SSRAA: A Swarm Intelligence Approach to Resource Allocation with Limited Knowledge
Is this project an undergraduate, graduate, or faculty project?
Graduate
Project Type
group
Campus
Daytona Beach
Authors' Class Standing
Fayruz Maysha, Graduate Student Julia Gorthey, Alumni
Lead Presenter's Name
Fayruz Maysha
Lead Presenter's College
DB College of Engineering
Faculty Mentor Name
Bryan Watson
Abstract
Effectively managing limited resources in uncertain, dynamic environments is a central challenge in systems engineering. This study introduces the Snapping Shrimp Resource Allocation Algorithm (SSRAA), a biologically inspired approach that adapts the defense mechanisms of snapping shrimp colonies to multi-agent systems with restricted situational awareness. Drawing on these natural cues, SSRAA promotes decentralized decision-making, enabling agents to allocate resources locally without requiring global information or centralized oversight. We evaluated SSRAA using an agent-based model of wildfire response, comparing it to three established methods: the Flood Approach (deploying all resources at once), an Auction-based strategy (assigning resources via bidding), and a Centralized Coordination model (leveraging comprehensive data). Simulation results across varied agent populations and sensing ranges reveal that SSRAA outperforms the Flood Approach in 86% of tested scenarios, saving on average 3% more forest area. Moreover, SSRAA’s efficiency closely matches that of both the Auction-based and centralized methods, highlighting its robustness and adaptability despite agents’ limited knowledge. By harnessing nature-inspired signaling thresholds, SSRAA selectively mobilizes resources, reducing redundancy and improving response times in high-stakes contexts. These findings suggest that biologically inspired algorithms offer valuable design principles for resource allocation in complex systems. Future research will refine SSRAA’s signaling parameters, explore scalability under more diverse emergency conditions, and investigate broader applications—such as humanitarian logistics and disaster relief—to further validate its potential for enhancing real-world resilience.
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?
Yes, SURF
SSRAA: A Swarm Intelligence Approach to Resource Allocation with Limited Knowledge
Effectively managing limited resources in uncertain, dynamic environments is a central challenge in systems engineering. This study introduces the Snapping Shrimp Resource Allocation Algorithm (SSRAA), a biologically inspired approach that adapts the defense mechanisms of snapping shrimp colonies to multi-agent systems with restricted situational awareness. Drawing on these natural cues, SSRAA promotes decentralized decision-making, enabling agents to allocate resources locally without requiring global information or centralized oversight. We evaluated SSRAA using an agent-based model of wildfire response, comparing it to three established methods: the Flood Approach (deploying all resources at once), an Auction-based strategy (assigning resources via bidding), and a Centralized Coordination model (leveraging comprehensive data). Simulation results across varied agent populations and sensing ranges reveal that SSRAA outperforms the Flood Approach in 86% of tested scenarios, saving on average 3% more forest area. Moreover, SSRAA’s efficiency closely matches that of both the Auction-based and centralized methods, highlighting its robustness and adaptability despite agents’ limited knowledge. By harnessing nature-inspired signaling thresholds, SSRAA selectively mobilizes resources, reducing redundancy and improving response times in high-stakes contexts. These findings suggest that biologically inspired algorithms offer valuable design principles for resource allocation in complex systems. Future research will refine SSRAA’s signaling parameters, explore scalability under more diverse emergency conditions, and investigate broader applications—such as humanitarian logistics and disaster relief—to further validate its potential for enhancing real-world resilience.