Logistics management system based on artificial intelligence

Abstract

The invention discloses a logistics management system based on artificial intelligence, and the system is used for improving the efficiency and intelligent level of logistics scheduling and resource configuration, and comprises the following steps: building a unified data set through collecting and positioning, sensor and operation log and other types of logistics data; a modeling method is used for fusion to form a data structure containing time-space and transportation state information, the system adopts a causal analysis model to identify key influence relations between logistics events, a cross-modal feature extraction technology is combined to generate representation vectors such as transportation tracks, loads and states, and the logistics events are subjected to self-adaptive analysis through a self-adaptive strategy. And the causal information and the multi-modal features are fused to generate a decision basis, and the fusion proportion is dynamically adjusted according to actual operation feedback, so that transportation path optimization, distribution task planning and reasonable resource allocation are realized. The system has a continuous learning capability, and can keep the accuracy and flexibility of scheduling decisions in a continuously updated data environment.

Description

Technical Field

[0001] The present invention relates to the technical field of logistics management, and in particular to a logistics management system based on artificial intelligence. Background Art

[0002] With the continuous development of artificial intelligence and information technology, modern logistics systems are facing unprecedented challenges in data complexity and real-time decision-making. Especially in the context of high-frequency flow of multi-source heterogeneous data, traditional logistics management systems have exposed obvious limitations in scheduling response, path optimization and resource allocation. Events in the logistics process are highly spatiotemporal dynamic and multimodal interactive. Single information processing strategies and static rules can no longer meet the intelligent decision-making needs of complex operation scenarios.

[0003] In the existing technology, traditional logistics management solutions generally adopt rule-based scheduling systems and static optimization algorithms. Such methods have the following major defects when facing the rapidly changing transportation environment and multi-source data fusion processing: 1. Lack of causal modeling capabilities: Existing methods mostly make scheduling decisions based on correlation analysis, which makes it difficult to identify the true causal relationship between logistics events, resulting in structural misjudgments in resource scheduling and path selection.

[0004] 2. Insufficient understanding of multimodal data: Traditional systems find it difficult to effectively process heterogeneous data from GPS, sensors, and operation logs at the same time. They rely on a large amount of manual annotation and lack automatic semantic extraction mechanisms.

[0005] 3. The fusion mechanism is rigid and not adaptive: Most methods use fixed weights or static fusion methods, and cannot dynamically adjust the weight ratio between causal information and feature expression based on actual operation feedback.

[0006] Therefore, how to provide a logistics management system based on artificial intelligence is an urgent problem that technical personnel in this field need to solve. Summary of the invention

[0007] One object of the present invention is to propose a logistics management system based on artificial intelligence. The present invention has the advantages of fast response speed, high scheduling accuracy and strong adaptability through modeling of causal relationships between logistics events, automatic extraction and dynamic fusion of cross-modal feature vectors, and combined with a real-time feedback mechanism for joint decision generation.

[0008] An artificial intelligence-based logistics management system according to an embodiment of the present invention includes the following steps: S1, data acquisition module, used to collect sensor, and multi-source logistics data of operation logs, perform standardized preprocessing on the collected multi-source logistics data, and construct a logistics data set; S2, data fusion module, used to align and fuse logistics data sets in time and space, and generate fused data based on the spatiotemporal dynamic logistics network modeling method, including logistics event characteristics, transportation status characteristics, resource usage characteristics and current network status data; S3, causal reasoning module, based on deep causal reasoning algorithm, uses multi-layer neural network to quantitatively model the causal relationship between logistics events in the fusion data and output causal reasoning parameters; S4, self-supervised multimodal representation learning module, constructs cross-modal self-supervised pre-training tasks, and automatically extracts feature vectors of fused data through cross-modal feature interaction mechanisms, including spatiotemporal trajectory features, vehicle load features, and transportation status features; S5, the fusion decision module, adopts an adaptive dynamic decision weight adjustment strategy, adaptively integrates causal reasoning parameters and feature vectors according to real-time online feedback, and generates logistics scheduling and resource allocation decisions, including the optimal transportation route, dynamic distribution plan and resource allocation plan.

[0009] Optionally, the S3 specifically includes: S31, event feature preprocessing unit, receives fused data, standardizes the original features of each logistics event in the fused data, extracts the event occurrence time, geographic location, status information and numerical indicators, and forms an event feature vector matrix ; S32, multi-layer neural network mapping unit, based on deep causal reasoning algorithm, event feature vector matrix A multi-layer neural network is used for layer-by-layer nonlinear mapping. Each hidden layer outputs abstract feature representation through a combination of linear transformation and activation function. The nonlinear mapping function is assumed to be , get the abstract feature matrix; S33, causal impact calculation unit, builds a quantitative model of causal relationships between events based on the abstract feature matrix, and calculates logistics events using the following formula Logistics events The causal influence weight of is: ; in, For logistics events Logistics events The causal influence weight of Logistics incident Logistics events The nonlinear mapping function composed of the abstract feature matrix, is the weight matrix, is the bias vector, is the activation function; S34, causal reasoning parameter generation unit, quantifies the causal relationship between each logistics event according to the causal influence weight, and generates causal reasoning parameters ,The causal reasoning parameters include event impact weights, resource scheduling influencing factors, and transportation path adjustment factors; S35, model parameter optimization unit, uses the back propagation algorithm to optimize the weight matrix of the quantitative model of causal relationship between logistics events With the bias vector Dynamic adjustments are made to ensure that the causal reasoning module can stably and accurately output the above causal reasoning parameters in different logistics scenarios.

