Solving Large-Scale TSP Problems with Parallel Artificial Immune System Framework

For solving large-scale Traveling Salesman Problem (TSP) instances, we developed the Tower-like Master-Slave Model (TMSM) and its corresponding Parallel Immune Memory Clonal Selection Algorithm (PIMCSA). These methodologies significantly enhance TSP solution efficiency while maintaining excellent scalability. In the TMSM architecture, the master node handles task allocation and scheduling operations through distributed computing protocols, while slave nodes perform specific computational tasks using parallel processing techniques. The PIMCSA algorithm employs clone operations that duplicate high-affinity antibodies (solution candidates) and selection mechanisms that retain optimal solutions through fitness-based evaluation, thereby improving solution quality through iterative optimization cycles. Implementation typically involves population initialization functions, affinity calculation modules, and dynamic memory cell update procedures. These approaches have been validated in practical applications, demonstrating remarkable performance improvements. Future research will focus on algorithmic enhancements to address more complex and larger-scale TSP variants, potentially incorporating adaptive parameter tuning and hybrid optimization techniques.