Intelligent logistics abnormity autonomous processing method based on operation planning and decision making intellectualization

Abstract

The invention discloses an intelligent logistics abnormity autonomous processing method based on operation planning and decision intelligence, and relates to the technical field of automatic logistics, and the method comprises the steps: collecting logistics data in real time through a logistics monitoring system; on the basis of an anomaly detection algorithm, abnormal conditions in the logistics transportation process are identified; performing preliminary intelligent analysis on the abnormal event; performing quantitative analysis on the abnormal event through analog simulation; generating at least one exception handling scheme based on an analysis result of the operation planning and decision intelligent model; predicting and evaluating the feasible exception handling scheme; selecting an optimal exception handling scheme, and issuing an execution instruction; and collecting a scheme execution result, and evaluating an exception handling effect. By introducing the operational research method and performing quantitative analysis on the abnormal events, the influence of the abnormal events on enterprise operation can be evaluated more accurately, a basis is provided for decision making, the decision making efficiency and accuracy are improved, and the processing cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of automated logistics, and specifically to an intelligent logistics anomaly autonomous processing method based on operation research and decision-making intelligence. Background Art

[0002] As an applied mathematics discipline, operations research provides a powerful tool for solving complex problems by establishing mathematical models to optimize resource allocation. In the logistics field, operations research methods can be used to simulate various possible abnormal situations, help decision-makers quantitatively analyze the impact of abnormal events, and formulate optimal abnormal handling solutions. Simulation technology can simulate the real logistics transportation process, including various links such as equipment operation, parameter adjustment, and fault troubleshooting, providing intuitive decision-making support for decision-makers.

[0003] Traditional logistics management methods often rely on manual monitoring and post-event processing, and it is difficult to accurately grasp abnormal situations in the logistics transportation process in real time. Facing the complex logistics transportation process and various possible abnormal situations, how to formulate the optimal abnormal handling solution is a difficult problem. Summary of the Invention

[0004] To solve the above technical problems, an intelligent logistics anomaly autonomous processing method based on operation research and decision-making intelligence is provided, and this technical solution solves the problems proposed in the above background art.

[0005] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] An intelligent logistics anomaly autonomous processing method based on operation research and decision-making intelligence, comprising:

[0007] Real-time collecting logistics data through a logistics monitoring system, where the logistics data includes transportation status, cargo location, and time nodes;

[0008] Based on an anomaly detection algorithm, identifying abnormal situations in the logistics transportation process and integrating the abnormal situation information into a logistics information management system to form an abnormal event record, where the abnormal situations include delays, losses, and damages;

[0009] Conducting a preliminary intelligent analysis on the abnormal event to determine the type, scope of influence, and possible causes of the abnormal event;

[0010] Introducing operations research methods to quantitatively analyze the abnormal event through simulation;

[0011] Generating at least one abnormal handling solution based on the analysis results of the operation research and decision-making intelligence model;

[0012] Using decision-making intelligent technology, based on the cost-benefit of the solution and the impact on the subsequent overall operation of the logistics system, predict and evaluate feasible abnormal handling solutions to obtain the feasibility indicators of each abnormal handling solution;

[0013] Based on the feasibility indicators of each abnormal handling solution, select the optimal abnormal handling solution and issue an execution instruction;

[0014] Collect the execution results of the solution, evaluate the effect of abnormal handling, and feedback the experiences and lessons in the handling process to the logistics information management system to optimize the future abnormal handling process.

