Operations Research Models for Public Policy

I have three projects for which I am seeking student research assistants:

1. Network Analysis of Fake News Propagation on Social Media: We are embedding our previously developed probability model for the propagation of fake news on social media into a network propagation framework.  We seek to understand the conditions causing untruthful or highly biased news to “cascade” through the network, and the influence such content has on population beliefs.  The work will involve finalizing the existing code base to run simulations through the network, analyzing data, simulating the effect of policy interventions, and drawing conclusions.

2. International Pricing of Pediatric Vaccines: GAVI is a Global Vaccine Alliance that facilitates the procurement of pediatric vaccines on behalf of low income countries worldwide.  To do so, GAVI negotiates price and quantity contracts directly with vaccine manufacturers.  Currently, pentavalent combination vaccines are available that provide, in a single administration, vaccination against five antigens.  Under development is a hexavalent vaccine that would add a sixth antigen to the combination vaccine.  Doing so would dramatically reduce operational costs associated with distributing these vaccines and could dramatically improve vaccination rates in low income countries. This project seeks to refine our previously developed model to determine procurement prices and quantities to incentivize pharmaceutical companies to develop the hexavalent vaccine.   The work will involve developing a code base for the model, running computational analysis over a range of scenarios, analyzing data, simulating the effect of policy interventions, and drawing conclusions.

3. Covert Network Disruption through Vitality Maximization: Covert networks, such as terrorist networks or drug cartels, often attempt to divert communication away from critical leaders of the network to protect the leaders from detection.  The vitality maximization (VIMAX) problem seeks to identify a subset of vertices that can be removed from a network so that communication is forced to flow through these covert operatives, rendering them more identifiable.  We are now completing computational tests of several approaches to this problem.  The work will involve developing code bases for testing the existing solutions to VIMAX, as well as developing and testing a meta-heuristic approach for solving VIMAX. 

Interested students should indicate in their application essay which project(s) they would like to work on and any relevant coursework.  Programming proficiency is a necessary skill; background or interest in operations research preferred.  This is applied research, so students should expect a fair number of computational tasks.

Name of research group, project, or lab
Prof. Martonosi
Why join this research group or lab?

This is an opportunity to work on important problems in public policy through the lens of operations research.

Logistics Information:
Project categories
Operations Research
Student ranks applicable
Student qualifications

- Programming proficiency in Python

- Willingness to learn a new language for mathematical optimization

- Comfort in R for data analysis

- Familiarity with linear programming and mixed integer programming is useful for project (3) but not required.

- Past experience in graph theory is useful for project (3).

Time commitment
Spring - Part Time
Academic Credit
Number of openings
Techniques learned

- Sensitivity analysis

- Large scale computational testing for policy analysis

- Mathematical Programming formulations

Contact Information:
Mentor name
Susan Martonosi
Mentor email
Mentor position
Professor of Mathematics
Name of project director or principal investigator
Prof. Susan Martonosi
Email address of project director or principal investigator
3 sp. | 1 appl.
Hours per week
Spring - Part Time
Project categories
Operations Research (+1)
AlgorithmsOperations Research