Search

CN-121998537-A - Intelligent logistics path optimization method and system

CN121998537ACN 121998537 ACN121998537 ACN 121998537ACN-121998537-A

Abstract

The invention discloses an intelligent logistics path optimization method and system, wherein the method comprises the steps of collecting current logistics environment data in real time through a first intelligent sensing component arranged in an intelligent logistics path optimization system, preprocessing the environment data collected in real time, removing noise in the environment data through data cleaning, denoising and normalization, constructing a distribution path optimization model according to historical environment data and based on a deep neural network, predicting and analyzing the preprocessed environment data through the distribution path optimization model, predicting distribution conditions of future path demands, and dynamically adjusting a path planning strategy based on edge calculation nodes through a reinforcement learning algorithm. The method effectively reduces the time cost and the calculation resource consumption of logistics distribution, and simultaneously improves the instantaneity of path planning and the overall operation efficiency of the system.

Inventors

  • LI CANHONG

Assignees

  • 江苏网进科技股份有限公司

Dates

Publication Date
20260508
Application Date
20260202

Claims (10)

  1. 1. An intelligent logistics path optimization method, characterized in that the method comprises the following steps: Collecting current logistics environment data in real time through a first intelligent sensing assembly arranged in an intelligent logistics path optimization system; Preprocessing environment data acquired in real time, and removing noise in the environment data through data cleaning, denoising and normalization; Constructing a distribution path optimization model according to historical environment data and based on a deep neural network, and predicting and analyzing the preprocessed environment data through the distribution path optimization model to predict the distribution condition of future path demands; And dynamically adjusting a path planning strategy through a reinforcement learning algorithm based on the edge calculation node.
  2. 2. The intelligent logistics path optimization method of claim 1, further comprising: And storing the commonly used path optimization result and the corresponding environment characteristics on the edge server to construct a task multiplexing table.
  3. 3. The intelligent logistics path optimization method of claim 1, further comprising: the distribution path optimization model is continuously optimized based on the real-time feedback data.
  4. 4. The intelligent logistics path optimization method of claim 1 wherein the first intelligent awareness component comprises a plurality of sensors, GPS, and edge nodes.
  5. 5. The intelligent logistics path optimization method of claim 1, wherein the environmental data comprises geographic location information, cargo information, traffic information, and edge server information of the conveyance.
  6. 6. The intelligent logistics path optimization method of claim 1, wherein the step of removing noise from the data by data cleaning, denoising and normalization comprises: data cleaning is carried out on the data acquired in real time to remove invalid data and abnormal values, and cleaned data are obtained; Denoising the cleaned data to eliminate interference and obtain denoised data; and carrying out normalization processing on the denoised data to normalize the data to a preset range and obtain normalized data.
  7. 7. The intelligent logistics path optimization method of claim 1, wherein the step of constructing a distribution path optimization model based on the deep neural network from the historical environmental data, predicting and analyzing the preprocessed environmental data by the distribution path optimization model, and predicting the distribution situation of the future path demand comprises: collecting historical environment data and real-time environment data, extracting key features through a feature selection algorithm, and inputting the key features into a distribution path optimization model; the incremental learning strategy is adopted, parameters are updated in real time, and the prediction precision is improved; and predicting and analyzing the input historical environment data and real-time environment data by using the distribution path optimization model, and predicting the distribution condition of future path demands.
  8. 8. The intelligent logistics path optimization method of claim 1, wherein the step of constructing a distribution path optimization model based on the deep neural network from the historical environmental data, predicting and analyzing the preprocessed environmental data by the distribution path optimization model, and predicting the distribution situation of the future path demand comprises: collecting historical environment data and real-time environment data, extracting key features through a feature selection algorithm, and inputting the key features into a distribution path optimization model; the incremental learning strategy is adopted, parameters are updated in real time, and the prediction precision is improved; and predicting and analyzing the input historical environment data and real-time environment data by using the distribution path optimization model, and predicting the distribution condition of future path demands.
  9. 9. The adaptive production parameter optimization system of claim 1, wherein the system further comprises: and the coordination multiplexing module is used for storing a common path optimization result and corresponding environmental characteristics on the edge server and constructing a task multiplexing table.
  10. 10. The adaptive production parameter optimization system of claim 1, wherein the system further comprises: and the feedback module is used for continuously optimizing the distribution path optimization model based on the real-time feedback data.

Description

Intelligent logistics path optimization method and system Technical Field The invention relates to the technical field of logistics optimization, in particular to an intelligent logistics path optimization method and system based on edge calculation. Background With the rapid development of electronic commerce and modern logistics industry, logistics transportation demands are increasing dramatically. However, conventional logistics path planning often faces problems of limited computing resources, poor real-time performance, insufficient adaptability to dynamic traffic environments and the like. With the development of edge computing technology, computing capacity is deployed near the logistics nodes, so that the efficiency and instantaneity of logistics path optimization can be improved while cloud computing pressure is reduced. However, how to fully utilize the edge computing resources and reasonably plan the logistics path is still a technical problem to be solved. Disclosure of Invention Aiming at the technical problems, the invention provides an intelligent logistics path optimization method and system based on edge calculation, which can maximally reduce distribution time, oil consumption and logistics cost. The first embodiment of the invention provides a self-adaptive production parameter optimization method, which comprises the steps of collecting current logistics environment data in real time through a first intelligent sensing component arranged in an intelligent logistics path optimization system, preprocessing the environment data collected in real time, removing noise in the environment data through data cleaning, denoising and normalization, constructing a distribution path optimization model according to historical environment data and based on a deep neural network, predicting and analyzing the preprocessed environment data through the distribution path optimization model, predicting distribution conditions of future path demands, and dynamically adjusting a path planning strategy through a reinforcement learning algorithm based on edge calculation nodes. Optionally, the method further comprises the step of storing commonly used path optimization results and corresponding environment characteristics on the edge server and constructing a task multiplexing table. Optionally, the method further comprises continuously optimizing the delivery path optimization model based on the real-time feedback data. Optionally, the first smart sensor component includes a plurality of sensors, a GPS, and an edge node. Optionally, the environmental data includes geographic location information, cargo information, traffic information, and edge server information of the vehicle. Optionally, the step of removing noise in the data through data cleaning, denoising and normalizing comprises the steps of cleaning the data acquired in real time to remove invalid data and abnormal values to obtain cleaned data, denoising the cleaned data to eliminate interference to obtain denoised data, and normalizing the denoised data to normalize the data to a preset range to obtain normalized data. Optionally, the step of constructing a distribution path optimization model according to the historical environment data and based on the deep neural network, and predicting and analyzing the preprocessed environment data through the distribution path optimization model comprises the steps of collecting the historical environment data and the real-time environment data, extracting key features through a feature selection algorithm, inputting the key features into the distribution path optimization model, adopting an incremental learning strategy, updating parameters in real time, improving prediction accuracy, and predicting and analyzing the input historical environment data and real-time environment data by utilizing the distribution path optimization model to predict the distribution situation of the future path demands. The second embodiment of the invention provides a self-adaptive production parameter optimization system which comprises a data acquisition module, a data processing module, a path prediction module and a path optimization module, wherein the data acquisition module is used for acquiring current logistics environment data in real time through a first intelligent sensing component arranged in the system, the data processing module is used for preprocessing the environment data acquired in real time, removing noise in the environment data through data cleaning, denoising and normalization, the path prediction module is used for constructing a distribution path optimization model according to historical environment data and based on a deep neural network, predicting and analyzing the preprocessed environment data through the distribution path optimization model and predicting distribution conditions of future path demands, and the path optimization module is used for dynamically adjusting a path planning strategy thro