BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:America/New_York BEGIN:DAYLIGHT DTSTART:20240310T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20231105T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:EST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20240308T001646Z UID:7E85D1C3-D270-40CE-8621-758AD30CE76C DTSTART;TZID=America/New_York:20240306T173000 DTEND;TZID=America/New_York:20240306T190000 DESCRIPTION:Resilience engineering is the ability to build and sustain a sy stem that can deal effectively with disruptive events. Previous resilience engineering research focuses on metrics to quantify resilience and models to characterize system performance. However\, resilience metrics are norm ally computed after disruptions have occurred and existing models lack the ability to predict one or more shocks and subsequent recoveries.\n\nTo ad dress these limitations\, this talk presents three alternative approaches to model system resilience with statistical techniques based on (i) regres sion\, (ii) time series\, and (iii) a combination of regression and time s eries to track and predict how system performance will change when exposed to multiple shocks and stresses of different intensity and duration\, pro vide structure for planning tests to assess system resilience against part icular shocks and stresses and guide data collection necessary to conduct tests effectively.\n\nThese modeling approaches are general and can be app lied to systems and processes in multiple domains. A historical data set o n job losses during the 1980 recessions in the United States is used to as sess the predictive accuracy of these approaches. Goodness-of-fit measures and confidence intervals are computed and interval-based and point-based resilience metrics are predicted to assess how well the models perform on the data set considered. The results suggest that resilience models based on statistical methods such as multiple linear regression and multivariate time series models are capable of modeling and predicting resilience curv es exhibiting multiple shocks and subsequent recoveries. However\, models that combine regression and time series account for changes in performance due to current and time-delayed effects from disruptions most effectively \, demonstrating superior performance in long-term predictions and higher goodness-of-fit despite increased parametric complexity.\n\nCo-sponsored b y: Boston Section Life Member Affinity Group\n\nSpeaker(s): Priscila Silva \, \n\nAgenda: \n5:30 PM Light repast and Networking\n\n6:00 PM Technical Presentation\n\n6:45 PM Questions and Answers\n\n7:00 PM Adjournment\n\nBl dg: Main Cafeteria\, Lincoln Laboratory\, 244 Wood St\, Lexington\, Massac husetts\, United States\, 02421\, Virtual: https://events.vtools.ieee.org/ m/406423 LOCATION:Bldg: Main Cafeteria\, Lincoln Laboratory\, 244 Wood St\, Lexingto n\, Massachusetts\, United States\, 02421\, Virtual: https://events.vtools .ieee.org/m/406423 ORGANIZER:james.yakura@ieee.org SEQUENCE:69 SUMMARY:Hybrid - Regression and Time Series Mixture Approaches to Predict R esilience URL;VALUE=URI:https://events.vtools.ieee.org/m/406423 X-ALT-DESC:Description: <br /><p align="justify">Resilience engineering is the ability to build and sustain a system that can deal effectively with d isruptive events. Previous resilience engineering research focuses on metr ics to quantify resilience and models to characterize system performance. However\, resilience metrics are normally computed after disruptions have occurred and existing models lack the ability to predict one or more shock s and subsequent recoveries.</p>\n<p align="justify">To address these limi tations\, this talk presents three alternative approaches to model system resilience with statistical techniques based on (i) regression\, (ii) time series\, and (iii) a combination of regression and time series to track a nd predict how system performance will change when exposed to multiple sho cks and stresses of different intensity and duration\, provide structure f or planning tests to assess system resilience against particular shocks an d stresses and guide data collection necessary to conduct tests effectivel y.</p>\n<p align="justify">These modeling approaches are general and can b e applied to systems and processes in multiple domains. A historical data set on job losses during the 1980 recessions in the United States is used to assess the predictive accuracy of these approaches. Goodness-of-fit mea sures and confidence intervals are computed and interval-based and point-b ased resilience metrics are predicted to assess how well the models perfor m on the data set considered. The results suggest that resilience models b ased on statistical methods such as multiple linear regression and multiva riate time series models are capable of modeling and predicting resilience curves exhibiting multiple shocks and subsequent recoveries. However\, mo dels that combine regression and time series account for changes in perfor mance due to current and time-delayed effects from disruptions most effect ively\, demonstrating superior performance in long-term predictions and hi gher goodness-of-fit despite increased parametric complexity.</p><br /><br />Agenda: <br /><p><strong>5:30 PM&nbsp\; &nbsp\; &nbsp\;</strong>Light r epast and Networking</p>\n<p><strong>6:00&nbsp\; PM</strong>&nbsp\; &nbsp\ ;Technical Presentation</p>\n<p><strong>6:45 PM</strong>&nbsp\; &nbsp\; Qu estions and Answers</p>\n<p><strong>7:00 PM</strong>&nbsp\; &nbsp\; Adjour nment</p> END:VEVENT END:VCALENDAR