[0010] Optionally, the S32 specifically includes: S321, input preprocessing unit, used to receive the output event feature vector matrix , for the event feature vector matrix The data in are normalized to form a normalized feature matrix ; S322, layer-by-layer mapping unit, based on deep causal reasoning algorithm, builds A multi-layer neural network with hidden layers, each hidden layer performs a linear transformation and activation function combination operation, and normalizes the feature matrix Perform layer-by-layer mapping and define the initial input as , for the layer , calculated using the following formula: ; in, Indicates The hidden representation of the layer, Indicates The weight matrix of the layer, Indicates The bias vector of the layer, is a nonlinear activation function, Indicates The hidden representation of the layer is mapped layer by layer to obtain the final hidden representation ; S323, feature abstraction unit, assuming that the nonlinear mapping function is , the final hidden representation of the output of the layer-by-layer mapping unit Perform nonlinear transformation to generate an abstract feature matrix.

[0011] Optionally, the S4 specifically includes: S41, a modality data construction unit, used to receive the generated fusion data, and divide the fusion data into a plurality of modality subsets, wherein the modality subsets include sensor modality, GPS modality and log modality, which correspond to environmental perception data, trajectory positioning data and operation scheduling data, respectively, to form a modality input set after division; S42, a modal feature encoding unit, based on the modal input set, constructs a corresponding feature encoding network for each modal subset, the feature encoding network includes several hidden layers, each layer contains trainable weight parameters and bias parameters, extracts its primary representation features, and sets The modal data is , which is coded as: ; in, Representing modality The encoding function, is the modal eigenvector; S43, cross-modal contrastive learning unit, for generating modal feature vectors based on Construct a cross-modal self-supervised learning task. By constructing positive and negative sample pairs and introducing a contrast loss function, the feature vector space aggregation of the same logistics event in different modes and the feature vector space separation of different logistics events are achieved to generate an optimized modal feature vector. S44, feature interaction fusion unit, is used to optimize the modal feature vector for feature alignment and fusion, use the cross-attention mechanism to calculate the correlation score between the modalities, and combine the fusion strategy to generate the feature vector , the feature vector Including spatiotemporal trajectory characteristics, vehicle load characteristics and transportation status characteristics.

[0012] Optionally, the S43 specifically includes: S431, sample pair construction unit, used to receive modal feature vector , construct positive sample pairs for the same logistics event between different modes, and construct negative sample pairs between different logistics events to form a sample pair set , where each Represents pairs of feature vectors from different modalities, where positive sample pairs correspond to the same logistics event and negative sample pairs correspond to different logistics events; S432, contrast loss function definition unit, used to define the loss function based on the sample pair set Construct a cross-modal self-supervised contrastive learning objective and define the contrastive loss function as follows: ; in, is the contrast loss function, is the similarity function between feature vectors, is the temperature coefficient, is the anchor feature vector Constitute the set of negative sample pairs, is the total number of sample pairs, is the interference feature vector and the anchor feature vector Constitute a negative sample pair, To match the feature vector with the anchor feature vector Constitute a positive sample pair; S433, a feature representation optimization unit, used to adjust the trainable weight parameters and bias parameters of the feature encoding network according to the defined contrast loss function, so that the feature vectors in the positive sample pairs converge in distance in the feature vector space, and the feature vectors in the negative sample pairs separate in distance in the feature vector space, so as to form a distinguishable cross-modal embedding representation structure; S434, optimized representation output unit, used to embed the distinguishable cross-modal representation structure to generate the optimized modal feature vector Output to the feature interaction fusion unit.

[0013] Optionally, the S5 specifically includes: S51, feature integration preparation unit, used to receive causal reasoning parameters and the eigenvector , and the causal inference parameters and the eigenvector Perform format conversion and dimension unification to construct an input alignment vector set ; S52, an adaptive weight generating unit, for aligning a vector set based on an input Establish a fusion strategy to perform weighted integration of causal inference parameters and feature vectors to construct a joint representation vector , the following formula is used for fusion: ; in, is an adjustable fusion weight coefficient, which is used to control the contribution ratio of causal inference parameters and feature vectors; S53, feedback control unit, builds a logistics execution system and generates real-time feedback information, which includes task completion time, resource usage status and execution deviation indicators, builds a weight update function based on the real-time feedback information, and dynamically adjusts the adjustable fusion weight coefficient , to adapt to the current operation status of the logistics system; S54, decision generation unit, used to receive the joint representation vector , and combines the current network status data for strategic reasoning to generate logistics scheduling and resource allocation decisions including optimal transportation routes, dynamic distribution plans and resource allocation plans.

[0014] Optionally, the S52 specifically includes: S521, feature similarity calculation unit, used to receive the output input alignment vector set , and calculate the feature representation similarity of the input alignment vector set, the feature representation similarity uses the normalized cosine similarity function: ; in, represents the feature representation similarity between the causal inference parameters and the fused feature vector, represents the vector dot product operation, Represents the L2 norm of the vector; S522, weight coefficient initialization unit, used to represent similarity based on the calculated feature Determine the adjustable fusion weight coefficient , through the monotone mapping function Calculate, the For the interval The monotone mapping function defined in , used to control the initial fusion ratio between causal information and fusion features; S523, fusion representation generation unit, used to receive adjustable fusion weight coefficient , input alignment vector set , and construct the joint representation vector .

