CN-122022627-A - Logistics transportation path assessment method, system, electronic equipment and storage medium based on map analysis

CN122022627ACN 122022627 ACN122022627 ACN 122022627ACN-122022627-A

Abstract

The invention belongs to the technical field of logistics transportation evaluation, and provides a logistics transportation path evaluation method, a system, electronic equipment and a storage medium based on map analysis, wherein the method comprises the steps of fusion data set construction, map construction, effective road network map construction, effective path information extraction, local prediction map sequence construction, multidimensional evaluation and grade matching; the method improves the prospective of the evaluation result by constructing the dynamic road network map, the historical road network map library and the multi-time step prediction, realizes the comprehensive evaluation of the path passing efficiency and the risk by constructing the local road network set, improves the integrity of the evaluation result, and enhances the reliability and the robustness of the evaluation result by integrating dynamic influence data and correcting by utilizing a multi-rule multi-dimensional processing mechanism.

Inventors

WANG YUE
ZHANG GAOPAN
HAN SHAOWU
ZHANG YANKUN
Yi Tongzheng
ZHU YISHUN

Assignees

山东省新联世纪信息科技有限责任公司

Dates

Publication Date: 20260512
Application Date: 20260130

Claims (10)

1. A method for assessing a physical distribution transportation path based on graph analysis, comprising: Constructing a multi-source data set, and carrying out standardization, abnormal value detection and repair treatment on the multi-source data set to obtain a fusion data set, wherein the multi-source data set comprises road network basic data, time sequence road network data, dynamic influence data, risk related data and local associated data; discretizing the road network state into a plurality of time steps according to the time dimension, and constructing a dynamic road network map and a historical road network map library corresponding to each time step according to the fusion data set; Performing state judgment and attribute correction on nodes and edges of the dynamic road network map according to a preset multi-rule multi-dimensional processing mechanism to obtain an effective road network map; carrying out path analysis and validity verification on the logistics transportation path to be evaluated to obtain effective path information; Setting a perception field, extracting a local road network from the effective road network map according to the perception field and the effective path information to obtain a local road network set, and preprocessing and predicting historical data corresponding to the local road network set in the historical road network map library to obtain a local prediction map sequence; Calculating a pre-constructed multidimensional quantitative evaluation system according to the effective road network map, the historical road network map library and the local prediction map sequence to obtain a comprehensive evaluation value; and performing level matching on the comprehensive evaluation value to obtain a target path evaluation level.
2. The method for evaluating a physical distribution transportation path based on graph analysis according to claim 1, wherein constructing a multi-source data set, and performing standardization, outlier detection and repair processing on the multi-source data set to obtain a fusion data set, comprises: collecting geographic position, connection relation, physical attribute and administrative region division of a target region to obtain road network basic data; collecting current data and historical data of the target area to obtain the time sequence road network data; Collecting current weather data, emergency data, policy restriction data and logistics business data of the target area to obtain the dynamic influence data; Collecting historical accident data, environment risk data and risk data of target goods of the target area to obtain risk related data; Collecting road network topological relation and state influence transfer rules of the target area to obtain the local association data; And performing outlier detection and outlier correction on the road network basic data, the time sequence road network data, the dynamic influence data, the risk related data and the local associated data by using an isolated forest algorithm and a time sequence interpolation method to obtain the fusion data set.
3. The method for evaluating a physical distribution transportation path based on graph analysis according to claim 1, wherein the step of discretizing the road network state into a plurality of time steps according to time dimension, and constructing a dynamic road network graph and a historical road network graph library corresponding to each time step according to the fusion data set comprises the steps of: constructing a dynamic road network map at each time step according to the fusion data set, wherein the expression of the dynamic road network map is as follows: Wherein, the method comprises the steps of, The dynamic road network map at the time t; 、 Respectively a node set and an edge set at the moment t; Embedding an attribute system in a node set and an edge set of the dynamic road network map, wherein the attribute system comprises a basic attribute, a state attribute, a performance attribute, a risk attribute and a time sequence attribute; and integrating the dynamic road network map before the current time step to obtain the historical road network map library.
