International multimodal transport teaching method, system and device and storage medium

Abstract

The invention discloses an international multimodal transport teaching method, system and device and a storage medium, and the key points are that the method comprises the steps: obtaining a transportation starting point, a transportation terminal point and a transportation demand, and analyzing the transportation demand to obtain demand feature information; generating a plurality of initial transportation schemes according to the transportation starting point and the transportation ending point; evaluating the plurality of initial transportation schemes by taking the demand feature information as a constraint condition of a decision tree model to obtain a pre-selected transportation scheme; selecting an optimal transportation scheme by adopting a linear programming algorithm; determining a student selection scheme from the initial transportation schemes according to a scheme selection instruction sent by a student, and comparing the student selection scheme with the optimal transportation scheme to obtain a comparison result; and obtaining a decision basis according to the decision path of the decision tree model and the information entropy of each node, and obtaining a selection basis according to the calculation process of the linear programming algorithm. According to the invention, a clear decision basis can be provided for students, preliminary evaluation and refined analysis can be effectively carried out on the transportation scheme, and the precision and robustness are higher.

Description

[0001] The present invention claims priority to Chinese patent application filed with the Chinese Patent Office on December 10, 2024, with application number 2024118119822 and application name “International multimodal transport teaching method, system, equipment and storage medium”, the entire contents of which are incorporated by reference into the present invention. Technical Field

[0002] The present invention belongs to the technical field of multimodal transport, and in particular relates to an international multimodal transport teaching method, system, equipment and storage medium. Background Art

[0003] International multimodal transport, also known as multimodal transport, is a form of transport that was developed on the basis of container transport. It refers to the transport of goods from a takeover location in one country to a designated delivery location in another country by a multimodal transport operator using at least two different modes of transport in accordance with an international multimodal transport contract. International multimodal transport is applicable to multiple modes of transport, including water, road, rail and air.

[0004] With the rapid development of logistics technology, such as the widespread application of 5G, Internet of Things, big data analysis and other technologies in the logistics field, the industry's requirements for talents are also more focused on comprehensive skills and technical application capabilities. However, in the process of multimodal transport teaching, traditional multimodal transport teaching methods lack dynamic and interactive content, resulting in students only being able to passively accept knowledge and finding it difficult to form an in-depth understanding of complex multimodal transport processes. Summary of the invention

[0005] The purpose of the present invention is to provide an international multimodal transport teaching method, system, equipment and storage medium, which can effectively conduct preliminary evaluation and refined analysis of transportation plans, provide students with clear decision-making basis, and realize dynamic interaction with students.

[0006] A first aspect of the present invention provides an international multimodal transport teaching method, comprising:

[0007] Obtaining the transportation starting point, transportation end point and transportation demand, analyzing the transportation demand and obtaining demand characteristic information;

[0008] Generate multiple initial transportation plans according to the transportation starting point and transportation end point, and calculate the plan characteristic information of each initial transportation plan;

[0009] The demand characteristic information is used as a constraint condition of a decision tree model, and the solution characteristic information of each initial transportation solution is input into the decision tree model to obtain a corresponding pre-selected transportation solution;

[0010] Selecting the best transportation plan from the pre-selected transportation plans by using a linear programming algorithm;

[0011] Determine a student selection plan from the initial transportation plan according to the plan selection instruction issued by the student, compare the student selection plan with the optimal transportation plan, and obtain a comparison result;

[0012] According to the decision path of the decision tree model and the information entropy of each node, the decision basis for the pre-selected transportation plan is obtained, and according to the calculation process of the linear programming algorithm, the selection basis for the optimal transportation plan is obtained, and the comparison result, decision basis and selection basis are output.

[0013] In some embodiments, the method for generating the decision tree model includes:

[0014] Acquire a training set formed by feature information of a plurality of sample transportation plans, wherein each sample transportation plan has a corresponding plan attribute, each feature information of the sample transportation plan corresponds to a different feature attribute, and the feature attributes of the demand feature information correspond one-to-one to the feature attributes of the sample transportation plan;

[0015] The training set is used to generate a decision tree model, the information gain of each feature attribute under each node is iteratively calculated, the training data set contained in the current node is divided according to the feature attribute corresponding to the maximum information gain, and multiple child nodes are split based on the current node until the solution attribute can be determined according to the training data set contained in the node.