[0015] The beneficial effects of the present invention are: (1) The present invention constructs a logistics management system that integrates causal reasoning, self-supervised multimodal representation learning and adaptive fusion strategy, thereby achieving modeling of causal relationships between logistics events, efficient expression of multi-source heterogeneous data and intelligent decision-making of scheduling strategies. The system can identify key logistics influencing factors and dynamically generate transportation routes, distribution plans and resource allocation plans, thereby effectively solving the problems of delayed response and low scheduling accuracy of traditional logistics systems in dynamic environments.

[0016] (2) The present invention introduces a feedback perception mechanism and an online parameter updating strategy to construct a weight control model driven by operational feedback, so that the fusion decision results can adapt to the current logistics network status in real time, improve the scheduling flexibility and stability of the system in different transportation scenarios, and have the ability of continuous learning and self-optimization, effectively improving resource utilization efficiency and scheduling response speed.

[0017] (3) The present invention adaptively constructs a joint representation vector from causal reasoning parameters and multimodal features, and combines it with online feedback to form a closed-loop decision-making system. This enables the system to automatically adapt to the logistics operation characteristics under different scenarios without relying on a large number of manual rules or annotations, providing more intelligent, accurate and dynamic logistics management capabilities, and significantly enhancing the system's decision-making capabilities and operating efficiency in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 This is a schematic diagram of the overall architecture of an artificial intelligence-based logistics management system proposed by the present invention. DETAILED DESCRIPTION

[0019] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0020] refer to Figure 1 , a logistics management system based on artificial intelligence, comprising the following steps: S1, data acquisition module, used to collect sensor, and multi-source logistics data of operation logs, perform standardized preprocessing on the collected multi-source logistics data, and construct a logistics data set; This implementation method collects multi-source logistics data from sensors, GPS and operation logs in real time by deploying data collection modules. Multi-source logistics data includes environmental perception data, trajectory positioning data and operation scheduling data. Environmental perception data includes the temperature in the car collected by the on-board temperature and humidity sensor, the impact intensity recorded by the vibration sensor, the light sensor to detect the opening state of the car, the air quality sensor to detect the gas concentration in the car, and the speed sensor to record the instant speed. The trajectory positioning data includes GPS real-time positioning coordinates, driving trajectory point set, stop event record, driving state identification, route deviation alarm data, and operation scheduling data including transportation task list, vehicle allocation record, scheduling plan schedule, distribution priority, and real-time task status; and unified standardized preprocessing, including time alignment, outlier removal and feature normalization, so as to construct a logistics data set with unified structure and consistent semantics. As the information entrance of the system, this module ensures the input quality of causal modeling and multimodal learning, so that the subsequent analysis model can accurately extract spatiotemporal, load and operation characteristics based on high-quality data, realize comprehensive perception of key behaviors and resource status in logistics scenarios, and improve the adaptability of the system in complex environments and the reliability of decision-making basis.

[0021] S2, data fusion module, used to align and fuse logistics data sets in time and space, and generate fused data based on the spatiotemporal dynamic logistics network modeling method, including logistics event characteristics, transportation status characteristics, resource usage characteristics and current network status data; This implementation adopts a spatiotemporal dynamic logistics network modeling method to perform spatiotemporal alignment and fusion processing on the collected logistics data sets, uniformly encode heterogeneous information such as GPS trajectories, sensor status, and operation logs according to timestamps and geographic locations, and model the relationship between nodes, edges, and resource flows in the network graph structure, and finally generate fused data, including logistics event characteristics, transportation status characteristics, resource usage characteristics, and current network status data. This processing method realizes the organic integration of multi-source logistics information in a unified spatial and temporal dimension, provides structured and high-density input data support for subsequent causal reasoning and scheduling decisions, and effectively enhances the system's ability to understand the global state of logistics and respond to real-time changes.

[0022] S3, causal reasoning module, based on deep causal reasoning algorithm, uses multi-layer neural network to quantitatively model the causal relationship between logistics events in the fusion data and output causal reasoning parameters; S4, self-supervised multimodal representation learning module, constructs cross-modal self-supervised pre-training tasks, and automatically extracts feature vectors of fused data through cross-modal feature interaction mechanism. The feature vectors include spatiotemporal trajectory features, vehicle load features, and transportation status features. S5, the fusion decision module, adopts an adaptive dynamic decision weight adjustment strategy, adaptively integrates causal reasoning parameters and feature vectors according to real-time online feedback, and generates logistics scheduling and resource allocation decisions, including the optimal transportation route, dynamic distribution plan and resource allocation plan.

