An Improved PBFT Consensus Algorithm Based on a Comprehensive Efficiency Model

With the continuous development of blockchain technology, it plays an increasingly important role in providing secure and reliable data storage, enhancing data security and transparency, which makes it applicable in areas such as port data security. Among these, consensus algorithms are key to ensuring data consistency in blockchain networks. Currently, one of the most widely used consensus algorithms is Practical Byzan- tine Fault Tolerance (PBFT). However, PBFT faces challenges such as high communication complexity, low consensus efficiency, and performance bottlenecks. To address these issues, this paper proposes an improved PBFT consensus algorithm based on a comprehensive efficiency model (MBFT). First, the traditional single-leader node model is replaced by a multiagent node model, optimizing the consensus process, improving fault tolerance, and reducing communication complexity. Second, the comprehensive efficiency model is introduced, utilizing the TOPSIS algorithm to select efficient consensus nodes and the K-medoids algorithm to provide optimized sharding solutions for these nodes. Finally, theoretical analysis and experimental results demonstrate that the MBFT algorithm achieves substantial performance improve- ments.