vol.30, No.6, pp. 713-720, 2015 (in Japanese).
11. [11] “leaflet,” https://leafletjs.com [accessed March 14, 2019]
12. [12] K. Takeuchi, R. Tomioka, K. Ishiguro, A. Kimura, and H. Sawada, “Non-negative Multiple Tensor Factorization,” In Proc. Int’l Conf. on Data Mining (ICDM 2013), pp. 1199-1204, 2013.
13. [13] Y. Wang, Y. Zheng, and Y. Xue, “Travel Time Estimation of a Path Using Sparse Trajectories,” Proc. ACM SIGKDD Int’l Conf. on Knowledge Discovery and Data Mining (KDD 2014), pp. 25-34, 2014.