[0023] In this implementation manner, S3 specifically includes: S31, event feature preprocessing unit, receives fused data, standardizes the original features of each logistics event in the fused data, extracts the event occurrence time, geographic location, status information and numerical indicators, and forms an event feature vector matrix ; S32, multi-layer neural network mapping unit, based on deep causal reasoning algorithm, event feature vector matrix A multi-layer neural network is used for layer-by-layer nonlinear mapping. Each hidden layer outputs abstract feature representation through a combination of linear transformation and activation function. The nonlinear mapping function is assumed to be , get the abstract feature matrix; S33, causal impact calculation unit, builds a quantitative model of causal relationships between events based on the abstract feature matrix, and calculates logistics events using the following formula Logistics events The causal influence weight of is: ; in, For logistics events Logistics events The causal influence weight of Logistics incident Logistics events The nonlinear mapping function composed of the abstract feature matrix, is the weight matrix, is the bias vector, is the activation function; This formula is used to quantify the strength of causal relationships between different logistics events and reflects the direct interaction between events. Its principle is based on the linear mapping and nonlinear activation mechanism in deep neural networks, which can model the relationship between complex high-dimensional abstract features, so that the system can not only identify the co-occurrence between events, but also explore their potential causal driving mechanisms, thereby providing a structured causal basis for subsequent scheduling optimization.

[0024] S34, causal reasoning parameter generation unit, quantifies the causal relationship between each logistics event according to the causal influence weight, and generates causal reasoning parameters ,The causal reasoning parameters include event impact weight, resource scheduling impact factor and transportation path adjustment factor, where the event impact weight reflects the causal effect strength of a certain type of logistics event on other event nodes in scheduling, the resource scheduling impact factor is used to describe the matching relationship and dependency relationship between tasks and resources such as vehicles and manpower, and the path adjustment factor indicates the degree of disturbance of external change factors on the feasibility of the current path plan, which is used to assist dynamic path reconstruction; S35, model parameter optimization unit, uses the back propagation algorithm to optimize the weight matrix of the quantitative model of causal relationship between logistics events With the bias vector Dynamic adjustments are made to ensure that the causal reasoning module can stably and accurately output the above causal reasoning parameters in different logistics scenarios.

[0025] This implementation method constructs an event modeling process based on a deep causal reasoning network. First, the logistics event features in the fused data are standardized, and information such as time, location, and status are extracted to form a feature vector matrix. Then, a multi-layer neural network is used to perform nonlinear mapping of the feature vectors layer by layer to generate an abstract feature representation, build a quantitative model of the causal relationship between events, calculate the causal influence weights between each logistics event, and further generate causal reasoning parameters for logistics scheduling and path planning, including the degree of event impact, scheduling factors, and path adjustment amounts. By introducing a back-propagation mechanism to continuously optimize the model parameters, the system can dynamically adapt to changes in event logic in different scenarios, effectively improving the modeling accuracy of complex logistics relationships and the reliability of causal decision-making.

[0026] In this implementation manner, the S32 specifically includes: S321, input preprocessing unit, used to receive the output event feature vector matrix , for the event feature vector matrix The data in are normalized to form a normalized feature matrix ; S322, layer-by-layer mapping unit, based on deep causal reasoning algorithm, builds A multi-layer neural network with hidden layers, each hidden layer performs a linear transformation and activation function combination operation, and normalizes the feature matrix Perform layer-by-layer mapping and define the initial input as , for the layer , calculated using the following formula: ; in, Indicates The hidden representation of the layer, Indicates The weight matrix of the layer, Indicates The bias vector of the layer, is a nonlinear activation function, Indicates The hidden representation of the layer is mapped layer by layer to obtain the final hidden representation ; This formula is used to perform nonlinear mapping of event feature vectors layer by layer in a multi-layer neural network, and to achieve gradual abstraction and extraction of features through a combination of weight transformation and activation function. The principle is to superimpose linear transformations and nonlinear activation operations of multiple hidden layers to enable the model to have stronger feature expression capabilities, thereby mining the implicit causal structure and high-dimensional feature relationships in logistics events. This process provides a high-quality feature input basis for subsequent causal reasoning parameter calculations.

[0027] S323, feature abstraction unit, assuming that the nonlinear mapping function is , the final hidden representation of the output of the layer-by-layer mapping unit Perform nonlinear transformation to generate an abstract feature matrix.

[0028] This implementation method constructs a multi-layer neural network model based on a deep causal reasoning architecture, performs a nonlinear mapping operation of layer-by-layer linear transformation and activation function combination on the normalized event feature vector matrix, gradually extracts hidden semantic information at different levels, and finally generates a high-level abstract feature matrix for causal relationship modeling through an abstract feature transformation function. This method effectively retains the deep feature representation of logistics events in different spatial, temporal and resource states, enables the model to accurately capture the potential causal structure between logistics events, improves the ability to understand event-driven mechanisms, and provides a reliable feature foundation for subsequent causal impact quantification and intelligent decision-making.

[0029] In this implementation manner, S4 specifically includes: S41, a modality data construction unit, used to receive the generated fusion data, and divide the fusion data into a plurality of modality subsets, wherein the modality subsets include sensor modality, GPS modality and log modality, which correspond to environmental perception data, trajectory positioning data and operation scheduling data, respectively, to form a modality input set after division; S42, a modal feature encoding unit, based on the modal input set, constructs a corresponding feature encoding network for each modal subset, the feature encoding network includes several hidden layers, each layer contains trainable weight parameters and bias parameters, extracts its primary representation features, and sets The modal data is , which is coded as: ; in, Representing modality The encoding function, is the modal eigenvector; S43, cross-modal contrastive learning unit, for generating modal feature vectors based on Construct a cross-modal self-supervised learning task. By constructing positive and negative sample pairs and introducing a contrast loss function, the feature vector space aggregation of the same logistics event in different modes and the feature vector space separation of different logistics events are achieved to generate an optimized modal feature vector. S44, feature interaction fusion unit, is used to optimize the modal feature vector for feature alignment and fusion, use the cross-attention mechanism to calculate the correlation score between the modalities, and combine the fusion strategy to generate the feature vector , the feature vector Including spatiotemporal trajectory characteristics, vehicle load characteristics and transportation status characteristics.