[0016] In some embodiments, the method for generating the decision tree model further includes:

[0017] Receiving user feedback information, the feedback information including: applicability evaluation and rationality evaluation;

[0018] Taking the transportation demand and the pre-selected transportation plan as new sample transportation plans, and updating the corresponding sample transportation plans in the training set according to the applicability evaluation and rationality evaluation;

[0019] Update the decision tree model using the updated sample transportation plan.

[0020] In some embodiments, the demand characteristic information includes: required transportation cost, required transportation time and required cargo safety, and the method of selecting the optimal transportation solution from the pre-selected transportation solutions using a linear programming algorithm includes:

[0021] Convert the required transportation cost, required transportation time and required cargo security into corresponding mathematical expressions respectively;

[0022] Construct an objective function based on transportation cost, transportation time, and cargo security;

[0023] The comprehensive value of each pre-selected transportation plan is calculated according to the mathematical expression of the required transportation cost, the mathematical expression of the required transportation time, the mathematical expression of the required cargo safety and the objective function, and the pre-selected transportation plan with the best comprehensive value is taken as the optimal transportation plan.

[0024] In some embodiments, generating a plurality of initial transportation plans according to the transportation starting point and the transportation destination includes:

[0025] Get map data;

[0026] Determine a route starting point and a route end point in the map data according to the transport starting point and the transport end point, perform route planning by a route planning algorithm, and obtain a plurality of transport routes;

[0027] Assign multiple means of transportation to each transportation route and obtain the corresponding initial transportation plan.

[0028] In some embodiments, the method further includes: teaching students about emergencies, specifically including the following steps:

[0029] receiving simulated burst data, wherein the simulated burst data is generated according to at least one of an actual case and external real-time data;

[0030] Obtaining a transportation node, where the transportation node is a current transportation location after simulated transportation according to an optimal transportation plan;

[0031] A plurality of initial transport plans are regenerated according to the simulated burst data, transport nodes and transport destinations.

[0032] In some embodiments, the method further includes: generating corresponding practice tasks and prompts according to the student's operation behavior, specifically including the following steps:

[0033] Record students' operation behaviors in real time, including: solution selection, operation steps, error frequency and completion time;

[0034] The operation steps with an error frequency greater than a frequency threshold are regarded as difficult steps, the types of schemes selected with an error frequency greater than the frequency threshold are regarded as difficult types, and the types of schemes selected with a completion time greater than a time threshold are regarded as timeout types;

[0035] Issue the practice tasks of difficulty type and timeout type to the students, and send the tips about the difficult steps when the operation steps of the students are the previous steps of the difficult steps during the practice of the students.

[0036] A second aspect of the present invention provides an international multimodal transport teaching system, comprising:

[0037] An acquisition and analysis module is used to obtain a transportation starting point, a transportation end point and transportation demand, analyze the transportation demand and obtain demand characteristic information;

[0038] A generating and calculating module is used to generate a plurality of initial transportation plans according to the transportation starting point and the transportation end point, and calculate and obtain the plan characteristic information of each initial transportation plan;

[0039] A scheme evaluation module, used to use the demand characteristic information as a constraint condition of a decision tree model, input the scheme characteristic information of each initial transportation scheme into the decision tree model, and obtain a corresponding pre-selected transportation scheme;

[0040] A scheme selection module, used for selecting the optimal transportation scheme from the pre-selected transportation schemes by using a linear programming algorithm;

[0041] A scheme comparison module is used to determine a student selection scheme from the initial transportation scheme according to the scheme selection instruction issued by the student, and compare the student selection scheme with the optimal transportation scheme to obtain a comparison result;

[0042] According to the output module, it is used to obtain the decision basis of the pre-selected transportation plan according to the decision path of the decision tree model and the information entropy of each node, obtain the selection basis of the optimal transportation plan according to the calculation process of the linear programming algorithm, and output the comparison result, decision basis and selection basis.

