Is this project an undergraduate, graduate, or faculty project?
Undergraduate
Project Type
group
Campus
Daytona Beach
Authors' Class Standing
Rachel Phillips, Senior Abigail Threatt Parneet Makkar Jasmine Cruz Matthew Skowronek
Lead Presenter's Name
Rachel Phillips
Lead Presenter's College
DB College of Arts and Sciences
Faculty Mentor Name
Barbara Chaparro
Abstract
In naval damage control scenarios, operators in Damage Control Central (DCC) must interpret information from numerous sensors at once to detect hazards such as fire, flooding, or smoke while also maintaining ship stability. Designing interfaces that effectively support these tasks typically requires significant experience in human factors and military design standards. This project investigates whether AI-based design tools can help novice users produce functional interface prototypes more efficiently. Participants were randomly divided into two groups: an experimental group that used an assigned AI tool (Figma Make, Visily, or Claude) to generate and refine interface layouts, and a control group that completed the same task manually using a platform of their choice. Participants in the manual group used Figma or sketched their designs on paper. Both groups could reference the MIL-STD-1472H military standards and a 3D ship view example. Participants were asked to design a “status-at-a-glance” dashboard that included a General Arrangement Plan (GAP), color-coded hazard overlays for fire, flooding, and smoke with live sensor data, automated emergency “Kill Cards”, and a real-time ship stability indicator. Measures included adherence to design standards, usability of the tool evaluated by the System Usability Scale (SUS), perceived workload measured with the NASA Task Load Index (NASA-TLX), as well as qualitative assessments of the interface’s aesthetic appearance and degree of interactivity. Preliminary observations suggest that participants using AI produced more detailed and interactive prototypes, while manually created designs were generally more focused on static layouts. These results may help clarify how AI tools could support early interface design and assist novice designers working on complex operational systems.
Did this research project receive funding support (Spark, SURF, Research Abroad, Student Internal Grants, Collaborative, Climbing, or Ignite Grants) from the Office of Undergraduate Research?
No
Shipboard Fire & Flooding: Evaluating AI-Assisted Interface Design for Novice Designers
In naval damage control scenarios, operators in Damage Control Central (DCC) must interpret information from numerous sensors at once to detect hazards such as fire, flooding, or smoke while also maintaining ship stability. Designing interfaces that effectively support these tasks typically requires significant experience in human factors and military design standards. This project investigates whether AI-based design tools can help novice users produce functional interface prototypes more efficiently. Participants were randomly divided into two groups: an experimental group that used an assigned AI tool (Figma Make, Visily, or Claude) to generate and refine interface layouts, and a control group that completed the same task manually using a platform of their choice. Participants in the manual group used Figma or sketched their designs on paper. Both groups could reference the MIL-STD-1472H military standards and a 3D ship view example. Participants were asked to design a “status-at-a-glance” dashboard that included a General Arrangement Plan (GAP), color-coded hazard overlays for fire, flooding, and smoke with live sensor data, automated emergency “Kill Cards”, and a real-time ship stability indicator. Measures included adherence to design standards, usability of the tool evaluated by the System Usability Scale (SUS), perceived workload measured with the NASA Task Load Index (NASA-TLX), as well as qualitative assessments of the interface’s aesthetic appearance and degree of interactivity. Preliminary observations suggest that participants using AI produced more detailed and interactive prototypes, while manually created designs were generally more focused on static layouts. These results may help clarify how AI tools could support early interface design and assist novice designers working on complex operational systems.