[0030] This implementation method divides the fused data into sensor modality, GPS modality and log modality, which correspond to environmental perception, trajectory positioning and operation scheduling data respectively, constructs a modal input set, and designs a feature encoding network for each modal subset to extract primary feature representation. Through the cross-modal contrast learning mechanism, positive and negative sample pairs are constructed and the contrast loss function is introduced to achieve the aggregation of representations of the same logistics event in different modalities and the distinction between different events, generate optimized modal features, and use the cross-attention mechanism to fuse the optimized feature vectors to generate a unified feature representation containing information such as spatiotemporal trajectory, vehicle load and transportation status. This method effectively solves the problems of weak processing capabilities of traditional methods for multi-source heterogeneous data, insufficient feature expression and difficulty in semantic alignment, and improves the adaptability of the logistics scheduling system to complex scenarios and the ability to understand multimodal data.

[0031] In this implementation manner, the S43 specifically includes: S431, sample pair construction unit, used to receive modal feature vector , construct positive sample pairs for the same logistics event between different modes, and construct negative sample pairs between different logistics events to form a sample pair set , where each Represents pairs of feature vectors from different modalities, where positive sample pairs correspond to the same logistics event and negative sample pairs correspond to different logistics events; S432, contrast loss function definition unit, used to define the loss function based on the sample pair set Construct a cross-modal self-supervised contrastive learning objective and define the contrastive loss function as follows: ; in, is the contrast loss function, is the similarity function between feature vectors, is the temperature coefficient, is the anchor feature vector Constitute the set of negative sample pairs, is the total number of sample pairs, is the interference feature vector and the anchor feature vector Constitute a negative sample pair, To match the feature vector with the anchor feature vector Constitute a positive sample pair; The principle of this formula is to maximize the similarity between positive sample pairs and minimize the similarity between the anchor point and all negative sample pairs, thereby guiding the model to align the multimodal features of the same logistics event in the shared embedding space and distinguish the features of different events. This mechanism can effectively improve the consistency and discriminability of inter-modal representations, providing a more stable feature foundation for subsequent fusion and scheduling.

[0032] S433, a feature representation optimization unit, used to adjust the trainable weight parameters and bias parameters of the feature encoding network according to the defined contrast loss function, so that the feature vectors in the positive sample pairs converge in distance in the feature vector space, and the feature vectors in the negative sample pairs separate in distance in the feature vector space, so as to form a distinguishable cross-modal embedding representation structure; S434, optimized representation output unit, used to embed the distinguishable cross-modal representation structure to generate the optimized modal feature vector Output to the feature interaction fusion unit.

[0033] This implementation achieves consistent modeling and enhanced distinguishability of feature representations of logistics events under different modalities by constructing a set of positive and negative sample pairs and introducing a cross-modal contrast loss function. Specifically, the system receives feature vectors generated by a multimodal encoding network, automatically constructs positive sample pairs of the same event and negative sample pairs of different events, optimizes the similarity between anchor features and positive and negative samples through a contrast loss function, and dynamically updates the weight parameters and bias parameters in the encoding network during training, so that the model can converge similar event representations and distance different event representations in a unified feature space, and finally generate an optimized modal feature vector. This mechanism significantly improves the distinguishing ability of feature representations and the cross-modal alignment effect, providing more stable and accurate input expressions for subsequent scheduling fusion and path decision-making.

[0034] In this implementation manner, S5 specifically includes: S51, feature integration preparation unit, used to receive causal reasoning parameters and the eigenvector , and the causal inference parameters and the eigenvector Perform format conversion and dimension unification to construct an input alignment vector set ; S52, an adaptive weight generating unit, for aligning a vector set based on an input Establish a fusion strategy to perform weighted integration of causal inference parameters and feature vectors to construct a joint representation vector , the following formula is used for fusion: ; in, is an adjustable fusion weight coefficient, which is used to control the contribution ratio of causal inference parameters and feature vectors; This formula realizes the linear weighted combination of causal information and fusion features. The value of can dynamically balance the importance of causal relationships and perceptual features in the final decision, ensuring that the model has flexible adaptability and representation capabilities in different scenarios. The principle of this mechanism is based on the idea of ​​linear superposition of features, which organically integrates the structural explanatory information provided by causal reasoning with data-driven multimodal features, thereby improving the expression integrity and discrimination ability of joint features in complex logistics scheduling tasks.

[0035] S53, feedback control unit, builds a logistics execution system and generates real-time feedback information, which includes task completion time, resource usage status and execution deviation indicators, builds a weight update function based on the real-time feedback information, and dynamically adjusts the adjustable fusion weight coefficient , to adapt to the current operation status of the logistics system; S54, decision generation unit, used to receive the joint representation vector , and combines the current network status data for strategic reasoning to generate logistics scheduling and resource allocation decisions including optimal transportation routes, dynamic distribution plans and resource allocation plans.