[0043] A third aspect of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0044] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the above method when executed by a processor.

[0045] The technical solution provided by the present invention has the following advantages and effects: the initial transportation plan is evaluated by a decision tree model, and the initial transportation plan can be preliminarily evaluated. Then, the pre-selected transportation plan is refined and analyzed by a linear programming algorithm to select the optimal transportation plan. Then, the transportation plan selected by the student is received, and the student's selected plan and the optimal transportation plan are compared to analyze the student's deficiencies. The decision tree model has the advantages of being intuitive and easy to explain. It can perform detailed attribute evaluation on the initial transportation plan through a series of rule branches, thereby providing students with a clear decision-making basis, using the calculation process of the linear programming algorithm as the basis for selecting the optimal transportation plan, and outputting the comparison results, decision basis and selection basis to students, thereby realizing dynamic interaction with students and promoting students' in-depth understanding of the multimodal transport process. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a schematic diagram of the flow chart of the international multimodal transport teaching method provided by the present invention;

[0047] Figure 2 is a schematic diagram of the structure of an exemplary decision tree model provided by the present invention;

[0048] Figure 3 It is a structural block diagram of the international multimodal transport teaching system provided by the present invention;

[0049] Figure 4 It is a diagram of the internal structure of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0050] In order to facilitate the understanding of the present invention, specific embodiments of the present invention will be described in more detail below with reference to the accompanying drawings.

[0051] Unless otherwise specified or defined, the "first, second..." used in this article is merely used to distinguish names and does not represent a specific quantity or order.

[0052] Unless specifically stated or defined otherwise, the term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0053] It should be noted that, in this document, “fixed to” or “connected to” may mean directly fixing or connecting to an element, or indirectly fixing or connecting to an element.

[0054] like Figure 1 As shown, this embodiment provides an international multimodal transport teaching method, including the following steps S1 to S6:

[0055] Step S1, obtaining the transportation starting point, transportation end point and transportation demand, analyzing the transportation demand, and obtaining demand characteristic information.

[0056] In actual applications, students input the transportation starting point, transportation end point and transportation demand through the human-computer interaction interface according to the published exercise tasks. The transportation demand includes cargo information, transportation cost, transportation time, etc. The trained large-scale language model is used to semantically understand and parse the transportation demand to extract key information such as cargo information, transportation cost and transportation time. Then, the key information is rechecked in combination with the industry-specific corpus to ensure the accuracy and professionalism of the data. Based on text correction and information extraction technology, possible input errors and omissions are supplemented or suggested to obtain demand feature information. For example, if the input transportation demand is "deliver cargo a within 7 days at the minimum cost", the key information extracted is "minimum cost", "7 days" and "cargo a". The "minimum cost", "7 days" and "cargo a" are rechecked in combination with the industry-specific corpus to obtain "minimum cost". " is corrected to "minimum transportation cost". According to the cargo characteristics of cargo a, "cargo a" is corrected to the corresponding transportation safety level. For example, if the cargo characteristics of cargo a are ordinary cargo, "cargo a" is corrected to "the third transportation safety level". According to the text correction and information extraction technology, "minimum transportation cost", "7 days" and "third transportation safety level" are checked for input errors or omissions. If there are no input errors and omissions, the demand characteristic information obtained is: minimum transportation cost, 7 days, third transportation safety level. The characteristic attribute corresponding to the minimum transportation cost is transportation cost, the characteristic attribute corresponding to 7 days is transportation time, and the characteristic attribute corresponding to the third transportation safety level is cargo safety. If the cargo characteristics of cargo a are fragile cargo, "cargo a" is corrected to "the second transportation safety level". If the cargo characteristics of cargo a are temperature-controlled cargo, "cargo a" is corrected to "the first transportation safety level".

[0057] Step S2: generating a plurality of initial transportation plans according to the transportation starting point and the transportation end point, and calculating and obtaining plan characteristic information of each initial transportation plan.