[0015] Preferably, based on the anomaly detection algorithm, identify abnormal situations in the logistics transportation process and integrate the abnormal situation information into the logistics information management system to form an abnormal event record, which specifically includes:

[0016] Obtain logistics data through the logistics monitoring system;

[0017] Clean the obtained logistics data, remove noise and outliers, and standardize the data;

[0018] Based on the time series characteristics of logistics transportation data and the requirements of anomaly detection, select the exponential smoothing method as the anomaly detection algorithm;

[0019] Use historical data to train the exponential smoothing method anomaly detection model to enable it to learn the characteristics of normal logistics transportation behavior;

[0020] Set the predicted value of the first period as the initial value of the historical data;

[0021] Use the exponential smoothing method prediction formula to calculate the data prediction values of each time node to obtain the first exponential smoothing of the historical data;

[0022] By adjusting the smoothing constant in the exponential smoothing method prediction formula, control the influence degree of historical data on the current predicted value;

[0023] Through double exponential smoothing, the re-smoothing of the first exponential smoothing, obtain a time series with a linear trend to capture the linear growth or decline trend of the time series;

[0024] Through triple exponential smoothing, on the basis of double smoothing, perform re-smoothing to obtain a time series with a non-linear trend or seasonal variation to simulate complex time series change patterns;

[0025] Input the real-time collected logistics transportation data into the trained exponential smoothing method anomaly detection model for real-time monitoring;

[0026] The model determines whether there is an abnormal situation based on the degree of difference between the input data and the predicted values of the normal behavior pattern. If so, it marks the specific location and characteristics of the abnormal data. If not, it does not output anything.

[0027] The prediction formula of the exponential smoothing method is as follows:

[0028] y' t+1 = ay t +(1 - a)y' t ,

[0029] In the formula, y' t+1 is the predicted value of the (t + 1)-th period, a is the adjustable smoothing constant, and y t is the actual value of the t-th period, and t is the time node.

[0030] Preferably, the preliminary intelligent analysis of the abnormal event to determine the type, scope of influence and possible causes of the abnormal event specifically includes:

[0031] Match the detected abnormal event with the historical abnormal event type library to determine the type of this abnormal event;

[0032] Determine the goods, means of transportation and personnel involved in this abnormal event;

[0033] Evaluate the impact degree of this abnormal event on logistics efficiency, cost and customer satisfaction;

[0034] Through data analysis, identify the risk points existing in the logistics process and output them as the possible causes of this abnormal event.

[0035] Preferably, the introduction of the operations research method to conduct quantitative analysis of the abnormal event through simulation specifically includes:

[0036] Based on the simulation objective and problem definition, establish a corresponding simulation model;

[0037] The simulation model reflects the characteristics and behaviors of the logistics system by simulating the order processing, inventory management and transportation and distribution links;

[0038] Based on historical data, business rules and operations research principles, build a quantitative analysis model for abnormal events in the logistics system;

[0039] Predict the occurrence probability, impact degree and the efficiency of the treatment plan of the abnormal event;

[0040] Input the collected logistics data into the quantitative analysis model of abnormal events in the logistics system, and run the model to obtain the quantitative analysis results.

[0041] Preferably, generating at least one exception handling solution based on the analysis result of the operation research and decision-making intelligent model specifically includes:

[0042] Identifying the key factors causing the exception, where the key factors include equipment failures, human errors, and environmental factors;

[0043] Based on the type, scope of influence, and possible causes of the exception event, formulating at least one targeted handling solution;

[0044] The handling solution includes specific operation steps, required resources, and expected effects.

[0045] Preferably, using decision-making intelligent technology, predicting and evaluating the feasible exception handling solutions based on the cost-benefit of the solutions and the impact on the overall subsequent operation of the logistics system, and obtaining the feasibility indicators of each exception handling solution specifically includes:

[0046] Predicting the net cash flow of each period after the implementation of the exception handling solution;

[0047] Based on the predicted values of the net cash flow of each period, calculating the discount rate indicator of each exception handling solution using the discount rate formula;

[0048] Based on the logistics throughput and response time, calculating the improvement degree indicator of the logistics system efficiency after the implementation of each exception handling solution using the system efficiency formula;

[0049] Calculating the system stability indicator by dividing the number of times without failures after the implementation of each exception handling solution by the total number of operations;