[0036] This implementation receives causal reasoning parameters and fusion feature vectors, unifies the format, constructs an input alignment vector set, and integrates the information of the two through an adjustable fusion weight coefficient based on this set to generate a joint representation vector for scheduling decisions. The system constructs a feedback control mechanism to collect feedback information such as task completion time, resource usage status, and execution deviation in real time, and dynamically adjusts the fusion weight to adapt to changes in the operating status of the logistics system. Based on the joint representation vector and the current network status data, the system infers and generates the optimal transportation path, dynamic distribution plan, and resource allocation plan, which effectively improves the response speed, adaptability, and execution accuracy of the decision, and realizes intelligent logistics scheduling optimization under the joint support of causal drive and data features.

[0037] In this implementation manner, the S52 specifically includes: S521, feature similarity calculation unit, used to receive the output input alignment vector set , and calculate the feature representation similarity of the input alignment vector set, the feature representation similarity uses the normalized cosine similarity function: ; in, represents the feature representation similarity between the causal inference parameters and the fused feature vector, represents the vector dot product operation, Represents the L2 norm of the vector; This formula is based on the normalized cosine similarity principle and measures the consistency of the directions of two vectors in the representation space. The output value is The value range is The closer the value is to 1, the more consistent the directions of the two vectors are and the stronger the semantic association is. This calculation can effectively characterize the correlation between causal information and fusion features, which is used to drive the adaptive initialization of subsequent fusion weights, and helps to improve the ability of fusion representation to retain key features and the pertinence of decision expression.

[0038] S522, weight coefficient initialization unit, used to represent similarity based on the calculated feature Determine the adjustable fusion weight coefficient , through the monotone mapping function Calculate, the For the interval The monotone mapping function defined in , used to control the initial fusion ratio between causal information and fusion features; S523, fusion representation generation unit, used to receive adjustable fusion weight coefficient , input alignment vector set , and construct the joint representation vector .

[0039] This implementation method uses a feature similarity calculation unit to perform normalized cosine similarity calculation on the causal reasoning parameter vector and the fused feature vector to obtain a correlation measure between the two in the feature space; the weight coefficient initialization unit then dynamically generates a fusion weight coefficient based on the similarity value through a monotonic mapping function to achieve proportional control of causal information and fused features, and the fusion representation generation unit weightedly combines the above weight coefficient with the input vector to construct a joint representation vector for subsequent decision reasoning. This method implements a weight initialization mechanism driven by feature similarity, enabling the system to automatically determine the fusion ratio of causal information and multimodal features under unsupervised conditions, improving the adaptability and expression accuracy of the fusion strategy, and providing a more reasonable decision-making basis for subsequent intelligent scheduling and resource allocation. Example

[0040] In order to verify the feasibility of the present invention in implementation, the present invention is applied to a large regional logistics center, which is responsible for express distribution and trunk transportation scheduling in multiple cities in the Yangtze River Delta region. The logistics center handles more than 200,000 parcels per day, covering more than 120 transportation routes and more than 300 transportation vehicles. The logistics data comes from a wide range of sources, including GPS tracks, storage status, order scheduling records, vehicle load information and other data sources. The traditional scheduling system mainly relies on fixed paths, manual experience judgment and static allocation rules. It is difficult to adapt to the operating environment with large order fluctuations and frequent changes in regional distribution loads. The scheduling accuracy is not high, the path planning is lagging, and the resource utilization efficiency is significantly reduced.

[0041] The logistics center deployed the system proposed in the present invention in the scheduling platform, focusing on testing the adaptive weight generation unit. The system automatically generates initial fusion weights based on historical causal reasoning parameters and real-time fusion feature vectors through feature similarity calculation, thereby constructing a joint representation vector for subsequent scheduling optimization and path planning. In actual operation, the system automatically collects and cleans nearly 60GB of multimodal logistics data every day, including vehicle trajectories, traffic congestion data, real-time package distribution density, etc., and uniformly performs format conversion and feature alignment.

[0042] For example, in a trunk transport task, the system detected that a large-scale temporary promotion in a certain area caused a surge in the storage of distribution stations, and there was severe congestion on the highway along the original planned route. The traditional system failed to respond in time, resulting in a delay of nearly 2 hours in some transportation tasks. After receiving a certain distribution data anomaly, this system integrated the event influencing factor of "promotional activities increase the pressure on path resources" identified by the causal reasoning module, combined with the latest traffic status and vehicle load information, dynamically adjusted the original scheduling plan, generated a new joint representation, and adaptively adjusted the fusion coefficient through the online feedback mechanism, and finally output the optimal path plan, significantly alleviating the transportation pressure.

