Factory Craft
Developers: This project is a collaboration between 91制片廠合集 Guildhall (Video Games), Lyle School of Engineering, and Simmons (Education)
Students with reading disabilities (RDs) face barriers to STEM education due to limited access support services. Accommodations may be provided to students during testing, but that same support might not exist for day-to-day classroom instruction. To address this gap, HuMIn Game Lab is designing, developing, and embedding AI-driven support into the educational game, Factory Craft, using scalable AI agents. AI agents will personalize video game-based learning for students with RDs and help teachers monitor progress.
Factory Craft is an educational video game, previously funded by the National Science Foundation (NSF), designed to teach middle school students STEM concepts—such as ratios, supply chains, climate change, and material sciences—through the familiar and engaging platform of Minecraft. To make these concepts accessible and appealing to diverse students, Factory Craft integrates them into a larger, multi-layered narrative that connects STEM learning to real-world challenges and encourages exploration.
Future Impacts: We hope that this work will increase access to STEM education, addressing a critical need for inclusivity in learning environments. Our work contributes to advancing computing research and also aligns with the NSF's core strategy of supporting teaching initiatives, making computing and computational thinking a core fluency in education.
Figure 1: An image of Tiny Town from Factory Craft. In this game, the citizens of Tiny Town—including the player—take on the role of aiding other communities affected by natural disasters, such as asteroid strikes. Each building serves as an outpost where the player can accept tasks. In future iterations of the game, AI agents will personalize lessons by generating access supports and also by generating new gameplay tasks based on a student’s knowledge/mastery of a subject.
Figure 2: An image of a few residents of Tiny Town. These non-playable characters (NPCs) give students tasks and evaluate students' work (ensuring it is complete before moving to the next task). NPCs also quiz the students to see whether they understand the lessons. Designing these NPCs as AI agents will provide new opportunities for dynamic gameplay.
Figure 3: A student is interacting with a simple one-step factory line, which demonstrates how input materials are processed into outputs. The system begins with an input chest, which feeds materials into a processing block via a hopper. Nearby, additional one-step factory lines can be seen, which students will later combine into larger, more complex systems to automate material production.