[0050] Calculating the customer satisfaction improvement rate indicator by dividing the difference between the predicted values of customer satisfaction before and after the implementation of each exception handling solution by the customer satisfaction before the implementation of each exception handling solution;

[0051] Presetting the weight coefficients of each indicator;

[0052] Calculating the final feasibility indicator of each exception handling solution using the feasibility indicator formula;

[0053] The discount rate formula is:

[0054]

[0055] In the formula, t is the t-th month after the implementation of the exception handling solution, T is the implementation time of the exception handling solution, W t is the net cash flow of the t-th month, r is the discount rate, and C is the investment cost required for the exception handling solution;

[0056] The system efficiency formula is:

[0057]

[0058] Wherein, α is the improvement degree index of the logistics system efficiency, N is the logistics throughput after the implementation of the exception handling plan, n is the logistics throughput before the implementation of the exception handling plan, and t N is the response time after the implementation of the exception handling plan, and t n is the response time before the implementation of the exception handling plan;

[0059] The formula for the feasibility index is as follows:

[0060] h = ω 1 r + ω 2 α + ω 3 β + ω 4 γ,

[0061] Wherein, h is the final feasibility index of each exception handling plan, and ω 1 , ω 2 , ω 3 , ω 4 are the weights of the discount rate index, the improvement degree index of the logistics system efficiency, the system stability index, and the customer satisfaction improvement rate index respectively, β is the system stability index, and γ is the customer satisfaction improvement rate index.

[0062] Preferably, collecting the execution results of the plan, evaluating the effect of exception handling, and feeding back the experience and lessons in the handling process to the logistics information management system to optimize the future exception handling process specifically includes:

[0063] Comparing and analyzing the actual handling results with the predicted feasibility index to evaluate the efficiency and effect of exception handling;

[0064] Comparing the effects of different exception handling plans to find potential improvement spaces;

[0065] According to the results of the comparative analysis, writing an evaluation report on the exception handling plan, summarizing the effectiveness and deficiencies of exception handling, and putting forward improvement suggestions;

[0066] Feeding back the summarized potential improvement spaces and improvement suggestions to the logistics information management system, and realizing knowledge sharing and process optimization through function optimization;

[0067] Using the means of knowledge graph technology to construct a logistics exception handling knowledge base and optimize the future exception handling process.

[0068] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0069] Based on the anomaly detection algorithm, it can quickly identify anomalies in the logistics transportation process, which helps enterprises discover and handle potential problems in a timely manner, reduce losses. By introducing the methods of operations research and conducting quantitative analysis of abnormal events through simulation, it can more accurately evaluate the impact of abnormal events on enterprise operations, provide a basis for decision-making, improve the efficiency and accuracy of decision-making, reduce processing costs, and feedback the lessons learned in the processing process into the logistics information management system, which can continuously optimize the future anomaly handling process and improve the overall operation efficiency of the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 It is a diagram of the intelligent logistics anomaly autonomous handling method based on operations research and decision-making intelligence of the present invention;

[0071] Figure 2 It is a flowchart of the method for identifying anomalies in the logistics transportation process of the present invention;

[0072] Figure 3 It is a flowchart of the method for preliminary intelligent analysis of the abnormal event of the present invention;

[0073] Figure 4 It is a flowchart of the method for quantitative analysis of the abnormal event through simulation of the present invention;

[0074] Figure 5 It is a flowchart of the method for generating at least one anomaly handling plan based on the analysis results of the operations research and decision-making intelligence model of the present invention;

[0075] Figure 6 It is a flowchart of the method for predicting and evaluating feasible anomaly handling plans of the present invention;

[0076] Figure 7 It is a flowchart of the method for collecting the execution results of the plan, evaluating the effect of anomaly handling, and feeding back the lessons learned in the processing process into the logistics information management system of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0077] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments in the following description are only examples, and those skilled in the art can think of other obvious variations.

