光華講壇——社會(huì)名流與企業(yè)家論壇第6541期
主 題：Learning Short and Long Term Failure Patterns from Massive Network Failure Data 從大量網(wǎng)絡(luò)故障數(shù)據(jù)中學(xué)習(xí)短期和長(zhǎng)期故障模式
主講人：新加坡國(guó)立大學(xué) 葉志盛教授
主持人：管理科學(xué)與工程學(xué)院 肖輝教授
時(shí)間：6月4日 9:30-11:30
舉辦地點(diǎn)：柳林校區(qū)通博樓D301會(huì)議室
主辦單位：管理科學(xué)與工程學(xué)院 科研處
主講人簡(jiǎn)介：
葉博士獲得清華大學(xué)材料科學(xué)與工程和經(jīng)濟(jì)學(xué)聯(lián)合學(xué)士學(xué)位(2008)。獲新加坡國(guó)立大學(xué)博士學(xué)位。他目前是新加坡國(guó)立大學(xué)工業(yè)系統(tǒng)工程與管理系教授兼系主任。他的研究領(lǐng)域包括可靠性建模、估計(jì)和優(yōu)化、基于狀態(tài)的維護(hù)和數(shù)據(jù)驅(qū)動(dòng)的決策。他的研究成果發(fā)表在可靠性、統(tǒng)計(jì)學(xué)和運(yùn)營(yíng)管理領(lǐng)域的旗艦期刊上，包括Technometrics、JQT、IISE Trans、ITR、RESS、JRSS-B、JASA、Biometrika、JMLR、MSOM和POMS。
內(nèi)容簡(jiǎn)介：
Many lifeline infrastructure systems consist of thousands of components configured in a complex directed network. Disruption of the infrastructure constitutes 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. In this talk, we focus on both short term cascading failures and long term ageing failures. Repair of a pipe might generate shocks to neighbouring pipes and cause short term cascading failures. Understanding the short-term cascading failure is important for the utility to allocate additional resources to monitor the neighbouring pipes after a repair. On the other hand, understanding long-term failures is helpful in risk analysis of the whole pipe network and prioritizing replacements of old pipes. Statistical modelling of the two failure modes are extremely challenging because of the large pipe network and the huge failure data set. We develop novel statistical methods that are computationally tractable to fit the data. Applying the methods to a large data set from the Scottish Water network, we demonstrate the usefulness of our models in aiding operation management and risk assessment of the water utility.
許多生命線(xiàn)基礎(chǔ)設(shè)施系統(tǒng)由數(shù)千個(gè)組件組成，這些組件配置在一個(gè)復(fù)雜的定向網(wǎng)絡(luò)中?；A(chǔ)設(shè)施的中斷構(gòu)成了有向網(wǎng)絡(luò)上反復(fù)出現(xiàn)的故障過(guò)程。由于大量節(jié)點(diǎn)之間具有不規(guī)則連接，因此對(duì)此類(lèi)網(wǎng)絡(luò)復(fù)發(fā)數(shù)據(jù)的統(tǒng)計(jì)推斷具有挑戰(zhàn)性。在本次演講中，我們將重點(diǎn)關(guān)注短期級(jí)聯(lián)故障和長(zhǎng)期老化故障。管道的修復(fù)可能會(huì)對(duì)鄰近的管道產(chǎn)生沖擊，并導(dǎo)致短期級(jí)聯(lián)故障。了解短期級(jí)聯(lián)故障對(duì)于公用事業(yè)公司在維修后分配額外資源來(lái)監(jiān)測(cè)鄰近管道非常重要。另一方面，了解長(zhǎng)期故障有助于對(duì)整個(gè)管網(wǎng)進(jìn)行風(fēng)險(xiǎn)分析，并優(yōu)先更換舊管道。由于管網(wǎng)龐大，故障數(shù)據(jù)集龐大，兩種失效模式的統(tǒng)計(jì)建模極具挑戰(zhàn)性。我們開(kāi)發(fā)了新的統(tǒng)計(jì)方法，這些方法在計(jì)算上易于處理以擬合數(shù)據(jù)。將這些方法應(yīng)用于蘇格蘭水網(wǎng)的大型數(shù)據(jù)集，我們證明了我們的模型在幫助水務(wù)公司的運(yùn)營(yíng)管理和風(fēng)險(xiǎn)評(píng)估方面的
