Matteo Pozzi
Professor, Civil and Environmental Engineering, Carnegie Mellon University
Value of Information Analysis for Optimal Sensing in Infrastructure Systems
Team
Carl Malings, Matteo Pozzi, Milad Memarzadeh
Description
This project investigates how to evaluate the impact of information gathering (through sensors and monitoring systems, or inspections) in the management of spatially distributed infrastructure systems.
We propose to perform pre-posterior analysis of the network to optimize the design of sensing networks. A rational metric is the Value of Information, a decision theoretic measure which explicitly quantifies the benefit of sensor measurements in reducing the expected losses to a managing agent in the context of a decision-making problem under uncertainty.
We are developing a scalable probabilistic framework to perform pre-posterior analysis in large infrastructure systems using this metric. We apply this framework to the seismic monitoring of the SF bay area.
An example of outcome of the sensing network lay-out for the SF bay area, and the corresponding analysis of the Value of Information
We also use the Value of Information as a metric to define priorities among inspections, integrating scheduling with sequential maintenance.
An example of outcome sequential decision making, where the Value of Information is tracked in time: when it is sufficiently high, an inspection is performed.
Source of Funding
Start-up funds
Publications
Memarzadeh, M., Pozzi, M. "Value of Information in Sequential Decision Making: component inspection, permanent monitoring and system-level scheduling," Reliability Engineering & System Safety 154:137-151. (Elsevier) doi:10.1016/j.ress.2016.05.014 (2016).
Malings, C., Pozzi, M. "Value of Information for Spatially Distributed Systems: application to sensor placement," Reliability Engineering & System Safety 154:219-233. (Elsevier) doi:10.1016/j.ress.2016.05.010 (2016).
Malings, C., Pozzi, M. "Conditional entropy and value of information metrics for optimal sensing in infrastructure systems," Structural Safety 60:77-90. (Elsevier) doi:10.1016/j.strusafe.2015.10.003 (2016).
Memarzadeh, M., Pozzi, M. "Integrated inspection scheduling and maintenance planning for infrastructure systems," Computer-Aided Civil and Infrastructure Engineering 31(6):403-415. (Wiley) doi: 10.1111/mice.12178 (2016).
Malings, C., Pozzi, M. “Sensor network optimization using Bayesian networks, decision graphs, and value of information,” 12th International Conference on Applied Statistics and Probability in Civil Engineering, ICASP12, Vancouver, Canada, July 12-15, 2015.
Malings, C., Pozzi, M. “Sensor placement optimization for structural health monitoring,” 10th International Workshop on Structural Health Monitoring (IWSHM 2015), Sept. 1-3, 2015, Stanford University, CA.
Malings, C., Pozzi, M. “Conditional entropy and value of information based metrics for sensor placement in infrastructure systems,” The Sixth World Conference on Structural Control and Monitoring, Barcelona, 15-17 July 2014.
Malings, C., Memarzadeh, M., Pozzi, M. “Optimal topology of sensor networks for management of infrastructure systems,” proceedings of the 6th International Conference on Structural Health Monitoring of Intelligent Infrastructure Hong Kong, 9-11 December 2013.