[0078] Referring to Figure 1 as shown, an intelligent logistics anomaly autonomous handling method based on operations research and decision-making intelligence includes:

[0079] Collecting logistics data in real time through a logistics monitoring system, where the logistics data includes transportation status, cargo location, and time nodes;

[0080] Based on the anomaly detection algorithm, identify anomalies in the logistics transportation process, and integrate the anomaly information into the logistics information management system to form an anomaly event record. The anomalies include delays, losses, and damages;

[0081] Conduct a preliminary intelligent analysis of the anomaly event to determine the type, scope of influence, and possible causes of the anomaly event;

[0082] Introduce operations research methods to conduct quantitative analysis of the anomaly event through simulation;

[0083] Generate at least one anomaly handling plan based on the analysis results of the operations research and decision-making intelligent model;

[0084] Utilize decision-making intelligent technology to predict and evaluate feasible anomaly handling plans based on the cost-benefit of the plans and the impact on the overall subsequent operation of the logistics system, and obtain the feasibility indicators of each anomaly handling plan;

[0085] Select the optimal anomaly handling plan based on the feasibility indicators of each anomaly handling plan and issue an execution instruction;

[0086] Collect the execution results of the plan, evaluate the effect of anomaly handling, and feedback the experiences and lessons in the handling process to the logistics information management system to optimize the future anomaly handling process.

[0087] Refer to Figure 2 As shown, based on the anomaly detection algorithm, identify anomalies in the logistics transportation process, and integrating the anomaly information into the logistics information management system to form an anomaly event record specifically includes:

[0088] Obtain logistics data through the logistics monitoring system;

[0089] Clean the obtained logistics data, remove noise and outliers, and standardize the data;

[0090] Based on the time series characteristics of logistics transportation data and the requirements of anomaly detection, select the exponential smoothing method as the anomaly detection algorithm;

[0091] Use historical data to train the exponential smoothing method anomaly detection model to enable it to learn the characteristics of normal logistics transportation behavior;

[0092] Set the predicted value of the first period as the initial value of the historical data;

[0093] Use the exponential smoothing method prediction formula to calculate the data prediction values at each time node to obtain the first exponential smoothing of the historical data;

[0094] Control the influence degree of historical data on the current predicted value by adjusting the smoothing constant in the exponential smoothing method prediction formula;

[0095] Through double exponential smoothing, which is the re - smoothing of single exponential smoothing, a time series with a linear trend is obtained to capture the linear growth or decline trend of the time series;

[0096] Through triple exponential smoothing, which is the re - smoothing based on double smoothing, a time series with a non - linear trend or seasonal variation is obtained to simulate the complex time series change pattern;

[0097] Input the real - time collected logistics transportation data into the trained exponential smoothing method - based anomaly detection model for real - time monitoring;

[0098] The model determines whether there is an abnormal situation according to the degree of difference between the input data and the predicted value of the normal behavior pattern. If so, it marks the specific location and characteristics of the abnormal data. If not, no output is made;

[0099] The prediction formula of the exponential smoothing method is as follows:

[0100] y' t+1 =ay t +(1 - a)y' t ,

[0101] In the formula, y' t+1 is the predicted value of the (t + 1) - th period, a is the adjustable smoothing constant, y t is the actual value of the t - th period, and t is the time node.

[0102] The exponential smoothing method is actually a special weighted moving average method, and its characteristics are as follows: First, the exponential smoothing method further strengthens the role of recent observed values in the observation period on the predicted value. The weights assigned to observed values at different times are unequal, so the weights of recent observed values are increased, enabling the predicted value to quickly reflect the actual changes in the market. The weights decrease in a geometric progression, with the first term of this progression being the smoothing constant and the common ratio being 1 minus the smoothing constant. Second, the weights assigned to observed values by the exponential smoothing method are flexible. Different smoothing constant values can be taken to change the rate of weight change. For example, if the smoothing constant takes a small value, the weight change is relatively rapid, and the recent change trend of the observed value can be quickly reflected in the exponentially moving average value. Therefore, by using the exponential smoothing method, different smoothing constant values can be selected to adjust the uniformity of time series observed values (i.e., the smoothness of trend changes).