4. The method for evaluating a physical distribution transportation path based on graph analysis according to claim 1, wherein the method for evaluating the physical distribution transportation path based on graph analysis is characterized by performing state judgment and attribute correction on nodes and edges of the dynamic road network graph according to a preset multi-rule multi-dimensional processing mechanism to obtain an effective road network graph, and comprises the following steps: Setting a node state processing rule and an edge state processing rule, wherein the node state processing rule comprises a traffic prohibition rule, a partial closing or temporary control rule and a risk correction rule, and the edge state processing rule comprises a traffic prohibition rule, a bearing capacity grading rule, a congestion state dynamic correction rule, a multi-factor influence fusion correction rule and a time sequence trend correction rule; and processing the nodes and edges of the dynamic road network map by using the node state processing rules and the edge state processing rules to obtain the effective road network map.
5. The method for evaluating a physical distribution transportation path based on graph analysis according to claim 1, wherein the steps of performing path analysis and validity verification on the physical distribution transportation path to be evaluated to obtain valid path information include: Carrying out path analysis on the logistics transportation path to be evaluated to obtain path basic information, wherein the path basic information comprises a node sequence, an edge sequence and transportation demand information, and the transportation demand information comprises a cargo type, a load, an aging requirement, a cost budget and a planned driving period; performing node and edge validity check, bearing suitability check and aging feasibility check on the path basic information to obtain a check result; and setting the path basic information which is checked to pass by the checking result as the effective path information.
6. The method for evaluating a physical distribution transportation path based on graph analysis according to claim 1, wherein a perception domain is set, local road network extraction is performed on the effective road network graph according to the perception domain and the effective path information to obtain a local road network set, and the historical data corresponding to the local road network set in the historical road network graph library is preprocessed and predicted to obtain a local prediction graph sequence, and the method comprises the steps of: extracting nodes and edges with the distances between nodes and edges in the effective road network map and the effective path information smaller than those in the perception field to obtain the local road network set; extracting historical data corresponding to the local road network set in the historical road network map library to obtain a historical data set; Performing feature extraction and normalization processing on the historical data set to obtain historical feature data; and inputting the historical characteristic data into a pre-trained ST-GCN model to perform multi-time-step prediction, and obtaining the local prediction spectrum sequence.
7. The method for evaluating a transportation path of a stream based on graph analysis according to claim 1, wherein the expression of the multi-dimensional quantitative evaluation system is: Wherein, the method comprises the steps of, ; The comprehensive evaluation value is obtained; Is an initial evaluation value; Is a weight correction threshold; outputting a random forest model; To evaluate the total number of dimensions; Is a single-dimension weight; representing an approximation ideal solution ordering method; representing the index vector in dimension d.
8. A system for assessing a physical distribution transportation path based on analysis of a map, for implementing the method for assessing a physical distribution transportation path based on analysis of a map of claim 1, the system comprising: The data acquisition module is used for constructing a multi-source data set, and carrying out standardization, abnormal value detection and restoration treatment on the multi-source data set to obtain a fusion data set, wherein the multi-source data set comprises road network basic data, time sequence road network data, dynamic influence data, risk related data and local associated data; The map construction module is used for discretizing the road network state into a plurality of time steps according to the time dimension, and constructing a dynamic road network map and a historical road network map library corresponding to each time step according to the fusion data set; The map correcting module is used for carrying out state judgment and attribute correction on the nodes and edges of the dynamic road network map according to a preset multi-rule multi-dimensional processing mechanism to obtain an effective road network map; The path information extraction module is used for carrying out path analysis and validity verification on the logistics transportation path to be evaluated to obtain effective path information; The information prediction module is used for setting a perception field, extracting a local road network from the effective road network map according to the perception field and the effective path information to obtain a local road network set, and preprocessing and predicting historical data in the historical road network map library and corresponding to the local road network set to obtain a local prediction map sequence; The multidimensional evaluation module is used for calculating a pre-constructed multidimensional quantitative evaluation system according to the effective road network map, the historical road network map library and the local prediction map sequence to obtain a comprehensive evaluation value; And the grade matching module is used for carrying out grade matching on the comprehensive evaluation value to obtain a target path evaluation grade.
9. An electronic device comprising at least one processor and a memory communicatively coupled to the processor, wherein the memory stores instructions executable by the processor to enable the processor to perform a method of assessing a transportation path of a stream based on profile analysis as claimed in any one of claims 1 to 7.
10. A non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform a method of assessing a transportation path of a stream based on a graph analysis as claimed in any one of claims 1 to 7.

