Digital operation management system and method for shipping document routes

Through the digital operation and management system of sea shipping document routes, data fusion and time series prediction algorithms are used to solve the problem of inefficient traditional sea shipping document management, automatic processing and risk management are realized, and the accuracy of shipping status prediction and scheduling decisions is improved. and flexibility.

CN118410946BActive Publication Date: 2024-11-05SHANGHAI COSCO SHIPPING CONTAINER TRANSPORT INFORMATION SERVICE CO LTD
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202410587035.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-11
Publication Date
2024-11-05
Estimated Expiration
2044-05-11

AI Technical Summary

Technical Problem

Traditional shipping document management relies on manual processing of paper documents, which is inefficient and easy to introduce human errors. The existing system lacks automated and effective risk management tools, which limits the real-time and flexibility of operations, affects resource allocation and scheduling decisions. optimization.

Method used

The digital operation and management system for shipping documents routes was designed, including data acquisition module, data processing and analysis module, shipping prediction module, scheduling recommendation module and implementation module, using data fusion algorithm, long and short-term memory network and improved generalized autoregression conditions. The variance model combines entropy weight method and time series prediction algorithm to automatically process and predict shipping status to generate the best scheduling scheme and implement it in real time.

Benefits of technology

It significantly improves the accuracy and real-time nature of data processing, provides a comprehensive data view and risk assessment, enhances the response ability to emergencies, optimizes resource allocation and scheduling decisions, and improves operational flexibility and efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118410946B_ABST
    Figure CN118410946B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of electronic digital data processing, and in particular to a digital operation management system and method for shipping document routes. The contents include: collecting and standardizing shipping data from different data sources, extracting and synthesizing feature information from different data sources; introducing time decay weights, and combining entropy weight method to obtain fused data; inputting fused data into a time series prediction algorithm, combining long short-term memory network and improved generalized autoregressive conditional heteroskedasticity model to predict future shipping status and potential delays; according to the predicted shipping status, obtaining the best scheduling plan and generating scheduling suggestions. It solves the technical problems that the traditional method relies on manual processing of paper documents, which is not only inefficient but also easy to introduce human errors; the existing system lacks automation, which affects the optimization of resource allocation and scheduling decisions; and the lack of effective risk management tools to predict and quantify potential operational risks.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the field of electronic digital data processing, and in particular to a digital operation management system and method for shipping document routes. Background Art

[0002] Ocean shipping operations rely on complex and sophisticated logistics and document management systems to ensure that goods can be safely and promptly transported from the point of shipment to the destination. With the continuous growth of global trade, ocean shipping operations are facing great pressure, and the number of documents to be processed and the complexity of operations are also increasing. Ocean shipping document management not only involves extensive regulatory compliance and diverse customer needs, but also needs to deal with many uncertainties, such as weather, political factors, economic fluctuations, etc., all of which may affect the efficiency and safety of shipping.

[0003] The rapid development of information technology, especially the advancement of data processing and communication technology, has provided a rare opportunity to optimize shipping operations and deepen the reform of state-owned enterprises. However, despite some improvements brought about by technological advances, the existing shipping management system still faces many challenges.

[0004] There are at least the following technical problems in the existing technology: the traditional method relies on manual processing of paper documents, which is not only inefficient but also prone to human errors, resulting in insufficient accuracy and reliability of data processing; the existing system lacks automation, which limits the real-time and efficiency of operations and affects the optimization of resource allocation and scheduling decisions; there is a lack of effective risk management tools to predict and quantify potential operational risks, and the ability to develop intelligent and adaptive scheduling strategies is limited, which reduces the ability to respond quickly to emergencies and the flexibility and efficiency of overall operations. Summary of the invention

[0005] The present invention provides a digital operation management system and method for shipping document routes to solve the problems that the traditional method relies on manual processing of paper documents, which is not only inefficient but also prone to human errors, resulting in insufficient accuracy and reliability of data processing; the existing system lacks automation, which limits the real-time and efficiency of operations and affects the optimization of resource allocation and scheduling decisions; it lacks effective risk management tools to predict and quantify potential operational risks, and has limited capabilities in intelligent and adaptive scheduling strategies, which reduces the ability to respond quickly to emergencies and the flexibility and efficiency of overall operations.