[0103] Refer to Figure 3 As shown, the preliminary intelligent analysis of the abnormal event to determine the type, scope of influence, and possible causes of the abnormal event specifically includes:

[0104] Match the detected abnormal event with the historical abnormal event type library to determine the type of this abnormal event;

[0105] Identify the goods, means of transportation, and personnel involved in the abnormal event;

[0106] Evaluate the impact of the abnormal event on logistics efficiency, cost, and customer satisfaction;

[0107] Through data analysis, identify the risk points in the logistics process and output them as possible causes of the abnormal event.

[0108] Match the detected abnormal event information with the historical abnormal event type library, which should contain detailed information on all previous recorded abnormal events, including event types, causes, and handling results; track the transportation routes and status of these goods to understand whether they have reached their destinations or are still in an abnormal state; check the qualifications, status, and historical records of these means of transportation and personnel to evaluate whether they may be the causes of the abnormal event; analyze the impact of the abnormal event on subsequent transportation plans, such as whether it is necessary to adjust transportation routes or increase means of transportation.

[0109] Refer to Figure 4 As shown, introduce operations research methods and conduct quantitative analysis of the abnormal event through simulation, specifically including:

[0110] Based on the simulation objectives and problem definition, establish a corresponding simulation model;

[0111] The simulation model reflects the characteristics and behaviors of the logistics system by simulating order processing, inventory management, and transportation and distribution links;

[0112] Based on historical data, business rules, and operations research principles, construct a quantitative analysis model for abnormal events in the logistics system;

[0113] Predict the occurrence probability, impact degree, and efficiency of the handling solution of the abnormal event;

[0114] Input the collected logistics data into the quantitative analysis model for abnormal events in the logistics system and run the model to obtain the quantitative analysis results.

[0115] Operations research, as a discipline used to solve practical problems, generally has the following steps when dealing with various problems: defining objectives, formulating solutions, establishing models, and developing solution methods. Although it is unlikely that there is an operations research that can handle an extremely wide range of objects, certain abstract models have been formed in the development process of operations research and can be applied to solve a relatively wide range of practical problems. With the development of science and technology and productivity, operations research has penetrated into many fields and is playing an increasingly important role. Operations research itself is also constantly evolving, covering branches such as linear programming, nonlinear programming, integer programming, combinatorial programming, graph theory, network flow, decision analysis, queuing theory, reliability mathematics theory, inventory theory, game theory, search theory, and simulation.

[0116] Refer to Figure 5 As shown, based on the analysis results of the operation research and decision-making intelligent model, generating at least one abnormal handling plan specifically includes:

[0117] Identifying the key factors causing the abnormality, where the key factors include equipment failures, human errors, and environmental factors;

[0118] Based on the type, scope of influence, and possible inducements of the abnormal event, formulating at least one targeted handling plan;

[0119] The handling plan includes specific operation steps, required resources, and expected effects.

[0120] For each abnormal event and its key factors, formulating at least one targeted handling plan, and the handling plan should include the following elements: Specific operation steps: Clearly define each step in the handling process to ensure the executability of the plan; Required resources: List the human, material, financial, and other resources required for implementing the plan for resource allocation; Expected effects: Predict the effects after implementing the plan, including the degree of problem solving, improvement of customer satisfaction, cost savings, etc.