Description

Logistics transportation path assessment method, system, electronic equipment and storage medium based on map analysis Technical Field The invention relates to the technical field of logistics transportation evaluation, in particular to a logistics transportation path evaluation method, system, electronic equipment and storage medium based on map analysis. Background The logistics transportation path evaluation is a core link of logistics supply chain optimization, and is directly related to transportation cost, timeliness, safety and customer satisfaction. The existing path evaluation technology has the key technical defects that 1) the road network modeling is lack of perspective, the existing method is mainly based on the current road network state, future state changes of road network nodes/edges in the path driving process, such as congestion aggravation, temporary sealing and weather mutation in the future period, are not fully considered, and the evaluation result only reflects the current situation, so that potential risks and efficiency losses in the path execution process cannot be predicted. 2) The local road gateway joint analysis is insufficient, namely the passing efficiency and risk of the path not only depend on the node/edge passing by the path but also are influenced by the states of peripheral associated road networks, such as parallel road sections and intersection junctions, a targeted local road network sensing and analyzing mechanism is not constructed in the prior art, and the indirect influence of the peripheral road networks of the path on the target path is difficult to comprehensively capture. 3) The prediction and the evaluation are not fused deeply, wherein part of the prior art introduces simple time sequence prediction, but a special prediction module is not designed based on a specific node/edge set of a path to be evaluated, the prediction result is not targeted, meanwhile, the fusion of prediction data and an evaluation index system is not tight enough, and future state data is not comprehensively incorporated into quantitative evaluation of cost, aging, risk and other dimensions. 4) The dynamic adaptability and the robustness are insufficient, namely, the accurate prediction of the dynamic change of the road network state in the path execution period is lacking, the potential risk of the path cannot be identified in advance, such as the congestion aggravation and the road section closure in the future period, and the adaptability of the path to the emergency is not considered, so that the estimation result is lack of foresight and reliability. 5) The historical path timing sequence rule mining is insufficient, the utilization of the historical data in the prior art is limited to the current state correction, the special historical timing sequence features are not extracted for the local road network of the target path, the periodicity and trend change rule of the local road network are difficult to accurately capture, and the prediction and evaluation accuracy is affected. Disclosure of Invention In order to overcome the defects of the prior art, the invention aims to provide a logistics transportation path evaluation method, a system, electronic equipment and a storage medium based on map analysis, which solve the problems of lack of prospective, insufficient local road gateway joint analysis, and insufficient dynamic adaptability and robustness of the existing method. In order to achieve the above object, the present invention provides the following solutions: a method for assessing a logistics transportation path based on graph analysis, comprising: Constructing a multi-source data set, and carrying out standardization, abnormal value detection and repair treatment on the multi-source data set to obtain a fusion data set, wherein the multi-source data set comprises road network basic data, time sequence road network data, dynamic influence data, risk related data and local associated data; discretizing the road network state into a plurality of time steps according to the time dimension, and constructing a dynamic road network map and a historical road network map library corresponding to each time step according to the fusion data set; Performing state judgment and attribute correction on nodes and edges of the dynamic road network map according to a preset multi-rule multi-dimensional processing mechanism to obtain an effective road network map; carrying out path analysis and validity verification on the logistics transportation path to be evaluated to obtain effective path information; Setting a perception field, extracting a local road network from the effective road network map according to the perception field and the effective path information to obtain a local road network set, and preprocessing and predicting historical data corresponding to the local road network set in the historical road network map library to obtain a local prediction map sequence; Calcula