path optimization, vertibi algorithm, logistics and distribution, genetic algorithm, vehicle path model with time window Abstract: The smart logistics system uses digital and information technology for efficient management of logistics operations like inventory and distribution. Guided by green low-carbon development, optimizing timewindow vehicle path models is core to smart logistics, effectively handling distribution settings. This paper focuses on multi-collection center logistics problems, devising a path optimization-based algorithm. It sets a penalty function for VRPTW, introduces and refines the Viterbi algorithm to generate candidate paths as the initial population for genetic algorithm selection and crossover via fitness and objective functions. After iterations, the optimal path is selected. Experimental results show the algorithm picks better paths with more iterations, with convergence speed improving by 26.7% and 16.3% vs control algorithms, demonstrating good optimization ability. https://mc.manuscriptcentral.com/jcmse Journal of Computational Methods in Science and Engineering For Peer Review Research on green logistics distribution network based on path optimization in the context of artificial intelligence Abstract: The smart logistics system uses digital and information technology for efficient management of logistics operations like inventory and distribution. Guided by green low-carbon development, optimizing time-window vehicle path models is core to smart logistics, effectively handling distribution settings. This paper focuses on multi-collection center logistics problems, devising a path optimization-based algorithm. It sets a penalty function for VRPTW, introduces and refines the Viterbi algorithm to generate candidate paths as the initial population for genetic algorithm selection and crossover via fitness and objective functions. After iterations, the optimal path is selected. Experimental results show the algorithm picks better paths with more iterations, with convergence speed improving by 26.7% and 16.3% vs control algorithms, demonstrating good optimization ability. Keywords: path optimization; vertibi algorithm; logistics and distribution; genetic algorithm; vehicle path model with time window 1.