[0006] The digital operation management system and method of the shipping document route of the present invention specifically includes the following technical solutions:

[0007] The digital operation management system of shipping document routes includes the following parts:

[0008] A data acquisition module for collecting and standardizing shipping-related data from different data sources; the data acquisition module is connected to the data processing and analysis module;

[0009] The data processing and analysis module is used to introduce data fusion algorithms, extract and synthesize feature information from different data sources, and obtain fused data by combining the entropy weight method; the data processing and analysis module is connected to the shipping forecast module;

[0010] The shipping forecasting module is used to combine the long short-term memory network and the improved generalized autoregressive conditional heteroskedasticity model to predict the future shipping status and obtain the prediction error variance; the risk value and confidence interval are calculated based on the prediction error variance; the shipping forecasting module is connected to the scheduling suggestion module;

[0011] The scheduling suggestion module is used to construct a scheduling suggestion optimization function according to the predicted shipping status, obtain the best scheduling plan, and automatically generate scheduling suggestions; the scheduling suggestion module is connected to the implementation module;

[0012] Implementation module, used to implement the optimal scheduling solution.

[0013] The digital operation management method of the shipping document route includes the following steps:

[0014] S1. Collect and standardize shipping data from different data sources, introduce data fusion algorithm to extract and synthesize feature information from different data sources; introduce time decay weights and combine them with entropy weight method to obtain fused data;

[0015] S2. Input the fused data into the time series prediction algorithm, combine the long short-term memory network and the improved generalized autoregressive conditional heteroskedasticity model to predict the future shipping status and potential delays; according to the predicted shipping status, obtain the best scheduling plan and generate scheduling recommendations.

[0016] Preferably, the S1 specifically includes:

[0017] The information content of each feature in each data source is calculated based on the entropy weight method, and the features are weighted in combination with the time decay weight.

[0018] Preferably, the S2 specifically includes:

[0019] The fused data is converted into time series format and input into the time series forecasting algorithm; the time series forecasting algorithm combines the long short-term memory network and the improved generalized autoregressive conditional heteroskedasticity model to use the fused data to predict future shipping status and potential delays.

[0020] Preferably, the S2 specifically includes:

[0021] The fused data are organized into time series and processed using long short-term memory networks; the predicted residuals are analyzed using an improved generalized autoregressive conditional heteroskedasticity model to evaluate the uncertainty and stability of the prediction, quantify the credibility of future predictions and adjust risk assessments.

[0022] Preferably, the S2 specifically includes:

[0023] The error between the predicted value of the LSTM network and the actual observed value of the shipping status is used as the input of the improved generalized autoregressive conditional heteroskedasticity model to obtain the prediction error variance.

[0024] Preferably, the S2 specifically includes:

[0025] The risk value and confidence interval are calculated for each forecast error variance, and the risk value, forecast error variance and confidence interval are provided to decision makers in the form of a report.

[0026] Preferably, the S2 specifically includes:

[0027] According to the predicted shipping status, the best scheduling solution is obtained by solving the scheduling suggestion optimization function.

[0028] The beneficial effects of the technical solution of the present invention are:

[0029] 1. Automatically transcribe paper documents using optical character recognition (OCR) technology, significantly reducing the time and error rate of manual input. Automatically update the real-time location of ships and cargo loading and unloading information through the global positioning system (GPS) and the API interface of shipping companies and port management systems, significantly improving real-time performance and data accuracy.

[0030] 2. The data fusion algorithm integrates information from multiple data sources, providing a comprehensive data view, making decisions more comprehensive and accurate; the time series prediction algorithm combines long short-term memory networks and improved generalized autoregressive conditional heteroskedasticity models to effectively predict future shipping status and potential delays, helping management make scheduling decisions in advance and optimize resource allocation.

[0031] 3. The calculated risk value and confidence interval help management evaluate and manage potential risks, and enhance the ability to respond to market fluctuations and other external uncertainties; the system's real-time monitoring and automatic adjustment functions can quickly respond to emergencies and cope with unforeseen changes by adjusting routes or changing cargo loading and unloading plans.

[0032] 4. By integrating machine learning and artificial intelligence technologies, the system can learn historical data, continuously optimize prediction and scheduling algorithms, and improve the scientificity and accuracy of decision-making; the system can provide dynamic scheduling suggestions and provide real-time feedback and adjustments based on actual execution conditions, making scheduling more flexible and efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a structural diagram of the digital operation and management system for shipping document routes according to the present invention;

[0034] Figure 2 The present invention is a flowchart of the digital operation management method for shipping document routes. DETAILED DESCRIPTION

[0035] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. Based on the embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0037] The specific scheme of the digital operation management system and method for shipping document routes provided by the present invention is described in detail below with reference to the accompanying drawings.

