報告題目：Data-Driven Risk Evaluation of a Large-Scale Pipe Network
報 告 人： 葉志盛 副教授 新加坡國立大學
報告時間：2022 年 05 月 13 日 下午 16:00-17:00
報告地點：騰訊會議 ID：122-570-145
或點擊鍊接直接加入會議：https://meeting.tencent.com/dm/wyPxjjdNru74
校内聯系人：趙世舜 zhaoss@jlu.edu.cn
報告摘要：A lifeline infrastructure system usually has thousands of elements configured in a complex network structure. The failures/repairs of the infrastructure constitute a recurrent failure process over a directed network. Statistical inference for such network recurrence data is challenging because of the large number of nodes with irregular connections among them. Based on 16-years of operation records in Scottish Water, we propose a network Gamma-Poisson Autoregressive NHPP (GPAN) model for recurrent failure data from a large-scale directed physical network. The model consists of two layers, where the temporal layer applies an NHPP with frailty for each node, and the spatial layer uses a well-orchestrated gamma-Poisson autoregressive scheme to establish correlations among the frailties. Under the network-GPAN model, we develop a sum-product algorithm to compute the marginal distribution for each frailty conditional on the recurrence data. The ability to rapidly compute these marginal distributions allows adoption of an EM algorithm for estimation. The developed model is applied to a subset of the Scottish Water network where we demonstrate the usefulness in aiding operation management and risk assessment of the water utility.
報告人簡介：葉志盛博士本科畢業于清華大學材料科學與工程系，博士就讀于新加坡國大工業與系統工程系。現在為新加坡國大工業系統工程與管理系副教授。他的主要研究方向包括剩餘壽命預測，可靠性建模，及數據驅動的運營決策。