[0058] In practical applications, multimodal transport plans usually include: a transport plan consisting of at least two modes of transport: water, road, rail and air. Therefore, the initial transport plan usually includes: multiple sub-transportation routes and the corresponding means of transport for each sub-transportation route. According to each sub-transportation route and its corresponding means of transport, the transport cost, time cost and cargo safety of each sub-transportation route are calculated. According to the transport cost (such as fuel consumption), time cost and cargo safety of all sub-transportation routes of each initial transport plan, the transport cost, time cost and cargo safety of each initial transport plan are calculated, that is, the plan characteristic information of the initial transport plan is obtained.

[0059] Specifically, the generating of a plurality of initial transportation plans according to the transportation starting point and the transportation end point includes:

[0060] Get map data;

[0061] Determine a route starting point and a route end point in the map data according to the transport starting point and the transport end point, perform route planning by a route planning algorithm, and obtain a plurality of transport routes;

[0062] Assign multiple means of transportation to each transportation route and obtain the corresponding initial transportation plan.

[0063] In practical applications, the map data obtained is GIS map data, including terrain data and road data. GIS (Geographical Information System) is based on geographic spatial data and adopts geographic model analysis methods to provide a variety of spatial and dynamic geographic information in real time. It is a computer system that analyzes and displays geographic reference information. The same place as the transportation starting point is screened out from the map data, and the place is used as the route starting point. The same place as the transportation end point is screened out from the map data, and the place is used as the route end point. The route planning algorithm is used to perform route planning to obtain multiple transportation routes, including:

[0064] Combine the route network information in the map data to establish a route network map. Specifically, each platform, dock, airport and intersection on the highway is used as a node, and the routes between adjacent nodes are used as edges to obtain a route network map.

[0065] The starting point of the route is taken as the starting node, and the end point of the route is taken as the ending node. Starting from the starting node, its adjacent nodes are traversed, and the node that meets the node selection rule is taken as the next node until the ending node is reached, thereby obtaining multiple transportation routes.

[0066] Since the nodes in each transport route may be road sections, water transport sections, railway sections or air transport sections, the same type of sections between multiple consecutive nodes in the transport route are taken as sub-transportation routes, and the corresponding transportation tools are matched to them according to the type of sub-transportation routes, which facilitates the subsequent calculation of the transportation cost, time cost and cargo safety of the transport route.

[0067] Furthermore, the international multimodal transport teaching method also teaches students about emergencies, specifically including the following steps:

[0068] receiving simulated burst data, wherein the simulated burst data is generated according to at least one of an actual case and external real-time data;

[0069] Obtaining a transportation node, where the transportation node is a current transportation location after simulated transportation according to an optimal transportation plan;

[0070] A plurality of initial transport plans are regenerated according to the simulated burst data, transport nodes and transport destinations.

[0071] In actual applications, there may be emergencies such as traffic accidents, natural disasters, policy or regulatory changes. The corresponding simulated emergency data can be generated based on external real-time emergencies and / or case emergencies that occurred in actual cases. The simulated emergency data includes: simulated events and simulated locations. If the simulated emergency data is generated based on actual cases, in the process of simulating transportation according to the optimal transportation plan, any one of the remaining transportation locations in the optimal transportation plan is selected as the simulated location. If the simulated emergency data is generated based on external real-time data, the location of the real-time emergency is used as the simulated location. By connecting with external real-time data sources, such as traffic data, meteorological data, and real-time operation data of logistics companies, the simulation is closer to the actual scene. The introduction of such real-time data allows students to experience the dynamic changes in the real environment and improve their flexibility and effectiveness in responding to emergencies.

[0072] Specifically, the method of regenerating multiple initial transportation plans based on simulated emergency data, transportation nodes and transportation destinations includes: first determining whether the severity of the simulated event will affect the optimal transportation plan, if so, generating multiple initial transportation plans based on the simulated locations, transportation nodes and transportation destinations; if not, generating multiple initial transportation plans based on the transportation nodes and transportation destinations, so that students can experience the entire process of the emergency in a virtual environment and observe the effects of different response strategies. For example, in the simulation, when traffic congestion occurs, the system displays different initial transportation plans and their impacts. Students can choose one of the plans, and the system will provide real-time feedback on its impact on the overall transportation.