[0121] Refer to Figure 6 As shown, using decision-making intelligent technology, based on the cost-benefit of the plan and its impact on the overall subsequent operation of the logistics system, predicting and evaluating the feasible abnormal handling plans to obtain the feasibility indicators of each abnormal handling plan specifically includes:

[0122] Predicting the net cash flow for each period after implementing the abnormal handling plan;

[0123] Based on the predicted values of the net cash flow for each period, using the discount rate formula to calculate the discount rate indicator of each abnormal handling plan;

[0124] Based on the logistics throughput and response time, using the system efficiency formula to calculate the improvement degree indicator of the logistics system efficiency after implementing each abnormal handling plan;

[0125] By dividing the number of times without failures after implementing each abnormal handling plan by the total number of operations, calculating the system stability indicator;

[0126] By dividing the difference between the predicted values of customer satisfaction before and after implementing each abnormal handling plan by the customer satisfaction before implementing each abnormal handling plan, calculating the customer satisfaction improvement rate indicator;

[0127] Presetting the weight coefficients of each indicator;

[0128] Using the feasibility indicator formula to calculate the final feasibility indicator of each abnormal handling plan;

[0129] The discount rate formula is:

[0130]

[0131] Wherein, t is the t-th month after the implementation of the exception handling solution, T is the implementation time of the exception handling solution, and W t is the net cash flow in the t-th month, r is the discount rate, and C is the investment cost required for the exception handling solution;

[0132] The system efficiency formula is as follows:

[0133]

[0134] Wherein, α is the index of the improvement degree of the logistics system efficiency, N is the logistics throughput after the implementation of the exception handling solution, n is the logistics throughput before the implementation of the exception handling solution, and t N is the response time after the implementation of the exception handling solution, and t n is the response time before the implementation of the exception handling solution;

[0135] The feasibility index formula is as follows:

[0136] h = ω 1 r + ω 2 α + ω 3 β + ω 4 γ,

[0137] Wherein, h is the final feasibility index of each exception handling solution, and ω 1 、ω 2 、ω 3 、ω 4 are the weights of the discount rate index, the improvement degree index of the logistics system efficiency, the system stability index, and the customer satisfaction improvement rate index respectively, β is the system stability index, and γ is the customer satisfaction improvement rate index.

[0138] Understood from the definition, the discount rate is the conversion of future money into present value. The ratio of the amount of money less or more to the future money is the discount rate. The higher the discount rate, the less money can be obtained.

[0139] Refer to Figure 7 As shown, collect the implementation results of the solution, evaluate the effect of exception handling, and feedback the experience and lessons in the handling process to the logistics information management system. Optimizing the future exception handling process specifically includes:

[0140] Compare and analyze the actual handling results with the predicted feasibility indicators to evaluate the efficiency and effect of exception handling;

[0141] Compare the effects of different exception handling solutions to find potential improvement space;

[0142] According to the results of the comparative analysis, prepare an evaluation report on the exception handling plan, summarize the effectiveness and deficiencies of the exception handling, and put forward improvement suggestions.

[0143] Feed back the summarized potential improvement space and improvement suggestions into the logistics information management system, and through function optimization, achieve knowledge sharing and process optimization.

[0144] Utilize knowledge graph technology to construct a logistics exception handling knowledge base and optimize future exception handling processes.

[0145] Essentially, a knowledge graph is a knowledge base of semantic networks. From another perspective, from the perspective of practical applications, a knowledge graph can actually be simply understood as a multi-relational graph, generally including various types of nodes and various types of edges.

[0146] Furthermore, this solution also proposes a computer-readable storage medium, on which a computer-readable program is stored. When the computer-readable program is called, it executes the above-mentioned intelligent logistics exception autonomous handling method based on operations research and decision-making intelligence.

[0147] It can be understood that the storage medium can be a magnetic medium, such as a floppy disk, a hard disk, or a magnetic tape; an optical medium, such as a DVD; or a semiconductor medium, such as a solid-state drive (SSD).

[0148] In summary, the advantages of the present invention are as follows: Based on the anomaly detection algorithm, it can quickly identify anomalies in the logistics transportation process, which helps enterprises discover and handle potential problems in a timely manner, reduce losses, introduce operations research methods, conduct quantitative analysis on anomaly events through simulation, and can more accurately evaluate the impact of anomaly events on enterprise operations, provide a basis for decision-making, improve decision-making efficiency and accuracy, reduce processing costs, and feed back the lessons learned in the processing process into the logistics information management system, which can continuously optimize future exception handling processes and improve the overall operation efficiency of enterprises.