[0038] Refer to the attached Figure 1 , which shows a structural diagram of a digital operation management system for shipping document routes provided by an embodiment of the present invention, the system includes the following parts:

[0039] Data collection module, data processing and analysis module, shipping forecast module, scheduling suggestion module, implementation module;

[0040] The data collection module is used to collect and standardize shipping-related data from multiple data sources, including shipping document data collection, real-time vessel location tracking, and port and cargo information access; the data collection module is connected to the data processing and analysis module;

[0041] The data processing and analysis module is used to introduce data fusion algorithms, extract and synthesize feature information from different data sources, and obtain fused data by combining the entropy weight method; the data processing and analysis module is connected to the shipping forecast module;

[0042] The shipping forecasting module is used to combine the long short-term memory network and the improved generalized autoregressive conditional heteroskedasticity model to predict the future shipping status and obtain the prediction error variance; the risk value and confidence interval are calculated based on the prediction error variance; the shipping forecasting module is connected to the scheduling suggestion module;

[0043] The scheduling suggestion module is used to construct a scheduling suggestion optimization function according to the predicted shipping status, obtain the best scheduling plan, and automatically generate scheduling suggestions; the scheduling suggestion module is connected to the implementation module;

[0044] An implementation module for implementing optimal scheduling solutions using automated tools and techniques.

[0045] Refer to the attached Figure 2 , which shows a flow chart of a method for digital operation and management of shipping document routes provided by an embodiment of the present invention, the method comprising the following steps:

[0046] S1. Collect and standardize shipping data from different data sources, introduce data fusion algorithm to extract and synthesize feature information from different data sources; introduce time decay weights and combine them with entropy weight method to obtain fused data;

[0047] The data collection module automatically collects and standardizes shipping data from multiple data sources, including: Shipping document data collection: using optical character recognition (OCR) technology to automatically scan and transcribe paper documents into digital format, reducing errors and time-consuming manual data entry; real-time vessel position tracking: automatically updating the real-time position information of the vessel through integration with the Global Positioning System (GPS) interface; port and cargo information access: automatically obtaining cargo loading and unloading information, estimated arrival time and other data through API interface connection with shipping companies and port management systems.

[0048] The data processing and analysis module uses data processing and analysis techniques to optimize shipping operations and convert the collected multi-source shipping data into a structured format for further analysis and processing. Considering the multi-source nature and processing complexity of multi-source shipping data, a data fusion algorithm is introduced to extract and synthesize useful information from different data sources so as to provide a more comprehensive data view for the processing of shipping documents and the management of routes. Considering the diversity of data sources and the real-time nature of data, time decay weights are introduced and combined with entropy weight methods to process and optimize OCR data from shipping documents, GPS tracking data, and information from port APIs.

[0049] First, various shipping data are cleaned and standardized, such as format unification and noise data removal; then features are extracted from each data source, such as text information in OCR data, coordinates and timestamps in GPS data, and cargo status and time information in API data, to identify and isolate data elements that have an impact on the subsequent decision-making process.

[0050] The information content of each feature is calculated based on the entropy weight method, and the time decay weight is combined to weight the features to optimize the relevance and real-time performance of the fusion results. The entropy weight H(i) represents the information content of the i-th feature, and the calculation formula is as follows:

[0051]

[0052] Among them, p ij represents the probability distribution of the i-th feature in the j-th sample, quantifying the information contribution of each feature in the overall data, and m is the number of samples.

[0053] Time decay weight w i The calculation formula is:

[0054]

[0055] Among them, t i is the difference between the data collection time and the current time, λ is the decay constant, and n is the number of features; the time decay weight formula takes into account the freshness and information content of the data to ensure that real-time data has a greater weight in decision making. The final calculation of the fused data F is:

[0056]

[0057] Among them, x i is the i-th feature; is the feature set variance of the ith feature; γ is the adjustment coefficient, which is used to control the influence of the variance on the overall fusion result. The above fusion formula not only combines the weight and actual value of the feature, but also considers the stability of the data through the variance term, making the fused data more reliable.

[0058] S2. Input the fused data into the time series prediction algorithm, combine the long short-term memory network and the improved generalized autoregressive conditional heteroskedasticity model to predict the future shipping status and potential delays; according to the predicted shipping status, obtain the best scheduling plan and generate scheduling recommendations.

