Multi-sensor based landslide monitoring via transfer learning

Abstract

Landslides are severe geographical activities that result in large quantities of rock and debris flowing down hill-slopes, leading to thousands of casualties and billions of dollars in infrastructure damage every year worldwide. For detecting landslides, on-site sensor systems are widely applied for data collection and many existing statistical process control methods can be adopted for modeling and monitoring. However, the conventional methods may perform poorly or even inapplicable when the sensors have different set-up times and end times, especially when the system includes newly deployed sensors with limited data collected. To make effective use of such new sensors immediately after deployment, we propose a novel multi-sensor based charting scheme for dynamic landslide modeling and monitoring by using transfer learning. A regularized parameter-based transfer learning approach integrated with the ordered LASSO is first proposed to effectively transfer information from old sensors with sufficient historical data to new ones with limited data. The approach considers the similarities not only between the autoregressive coefficients of different sensors, but also between the temporal correlation patterns. A control chart is then proposed for monitoring the newly deployed sensors sequentially based on the generalized likelihood ratio. Extensive simulation results and a real data example of landslide monitoring demonstrate the effectiveness of our proposed method.

Keywords:

• Dynamic landslide monitoring
• generalized likelihood ratio
• ordered Lasso
• transfer learning

Acknowledgments

The authors sincerely acknowledge the efforts of the Editor, the Associate Editor and three anonymous referees that have resulted in significant improvements of this paper. The authors also thank Prof. Yu-Hsing Wang, director of the Data-Enabled Scalable Research (DESR) Laboratory, at HKUST for providing the context of the landslide example.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This paper is partly supported by National Natural Science Foundation of China [12071144; 71931004; 71931006; 11871324; 11771145], RGC GRF [16216119; 16201718], National Science Foundation of Shanghai [19ZR1414400], China Postdoctoral Science Foundation [2020M671064], National Bureau of Statistics of China [2020LD03], and the Fundamental Research Funds for the Central Universities.

Notes on contributors

Wendong Li

Wendong Li is an assistant professor of School of Statistics and Management, Shanghai University of Finance and Economics. His research interest is statistical quality control.

Fugee Tsung

Fugee Tsung is a Chair Professor and Acting Dean of the Information Hub, Guangzhou Campus, at the Hong Kong University of Science and Technology (HKUST). Based on his pioneering contribution to Quality Analytics research and education, he has been elected Academician of the International Academy for Quality (IAQ), Fellow of the American Statistical Association (ASA), Fellow of the Institute of Industrial and Systems Engineers (IISE), Fellow of the American Society for Quality (ASQ), Elected Member of the International Statistical Institute (ISI), and Fellow of the Hong Kong Institution of Engineers (HKIE). His research interests include Quality Analytics; Industrial Big Data; Statistical Process Control (SPC), Monitoring, and Diagnosis; Helping Solve the World's Problems Using Quality, Analytics, and Innovation.

Zhenli Song

Zhenli Song and Ke Zhang are PhDs of HKUST under the supervision of Professor Fugee Tsung.

Ke Zhang

Zhenli Song and Ke Zhang are PhDs of HKUST under the supervision of Professor Fugee Tsung.

Dongdong Xiang

Dongdong Xiang is an associate professor of School of Statistics, East China Normal University, Shanghai, China. His research interests include statistical quality control and multiple testing.