[0073] Step S3: using the demand characteristic information as a constraint condition of a decision tree model, inputting the solution characteristic information of each initial transportation solution into the decision tree model, and obtaining a corresponding pre-selected transportation solution.

[0074] Specifically, the method for generating the decision tree model includes:

[0075] Acquire a training set formed by feature information of a plurality of sample transportation plans, wherein each sample transportation plan has a corresponding plan attribute, each feature information of the sample transportation plan corresponds to a different feature attribute, and the feature attributes of the demand feature information correspond one-to-one to the feature attributes of the sample transportation plan;

[0076] The training set is used to generate a decision tree model, the information gain of each feature attribute under each node is iteratively calculated, the training data set contained in the current node is divided according to the feature attribute corresponding to the maximum information gain, and multiple child nodes are split based on the current node until the solution attribute can be determined according to the training data set contained in the node.

[0077] In practical applications, the sample transportation scheme of the training set uses historical data of actual logistics, so that the branch rules of the trained decision tree model are closer to the real business scenario, thereby improving the reliability and accuracy of the model in specific logistics tasks. The scheme attributes include: low transportation cost, fast transportation time, high cargo safety, both transportation cost and transportation time, both transportation time and cargo safety, and both transportation cost and cargo safety. The characteristic attributes include: transportation cost, transportation time and cargo safety. For example, the scheme attribute of sample transportation scheme A is fast transportation time. The characteristic information of sample transportation scheme A includes transportation cost x, transportation time y, and cargo safety as the second transportation safety level. In the decision tree model, information gain is the difference between the information entropy before and after the training set is divided by a certain feature, and information entropy is used to measure the uncertainty of random variables. Information gain can be used to evaluate the effect of using a certain feature to divide the training set. By continuously dividing the training set according to the characteristic attributes, the nodes of the decision tree model are generated accordingly, that is, the nodes are continuously split until the scheme attributes can be determined according to the data set contained in the child nodes obtained by the split, thereby obtaining a decision tree model with evaluation function.

[0078] Specifically, the method for generating the decision tree model further includes:

[0079] Receiving user feedback information, the feedback information including: applicability evaluation and rationality evaluation;

[0080] Taking the transportation demand and the pre-selected transportation plan as new sample transportation plans, and updating the corresponding sample transportation plans in the training set according to the applicability evaluation and rationality evaluation;

[0081] Update the decision tree model using the updated sample transportation plan.

[0082] In actual applications, users include students and teachers. Receiving feedback from students and teachers, that is, users' suitability evaluation and rationality evaluation, and analyzing it can identify possible problems in the decision tree model, such as the recommended pre-selected transportation plan under certain conditions does not meet expectations. Through continuous data collection and feedback analysis, the branch logic of the decision tree can be adaptively adjusted to ensure that the model responds more accurately to different input conditions, and can continuously learn and optimize, gradually improve decision rules, so that the decision tree model is more adaptable to dynamic logistics scenarios. In this application, the updated sample transportation plan is used to update the decision tree model, that is, the incremental learning method is used to continuously learn from new interactive data and update the parameters of the model so that the system can adapt to new learning needs and logistics scenarios. For example, the priority of certain transportation conditions may change over time or market changes, and incremental learning can help the model automatically adapt to such changes.

[0083] Step S4: Select the optimal transportation plan from the pre-selected transportation plans using a linear programming algorithm.

[0084] Specifically, the demand characteristic information includes: required transportation cost, required transportation time and required cargo safety. The linear programming algorithm is used to select the optimal transportation scheme from the pre-selected transportation schemes, including:

[0085] Convert the required transportation cost, required transportation time and required cargo security into corresponding mathematical expressions respectively;

[0086] Construct an objective function based on transportation cost, transportation time, and cargo security;

[0087] The comprehensive value of each pre-selected transportation plan is calculated according to the mathematical expression of the required transportation cost, the mathematical expression of the required transportation time, the mathematical expression of the required cargo safety and the objective function, and the pre-selected transportation plan with the best comprehensive value is taken as the optimal transportation plan.