[0059] The fused data is converted into a time series format and input into a time series forecasting algorithm that combines a long short-term memory network and an improved generalized autoregressive conditional heteroskedasticity model. The shipping forecasting module uses the fused data to predict future shipping status and potential delays to handle the nonlinear characteristics and potential high volatility of time series data. The fused data is organized into a time series to capture the time dependency and periodicity characteristics in the data; the time series is processed using a long short-term memory network to learn the long-term and short-term patterns therein, and to manage the time gaps and vanishing gradient problems in the time series; the residuals of the forecast are analyzed in combination with an improved generalized autoregressive conditional heteroskedasticity model to assess the uncertainty and stability of the forecast, quantify the credibility of future forecasts and adjust the risk assessment.

[0060] The long short-term memory network prediction model expression is:

[0061]

[0062] in, It represents the predicted value at time point t, which is the predicted output of future shipping status; LSTM is used to characterize the calculation of long short-term memory network; F t is the sequence data output from the data fusion algorithm, representing the fused data at time point t; Θ represents the long short-term memory network model parameter, indicating the predicted value It is a function based on the historical data sequence and parameterized by the long short-term memory network. The error between the predicted value and the actual observed value of the shipping status is used as the input of the improved generalized autoregressive conditional heteroskedasticity model to evaluate the volatility of the forecast. The expression of the improved generalized autoregressive conditional heteroskedasticity model is:

[0063]

[0064] in, is the forecast error variance at time t, that is, the volatility of the forecast value, which is used to measure the uncertainty and risk standard of the forecast value; α0 is a constant term used to describe the baseline fluctuation; α1 is the influence coefficient of the previous period's forecast error, which is used to capture the continuous impact of the error term; ∈ t-1 is the forecast error at time t-1, that is, the difference between the actual observed value and the predicted value; β1 is the coefficient of the variance of the previous period's forecast error, which is used to model the autoregressive nature of fluctuations; δ is the introduced nonlinear adjustment parameter, which is used to adjust the degree of influence of nonlinear effects; φ k is the nonlinear coefficient used to simulate the periodic fluctuations in the data; k is the index of the periodic term; q is the number of periodic terms; f is the frequency parameter related to the period, which is used to adjust the frequency of the periodic component.

[0065] Forecast Error Variance It is used to quantify the uncertainty of the forecast results and calculate the value at risk (VaR) and confidence interval for each forecast error variance. VaR is a commonly used risk management tool used to evaluate the maximum possible loss under normal market conditions. The calculation of VaR can use the historical simulation method, variance-covariance method or Monte Carlo simulation method to determine the potential maximum loss at a certain confidence level based on the volatility of the forecast and historical data; the confidence interval provides management with a range of possible fluctuations in the forecast results, helping them better understand the uncertainty of the forecast. The specific calculation is:

[0066]

[0067] in, Indicates the confidence level The risk value under is used to quantify the potential maximum loss; μ t It is the expected value of the forecast, representing the average level that may be achieved in the future; is normally distributed For example, for a 95% confidence level, z 0.05 Take 1.645. The calculated risk value and volatility forecast results are provided to decision makers in the form of reports, so that they can make more informed decisions based on sufficient information. The volatility forecast results include the forecast error variance and confidence interval.

[0068] The scheduling suggestion module automatically generates scheduling suggestions based on the predicted shipping status, including cargo arrival time, vessel location, and possible delays. For example, if a route is predicted to be delayed, the digital operation management system for shipping document routes will recommend adjusting the route or rescheduling cargo loading and unloading plans to maximize resource efficiency and respond to market changes. The scheduling suggestion optimization function is as follows:

[0069]

[0070] Among them, C t is the total scheduling cost at time t, and the goal is to minimize the total scheduling cost; s is an element in the set S of possible scheduling solutions; c s is the basic cost of scheme s; is a weight factor used to adjust the penalty intensity of time deviation; is the importance weight of cargo r, reflecting the priority of different cargoes; t sr is the actual arrival time of cargo r under plan s; d r is the expected arrival time of cargo r. The scheduling suggestion optimization function is solved by the existing objective function optimal solution method (such as genetic algorithm) to obtain the best scheduling solution.

[0071] According to the best scheduling solution obtained by the scheduling optimization function, a detailed implementation plan is formulated, including a specific cargo loading and unloading schedule, a ship scheduling route map, and necessary alternative plans. Each scheduling operation is assigned corresponding resources and personnel to ensure the effective implementation of the scheduling plan.

[0072] Integrate the scheduling execution plan with the existing operating system of the shipping company, such as exchanging data with the port management system, freight tracking system, etc. through API, to ensure data synchronization and real-time update of all operation steps. The implementation module uses automated tools and technologies (such as intelligent scheduling software and automated logistics management systems) to implement the best scheduling plan, reduce human errors and improve operational efficiency.