[0149] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification is only the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection required by the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent logistics abnormality autonomous processing method based on intelligent operations and decision-making, characterized in that: include: Collect logistics data in real time through a logistics monitoring system, the logistics data including transportation status, cargo location and time node; Based on the anomaly detection algorithm, identify abnormal situations in the logistics transportation process and integrate the abnormal situation information into the logistics information management system to form abnormal event records. The abnormal situations include delays, losses and damages; Conduct preliminary intelligent analysis of the abnormal events to determine the type, impact scope and possible causes of the abnormal events; Introducing operations research methods to conduct quantitative analysis of the abnormal events through simulation; Generate at least one exception handling plan based on the analysis results of the operations research and decision-making intelligent model; Using intelligent decision-making technology, based on the cost-effectiveness of the plan and the impact on the overall subsequent operation of the logistics system, feasible exception handling plans are predicted and evaluated, and the feasibility indicators of each exception handling plan are obtained; Based on the feasibility indicators of each exception handling solution, select the best exception handling solution and issue execution instructions; Collect the results of program execution, evaluate the effectiveness of exception handling, and feed back the lessons learned from the handling process to the logistics information management system to optimize future exception handling processes.

2. According to claim 1, a method for autonomously handling intelligent logistics anomalies based on intelligent operations and decision-making, characterized in that: The abnormal situation in the logistics transportation process is identified based on the abnormal detection algorithm, and the abnormal situation information is integrated into the logistics information management system to form abnormal event records, which specifically include: Obtain logistics data through logistics monitoring system; Clean the acquired logistics data, remove noise and outliers, and standardize the data; Based on the time series characteristics of logistics transportation data and the need for anomaly detection, exponential smoothing is selected as the anomaly detection algorithm; The exponential smoothing anomaly detection model is trained using historical data to learn the characteristics of normal logistics and transportation behavior; The forecast value for the first period is set as the initial value of the historical data; The exponential smoothing prediction formula is used to calculate the data prediction value of each time node to obtain an exponential smoothing of the historical data; By adjusting the smoothing constant in the exponential smoothing forecasting formula, the influence of historical data on the current forecast value can be controlled; Through the second exponential smoothing, the first exponential smoothing is smoothed again to obtain a time series with a linear trend, which captures the linear growth or decline trend of the time series; Through triple exponential smoothing, re-smoothing is performed on the basis of quadratic smoothing to obtain a time series with nonlinear trends or seasonal changes, simulating the change pattern of complex time series; Input the real-time collected logistics transportation data into the trained exponential smoothing anomaly detection model for real-time monitoring; The model determines whether there is an abnormality based on the difference between the input data and the predicted value of the normal behavior pattern. If so, the specific location and characteristics of the abnormal data are marked. If not, no output is made. The exponential smoothing prediction formula is: you t+1 =is t +(1-a)y' t , In the formula, y' t+1 is the forecast value for the t+1 period, a is an adjustable smoothing constant, y t is the actual value of the tth period, and t is the time node.

3. The method for autonomously handling intelligent logistics anomalies based on intelligent operations and decision-making according to claim 2 is characterized in that: The preliminary intelligent analysis of the abnormal event to determine the type, impact range and possible causes of the abnormal event specifically includes: Match the detected abnormal event with the historical abnormal event type library to determine the type of the abnormal event; Determine the goods, means of transport and personnel involved in the abnormal event; Assess the impact of the abnormal event on logistics efficiency, cost and customer satisfaction; Through data analysis, the risk points in the logistics process are identified and output as possible causes of the abnormal event.

