“Statistical and Machine Learning Models for System Reliability and Resilience”
Please join the Boston IEEE Reliability Chapter for the following Technical Presentation on June 11, 2025!
Doors open at 5pm, with food and refreshments served at 5:30.
Location: This Meeting is to be delivered in-person at MIT Lincoln Lab Main Cafeteria, 244 Wood St, Lexington, MA 02421, and virtually. If attending in person, you must show a valid photo ID at the gate.
System reliability and resilience are crucial for ensuring dependable performance, especially in response to evolving demands and unexpected disruptions. Traditional reliability models, such as the Non-Homogeneous Poisson Process (NHPP), are widely used to predict defect occurrence based on testing time or effort. However, these models often fail to capture the complexities of real-world systems. Resilience engineering, which focuses on a system's ability to respond to and recover from shocks, has gained significant attention as a complementary approach to traditional reliability methods. Although statistical models provide foundational insights, their rigid assumptions can limit flexibility and fail to capture dynamic patterns in defect occurrence and recovery processes. Conversely, machine learning methods, such as neural networks, offer the potential to model intricate dependencies and non-linear trends. However, these models often require extensive data, which may not always be available in resilience engineering contexts, and they can lack robustness in long-term predictions. This limitation underscores the need for integrated approaches that effectively tackle the challenges of modeling resilience in systems experiencing various types and intensities of shocks.
To address these challenges, this talk explores hybrid approaches that enhance defect prediction in both regression and classification tasks and improve resilience assessment. We introduce flexible time series techniques that account for multiple stressors and recovery patterns. By integrating machine learning and statistical methods, this presentation aims to advance the assessment of both reliability and resilience in systems, providing robust, adaptable models capable of predicting defects and tracking recovery under complex conditions.
Date and Time
Location
Hosts
Registration
- Date: 11 Jun 2025
- Time: 09:00 PM UTC to 11:00 PM UTC
-
Add Event to Calendar
- 244 Wood Street
- Lexington, Massachusetts
- United States
- Building: Main Cafeteria
- Starts 20 May 2025 01:00 PM UTC
- Ends 11 June 2025 09:00 PM UTC
- No Admission Charge
Speakers
Fatemeh Salboukh
“Statistical and Machine Learning Models for System Reliability and Resilience”
Biography:
Fatemeh Salboukh is a PhD candidate working under the supervision of Professor Lance Fiondella at the University of Massachusetts at Dartmouth. Currently, she is completing an internship as a Data Scientist at Northwestern University. Her doctoral research focuses on "Statistical and Machine Learning Approaches for System Reliability and Resilience," with an anticipated defense and graduation date of May 2026.