[0043] During the experimental stage, the system was run for 14 days during peak and off-peak periods, respectively. The traditional system and the system of the present invention were set up for parallel testing, and indicators such as scheduling response time, path re-planning frequency, resource utilization, and order delivery timeliness were compared, and simulation experiments were conducted on the system's operating performance. The experiment uses preset multimodal logistics data, including simulated GPS track points, sensor status data, and task scheduling records. The data type is consistent with the data collected by the S1 module data acquisition module in the manual, and is uniformly input into the fusion system after preprocessing. After the system completes reasoning and fusion, it outputs the scheduling results, and statistics indicators such as scheduling response time, path adjustment frequency, resource allocation rate, and task completion time to form the comparative data in Table 1. The experimental data is shown below: Table 1 Comparison of the effects of the traditional logistics scheduling system and the system of the present invention (14-day experimental period) index Traditional dispatch system System of the present invention Average dispatch response time (seconds) 135 47 Path re-planning frequency (times / day) 1.2 3.8 Vehicle resource utilization rate (%) 71.4 89.6 Order delivery timeliness rate (%) 85.2 96.7 Average distance saved per single path (km) 0 11.3 Real-time exception handling success rate (%) 63.5 91.8 As can be seen from the table, the system of the present invention shortened the average dispatch response time from 135 seconds to 47 seconds, and the dispatch response speed increased by more than 65%. The system greatly enhanced the path replanning capability through dynamic fusion decision-making strategy, and was able to make adjustments to emergencies more timely. The resource utilization rate increased from 71.4% to 89.6%, reflecting that the system is more efficient in vehicle allocation, path design and transport batch integration. At the same time, the on-time rate of order delivery increased from 85.2% to 96.7%, indicating that the system has significant advantages in service quality assurance.

[0044] Further analysis of the correlation between "fusion weight change" and system output shows that during the peak promotion period, the average similarity value calculated by the system is 0.43, and the weight after mapping is stable at 0.64, indicating that causal information has a significant impact on decision-making; during non-peak hours, the similarity value increases to 0.76, and the weight drops to 0.42, and the system relies more on fusion feature expression for scheduling. This strategy of adaptively adjusting decision weights based on similarity effectively avoids the single dependence of causal drive or feature drive, and improves the overall scheduling robustness of the system.

[0045] In addition, in response to the historical records of frequent peaks in a certain type of night orders but a high delivery failure rate, the system conducted similarity clustering modeling on hundreds of historical tasks and found that there was a key causal path between the "vehicle's nighttime remaining load rate" and the "order accumulation density". The system automatically injected it as a parameter into the fusion vector, and ultimately reduced the delayed delivery rate of this type of orders from 8.7% to 2.3%.

[0046] In summary, this embodiment shows that the adaptive fusion weight generation mechanism based on feature similarity driven proposed in the present invention can automatically adjust the information fusion strategy according to the correlation between causal parameters and multimodal features, and combined with the feedback optimization mechanism, it can effectively improve the logistics scheduling efficiency, resource allocation rationality and path reconstruction capabilities in complex scenarios, and verify the practicality and advancement of the present invention in a real logistics operation environment.

[0047] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A logistics management system based on artificial intelligence, characterized in that: The steps include: S1, data acquisition module, used to collect sensor, and multi-source logistics data of operation logs, perform standardized preprocessing on the collected multi-source logistics data, and construct a logistics data set; S2, data fusion module, used to align and fuse logistics data sets in time and space, and generate fused data based on the spatiotemporal dynamic logistics network modeling method, including logistics event characteristics, transportation status characteristics, resource usage characteristics and current network status data; S3, causal reasoning module, based on deep causal reasoning algorithm, uses multi-layer neural network to quantitatively model the causal relationship between logistics events in the fusion data and output causal reasoning parameters; S4, self-supervised multimodal representation learning module, constructs cross-modal self-supervised pre-training tasks, and automatically extracts feature vectors of fused data through cross-modal feature interaction mechanisms, including spatiotemporal trajectory features, vehicle load features, and transportation status features; S5, the fusion decision module, adopts an adaptive dynamic decision weight adjustment strategy, adaptively integrates causal reasoning parameters and feature vectors according to real-time online feedback, and generates logistics scheduling and resource allocation decisions, including the optimal transportation route, dynamic distribution plan and resource allocation plan.

2. According to claim 1, a logistics management system based on artificial intelligence is characterized in that: The S3 specifically includes: S31, event feature preprocessing unit, receives fused data, standardizes the original features of each logistics event in the fused data, extracts the event occurrence time, geographic location, status information and numerical indicators, and forms an event feature vector matrix ; S32, multi-layer neural network mapping unit, based on deep causal reasoning algorithm, event feature vector matrix A multi-layer neural network is used for layer-by-layer nonlinear mapping. Each hidden layer outputs abstract feature representation through a combination of linear transformation and activation function. The nonlinear mapping function is assumed to be , get the abstract feature matrix; S33, causal impact calculation unit, builds a quantitative model of causal relationships between events based on the abstract feature matrix, and calculates logistics events using the following formula Logistics events The causal influence weight of is: ； in, For logistics events Logistics events The causal influence weight of Logistics incident Logistics events The nonlinear mapping function composed of the abstract feature matrix, is the weight matrix, is the bias vector, is the activation function; S34, causal reasoning parameter generation unit, quantifies the causal relationship between each logistics event according to the causal influence weight, and generates causal reasoning parameters ,The causal reasoning parameters include event impact weights, resource scheduling influencing factors, and transportation path adjustment factors; S35, model parameter optimization unit, uses the back propagation algorithm to optimize the weight matrix of the quantitative model of causal relationship between logistics events With the bias vector Dynamic adjustments are made to ensure that the causal reasoning module can stably and accurately output the above causal reasoning parameters in different logistics scenarios.