[0073] In summary, the digital operation management system and method for shipping document routes have been completed.

[0074] The order of the embodiments of the invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0075] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0076] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. Digital operation and management methods for shipping document routes, It is characterized in that The following steps are involved: S1. Collect and standardize shipping data from different data sources, and introduce data fusion algorithms to extract and synthesize feature information from standardized shipping-related data; The information content of each feature in each data source is calculated based on the entropy weight method. The calculation formula is as follows: Among them, the entropy weight H(i) represents the information content of the i-th feature; p ij represents the probability distribution of the i-th feature in the j-th sample; m is the number of samples; Combine the time decay weight to weight the features, the time decay weight w i The calculation formula is as follows: Among them, t i is the difference between the data acquisition time and the current time; λ is the decay constant; n is the number of features; Finally, the fused data F is obtained, and the calculation formula is as follows: Among them, x i is the i-th feature; is the feature set variance of the i-th feature; γ is the adjustment coefficient; S2. Design a time series prediction algorithm by combining the long short-term memory network and the improved generalized autoregressive conditional heteroskedasticity model; convert the fused data into a time series format and input it into the long short-term memory network to obtain the predicted value; use the prediction error between the predicted value and the actual observed value of the shipping status as the input of the improved generalized autoregressive conditional heteroskedasticity model to obtain the prediction error variance and predict the future shipping status and potential delays; obtain the best scheduling plan based on the predicted shipping status and generate scheduling recommendations; The specific formula of the improved generalized autoregressive conditional heteroskedasticity model is: in, is the forecast error variance at time t; α 0 is a constant term describing the baseline fluctuation; α 1 is the influence coefficient of the previous period’s forecast error; ∈ t-1 is the prediction error at time t-1; β 1 is the coefficient of the previous period forecast error variance; δ is the adjustment parameter of the nonlinear effect; φ k is the nonlinear coefficient; k is the index of the periodic term; q is the number of periodic terms; and f is the frequency parameter associated with the period.

2. According to the method for digital operation and management of shipping document routes according to claim 1, It is characterized in that The S2 specifically includes: The fused data are organized into time series and processed using long short-term memory networks; the predicted residuals are analyzed using an improved generalized autoregressive conditional heteroskedasticity model to evaluate the uncertainty and stability of the prediction, quantify the credibility of future predictions and adjust risk assessments.

3. According to the digital operation and management method of shipping document routes as claimed in claim 1, It is characterized in that The S2 specifically includes: The risk value and confidence interval are calculated for each forecast error variance, and the risk value, forecast error variance and confidence interval are provided to decision makers in the form of a report.

4. A digital operation management system for shipping document routes, applied to the digital operation management method for shipping document routes as claimed in claim 1, It is characterized in that Includes the following sections: A data collection module to collect and standardize shipping-related data from different data sources; The data acquisition module is connected to the data processing and analysis module; The data processing and analysis module is used to introduce data fusion algorithm, extract and synthesize feature information from standardized shipping-related data, and obtain fused data by combining entropy weight method; the data processing and analysis module is connected to the shipping forecast module; Shipping forecasting module, used to predict the future shipping status based on fusion data combined with long short-term memory network and improved generalized autoregressive conditional heteroskedasticity model, and obtain the prediction error variance; calculate the risk value and confidence interval based on the prediction error variance; the shipping forecasting module is connected to the scheduling suggestion module; The scheduling suggestion module is used to construct a scheduling suggestion optimization function according to the predicted shipping status, obtain the best scheduling plan, and automatically generate scheduling suggestions; the scheduling suggestion module is connected to the implementation module; the scheduling suggestion optimization function is as follows: Among them, C t is the total scheduling cost at time t; s is an element in the scheduling solution set S; c s is the basic cost of scheme s; is the weight factor that adjusts the intensity of the time deviation penalty; is the importance weight of item r; t sr is the actual arrival time of cargo r under plan s; d r is the expected arrival time of cargo r; Implementation module, used to implement the optimal scheduling solution.

Citation Information

Patent Citations

  • Stock prediction system integrating volatility

    CN114819296A

  • Global shipping information monitoring system based on big data

    CN116911468A

  • Track prediction method based on bidirectional long short-term memory network

    CN117217089A

  • System and method for predicting sea freight based on artificial intelligence

    KR102558105B1

  • Route planning system, route planning method, article-arrangement planning system, article-arrangement planning method, decision-making support system, computer program, and storage medium

    US20180080782A1