KINGSTON, R.I. — July 31, 2018 — University of Rhode Island Assistant Professor of Engineering Gretchen Macht is using a $226,942 grant to conduct research into how to streamline the voting process in Rhode Island.
Titled RI VOTES (Voter OperaTions & Election Systems), the grant was funded by the Rhode Island Board of Elections, the Secretary of State’s office, the URI College of Engineering, and The Democracy Fund, an independent, private foundation.
Two years ago, significant delays in the voting process were reported at some locations, including East Providence, Jamestown, Pawtucket, Providence and Warren. In some cases, it took more than two hours for people to cast their ballot. According to the Presidential Commission on Elections Administration, a voter should not have to wait more than 30 minutes.
Secretary of State Nellie M. Gorbea convened a task force in February 2017 to investigate the cause of the delays, and reached out to URI’s College of Engineering for expertise.
A 14-page report released in April 2017 by the task force recommended that the Board of Elections work with Macht, and while she was honored and excited to be selected, she was well aware of the difficulty of the task.
“Every election is unique, making it difficult to plan for the next one,” Macht said. “There are many factors to consider, such as the growth in population, challenges in security, the limited space at some precincts and the possibility that a scanner could jam.”
Valerie Maier-Speredelozzi, associate professor of mechanical, industrial and systems engineering, also joined the effort by convincing Simio, LLC to donate a simulation software package for use on the RI Votes project. Valued at $132,000, the package includes 55 licenses and technical support. It enables Macht and her team to create a three-dimensional simulation of any voting precinct in Rhode Island by entering the number of check-in stations, voting booths, scanners and the physical layout of the precinct.
“By using voting data from the 2016 election precincts where everything ran smoothly, we are able to create a baseline trend and compare that to the locations where there were delays,” Macht said.
Multiple precincts in Rhode Island reported scanners that jammed, causing long delays.
“One of the things we’re looking into is whether the paper jams were a result of human error or machine error,” Macht said. “The thought is that the precincts that had especially long ballots, requiring more pages, experienced the most problems.”
Seniors and graduate students in Macht’s Introduction to Human Factors and Ergonomics course conducted studies in the fall 2017 semester on the time it took to check a voter into the polling location, time spent in a voting booth, and at the scanner.
Maier-Speredelozzi had students in her Facilities Planning and Material Handling course work with the Simio software in the spring 2018 semester, in association with Macht’s RI Votes research project.
“Each team of students was assigned a different voting precinct,” Maier-Speredelozzi explained. “They visited the polling location and assessed the space, layout, electricity, parking, and accessibility with respect to the Americans with Disabilities Act. Then they built models of the different precincts using the Simio software.”
Two of Macht’s students were original members of the team — Cherish Prickett of Lilburn, Georgia and James Houghton of Holtsville, New York. Prickett has since graduated, having completed URI’s International Engineering Program. Houghton graduated in the spring of 2017. After working in industry for almost nine months, he rejoined the team in spring 2018.
Nicholas Bernardo, of North Haven, Connecticut, joined the team his senior year. As an undergraduate, he worked on the team as part of an independent study project and now as a graduate student, he is working on the project for his thesis.
Bernardo and Houghton recently spent four days at a training at Simio’s headquarters in Sewickley, Pennsylvania.
The newest team member is undergraduate mechanical engineering student Ahmad Siddiqi, from Lincoln, Rhode Island. Siddiqi has completed several two- and three-dimensional sketches of the precincts and has worked on models using the Oculus Rift virtual reality system.
The RI VOTES project won’t be completed until June 2019, but the November 2018 elections will play a significant role in advancing the research.
“We’ll apply what we have at the time of the 2018 elections and use the results to improve and adapt the models,” said Bernardo. “This can act as a small-scale trial of the tools we have developed.”
With the initial task of collecting, entering and processing the 2016 election data complete, the research team is finishing the time-consuming task of making sure the data is accurate and interpreting the information.
The next step is to develop a simulation model that can be applied to any precinct in the state, regardless of population, the number of machines or workers, or the length of the ballot.
“We hope to reduce the maximum time a voter spends in a polling location in our simulated models, and then replicate these models in actual polling locations to improve the voter experience,” Bernardo said. “We really want to encourage people to vote by making the experience quick and easy, rather than a seemingly inconvenient obligation.”