3. The artificial intelligence-based logistics management system according to claim 2 is characterized in that: The S32 specifically includes: S321, input preprocessing unit, used to receive the output event feature vector matrix , for the event feature vector matrix The data in are normalized to form a normalized feature matrix ; S322, layer-by-layer mapping unit, based on deep causal reasoning algorithm, builds A multi-layer neural network with hidden layers, each hidden layer performs a linear transformation and activation function combination operation, and normalizes the feature matrix Perform layer-by-layer mapping and define the initial input as , for the layer , calculated using the following formula: ； in, Indicates The hidden representation of the layer, Indicates The weight matrix of the layer, Indicates The bias vector of the layer, is a nonlinear activation function, Indicates The hidden representation of the layer is mapped layer by layer to obtain the final hidden representation ; S323, feature abstraction unit, assuming that the nonlinear mapping function is , the final hidden representation of the output of the layer-by-layer mapping unit Perform nonlinear transformation to generate an abstract feature matrix.

4. The artificial intelligence-based logistics management system according to claim 2, characterized in that: The S4 specifically includes: S41, a modality data construction unit, used to receive the generated fusion data, and divide the fusion data into a plurality of modality subsets, wherein the modality subsets include sensor modality, GPS modality and log modality, which correspond to environmental perception data, trajectory positioning data and operation scheduling data, respectively, to form a modality input set after division; S42, a modal feature encoding unit, based on the modal input set, constructs a corresponding feature encoding network for each modal subset, the feature encoding network includes several hidden layers, each layer contains trainable weight parameters and bias parameters, extracts its primary representation features, and assumes The modal data is , which is coded as: ; in, Representing modality The encoding function, is the modal eigenvector; S43, cross-modal contrastive learning unit, for generating modal feature vectors based on Construct a cross-modal self-supervised learning task. By constructing positive and negative sample pairs and introducing a contrast loss function, the feature vector space aggregation of the same logistics event in different modes and the feature vector space separation of different logistics events are achieved to generate an optimized modal feature vector. S44, feature interaction fusion unit, is used to optimize the modal feature vector for feature alignment and fusion, use the cross-attention mechanism to calculate the correlation score between the modalities, and combine the fusion strategy to generate the feature vector , the feature vector Including spatiotemporal trajectory characteristics, vehicle load characteristics and transportation status characteristics.

5. The artificial intelligence-based logistics management system according to claim 4, characterized in that: The S43 specifically includes: S431, sample pair construction unit, used to receive modal feature vector , construct positive sample pairs for the same logistics event between different modes, and construct negative sample pairs between different logistics events to form a sample pair set , where each Represents pairs of feature vectors from different modalities, where the positive sample pairs correspond to the same logistics event and the negative sample pairs correspond to different logistics events; S432, contrast loss function definition unit, used to define the loss function based on the sample pair set Construct a cross-modal self-supervised contrastive learning objective and define the contrastive loss function as follows: ； in, is the contrast loss function, is the similarity function between feature vectors, is the temperature coefficient, is the anchor feature vector Constitute the set of negative sample pairs, is the total number of sample pairs, is the interference feature vector and the anchor feature vector Constitute a negative sample pair, To match the feature vector with the anchor feature vector Constitute a positive sample pair; S433, a feature representation optimization unit, used to adjust the trainable weight parameters and bias parameters of the feature encoding network according to the defined contrast loss function, so that the feature vectors in the positive sample pairs converge in distance in the feature vector space, and the feature vectors in the negative sample pairs separate in distance in the feature vector space, so as to form a distinguishable cross-modal embedding representation structure; S434, optimized representation output unit, used to embed the distinguishable cross-modal representation structure to generate the optimized modal feature vector Output to the feature interaction fusion unit.

6. The artificial intelligence-based logistics management system according to claim 4, characterized in that: The S5 specifically includes: S51, feature integration preparation unit, used to receive causal reasoning parameters and the eigenvector , and the causal inference parameters and the eigenvector Perform format conversion and dimension unification to construct an input alignment vector set ; S52, an adaptive weight generating unit, for aligning a vector set based on an input Establish a fusion strategy to perform weighted integration of causal inference parameters and feature vectors to construct a joint representation vector , the following formula is used for fusion: ； in, is an adjustable fusion weight coefficient, which is used to control the contribution ratio of causal inference parameters and feature vectors; S53, feedback control unit, builds a logistics execution system and generates real-time feedback information, which includes task completion time, resource usage status and execution deviation indicators, builds a weight update function based on the real-time feedback information, and dynamically adjusts the adjustable fusion weight coefficient , to adapt to the current operation status of the logistics system; S54, decision generation unit, used to receive the joint representation vector , and combines the current network status data for strategic reasoning to generate logistics scheduling and resource allocation decisions including optimal transportation routes, dynamic distribution plans and resource allocation plans.

7. The artificial intelligence-based logistics management system according to claim 6, characterized in that: The S52 specifically includes: S521, feature similarity calculation unit, used to receive the output input alignment vector set , and calculate the feature representation similarity of the input alignment vector set, the feature representation similarity uses the normalized cosine similarity function: ； in, represents the feature representation similarity between the causal inference parameters and the fused feature vector, represents the vector dot product operation, Represents the L2 norm of the vector; S522, weight coefficient initialization unit, used to represent similarity based on the calculated feature Determine the adjustable fusion weight coefficient , through the monotone mapping function Calculate, the For the interval The monotone mapping function defined in , used to control the initial fusion ratio between causal information and fusion features; S523, fusion representation generation unit, used to receive adjustable fusion weight coefficient , input alignment vector set , and construct the joint representation vector .

Images