A Resilient RAP Solution Using Market-based Multi-Agent Systems

This paper presents a market-based multi-agent interpretation to solve the Redundancy Allocation Problem (RAP). Critical for designing reliable and cyber-resilient collaborative multiple subsystems, RAP is a NP-hard problem [3]. We aim to reinterpret RAP through a Multi-Agent System and solve using a Multiple Weighted Objective (MWO) approach. In this framework, individual agents aim to optimize individual subsystem objectives and negotiate with others to enhance the overall system reliability and cyber resilience. Our solution involves a three-tier process: local optimization using Linear Integer Programming (LIP), group optimization through agent negotiation, and iterative system enhancement. We have also shown using benchmark dataset, that our MWO approach delivers robust results, faster convergence with minimal variation and improved resilience against potential cyber threat.