4. According to claim 3, a method for autonomously handling intelligent logistics anomalies based on intelligent operations and decision-making, characterized in that: The introduction of operations research methods to quantitatively analyze the abnormal events through simulation specifically includes: Based on the simulation objectives and problem definition, establish the corresponding simulation model; The simulation model reflects the characteristics and behaviors of the logistics system by simulating order processing, inventory management, and transportation and distribution; Based on historical data, business rules and operation research principles, a quantitative analysis model for abnormal events in the logistics system is constructed; Predict the probability of occurrence of abnormal events, the extent of their impact, and the efficiency of treatment plans; The collected logistics data is input into the quantitative analysis model of abnormal events in the logistics system, and the model is run to obtain the quantitative analysis results.

5. The method for autonomously handling intelligent logistics anomalies based on intelligent operations and decision-making according to claim 4 is characterized in that: The generating of at least one exception handling solution based on the analysis results of the operations research and decision-making intelligent model specifically includes: Identify key factors that cause abnormalities, including equipment failure, human error, and environmental factors; Develop at least one targeted treatment plan based on the type, scope of impact and possible causes of the abnormal event; The treatment plan includes specific operational steps, required resources and expected results.

6. The method for autonomously handling intelligent logistics anomalies based on intelligent operations and decision-making according to claim 5 is characterized in that: The decision-making intelligent technology is used to predict and evaluate feasible exception handling solutions based on the cost-effectiveness of the solutions and the degree of impact on the overall subsequent operation of the logistics system, and the feasibility indicators of each exception handling solution are obtained, including: Predict the net cash flow of each period after the implementation of the abnormal handling plan; Based on the forecast value of net cash flow in each period, the discount rate indicator of each abnormal handling plan is calculated using the discount rate formula; Based on logistics throughput and response time, the system efficiency formula is used to calculate the improvement index of logistics system efficiency after the implementation of each exception handling solution; The system stability index is calculated by taking the quotient of the number of times no failure occurs after each abnormal handling scheme is implemented and the total number of operations; The customer satisfaction improvement rate index is calculated by taking the quotient of the difference between the customer satisfaction prediction values ​​before and after the implementation of each exception handling plan and the customer satisfaction before the implementation of each exception handling plan; Preset the weight coefficient of each indicator; The feasibility index formula is used to calculate the final feasibility index of each exception handling solution; The discount rate formula is: Where t is the tth month after the implementation of the exception handling plan, T is the implementation time of the exception handling plan, and W t is the net cash flow in the tth month, r is the discount rate, and C is the investment cost required for the abnormal handling plan; The system efficiency formula is: In the formula, α is the improvement index of logistics system efficiency, N is the logistics throughput after the implementation of the abnormal handling plan, n is the logistics throughput before the implementation of the abnormal handling plan, t N is the response time after the exception handling solution is implemented, t n The response time before the exception handling plan is implemented; The feasibility index formula is: h=ω1r+ω2α+ω3β+ω4γ, Where h is the final feasibility index of each exception handling solution, ω1, ω2, ω3, and ω4 are the weights of the discount rate index, the improvement degree index of the logistics system efficiency, the system stability index, and the customer satisfaction improvement rate index, respectively, β is the system stability index, and γ is the customer satisfaction improvement rate index.

7. The method for autonomously handling intelligent logistics anomalies based on intelligent operations and decision-making according to claim 6 is characterized in that: The collection of program execution results, evaluation of the effectiveness of exception handling, and feedback of lessons learned from the handling process to the logistics information management system to optimize future exception handling processes specifically include: Compare and analyze the actual processing results with the predicted feasibility indicators to evaluate the efficiency and effectiveness of exception processing; Compare the effects of different exception handling solutions to find potential areas for improvement; Based on the results of the comparative analysis, write an exception handling solution evaluation report, summarize the effectiveness and shortcomings of exception handling, and put forward improvement suggestions; Feedback the potential improvement space and improvement suggestions summarized into the logistics information management system to achieve knowledge sharing and process optimization through functional optimization; Utilize knowledge graph technology to build a logistics exception handling knowledge base and optimize future exception handling processes.

Images