"The best-laid plans of mice and men often go awry," and we are all familiar with the subsequent chaos that follows, teams dedicated to firefighting in war rooms executing predefined plans that "best match" current circumstances.
Disruption Management seamlessly recommends the optimal solution to the precise disruption(s) occurring, factoring in all existing resources, dependencies, and requirements.
The co-founders of BIAS Intelligence dove deep into research in multi-agent consensus optimization working with members from the Department of Industrial & Systems Engineering and the Department of Mechanical Engineering at the University of Washington in Seattle. Once their joint research project completed (results published in IEEE Robotics and Automation Letters), BIAS developed the concept further for practical application in the Transportation sector.
The result is BIAS' Disruption Management module, leveraging asynchronous Consensus-Based Bundle Algorithm (CBBA) in a decentralized decision-making environment to assign tasks in near real-time to independent resources (humans, autonomous machines, or entire systems) to avoid expensive delays caused by unanticipated disruptions.
BIAS Disruption Management
Existing solutions often rely on heuristics, Linear Programs, and Mixed Integer Linear Programs which:
Don't account for communication constraints
Relying on centralized decision making
Requiring large computation times that don't scale exponentially
Don't account for "competition" between independent resources to secure tasks
Resilient to failures and uncertainties due to decentralized and consensus foundation
Added protocols allowing for the breakdown of tasks into smaller components, and elimination of tasks, when working with transferring tasks due to disruptions (example: if a drone goes offline, but a human in an aircraft is available to complete a task, there would likely be different execution)