[0088] In practical applications, the required transportation cost, required transportation time and required cargo safety in the demand characteristic information can be expressed as a set of linear equations or inequalities. According to the mathematical expressions of transportation cost, transportation time and cargo safety, the mathematical expressions of the required transportation cost, transportation time and cargo safety can be obtained. The expression of transportation cost can be:

[0089] C=C t +C l +C w +C e

[0090] Among them, C t : Transportation costs, which are calculated by calculating the fuel costs, mileage costs and other additional costs (such as tolls) of the means of transportation; C l : Loading and unloading costs, calculated based on the characteristics of the cargo and the efficiency of the equipment at the loading and unloading point; C w C: Storage costs, taking into account the storage needs of goods at the transit station or destination; e : Additional costs, including insurance, tariffs, and fines. Students can learn specific strategies to optimize costs by adjusting each cost parameter (such as reducing storage time or choosing alternative transportation) and observing its impact on total costs.

[0091] The expression for transportation time can be:

[0092] T=T d +T o +T m +T s

[0093] Among them, T d: Delay time of shipment; T o : Actual transportation time; T m : Transit waiting time; T s : Emergency processing time.

[0094] The expression of cargo safety can be:

[0095] S=1-(R h +R t +R e +R o )

[0096] Among them, R h : Human factor risk; R t : Risk of transportation tool factors; R e : Environmental risks, such as weather or geographical conditions; R o : Other risks, such as policy changes or market fluctuations. Risks are divided into high, medium and low levels, and students are provided with corresponding coping strategies. High risk corresponds to the third transportation safety level, medium risk corresponds to the second transportation safety level, and low risk corresponds to the first transportation safety level.

[0097] If the demand characteristic information is that the transportation cost is no more than 1,500 yuan, the transportation time is within 7 days, and the cargo safety is the third transportation safety level, then the mathematical expression of the demand transportation cost is C t +C l +C w +C e ≤1500, the mathematical expression of the required transportation time is T d +T o +T m +T s ≤7, the mathematical expression of the required cargo safety is 1-(R h +R t +R e +R o )≤ high risk, and then construct an objective function based on transportation cost, transportation time and cargo security. The objective function can be to minimize cost or maximize comprehensive benefits. In the social e-commerce environment, the goal is to find a transportation plan that achieves the best balance between cost, timeliness and safety. Combine these constraints with the objective function, and use linear programming solving tools (such as the Simplex algorithm) to find the optimal solution, so as to select the optimal transportation plan based on meeting transportation needs. In order to enhance students' understanding and operational capabilities of multimodal transport, this application requires a model that can intuitively present the decision-making basis to help students understand the selection process of logistics solutions. The combination of decision trees and linear programming algorithms is more in line with teaching objectives.

[0098] Furthermore, historical big data analysis is used to provide dynamic weights for transportation costs, transportation time, and cargo security. For example, if students set budget constraints, the priority of high-cost options will be automatically reduced, and sensitivity analysis can be used to show how changes in cost parameters affect the overall plan. If students choose a high-time priority plan, the system will automatically increase the recommended weight of high-speed transportation tools (such as air transportation) and prompt the corresponding cost increase. For example, when the student's selected plan involves a high-risk path, the system will provide an alternative low-risk path and its cost and time comparison.

[0099] Step S5: Determine the student selection plan from the initial transportation plan according to the plan selection instruction issued by the student, compare the student selection plan with the optimal transportation plan, and obtain a comparison result.

[0100] In practical applications, if the student chooses the initial transportation plan B and the optimal transportation plan is the initial transportation plan C, the transportation cost, transportation time and cargo safety of the initial transportation plans B and C are compared, and the differences between the two transportation plans are analyzed from multiple angles to help students discover their own shortcomings. Personalized improvement suggestions can also be provided based on the comparison results.

[0101] Specifically, the international multimodal transport teaching method further includes:

[0102] Record students' operation behaviors in real time, including: solution selection, operation steps, error frequency and completion time;

[0103] The operation steps with an error frequency greater than a frequency threshold are regarded as difficult steps, the types of schemes selected with an error frequency greater than the frequency threshold are regarded as difficult types, and the types of schemes selected with a completion time greater than a time threshold are regarded as timeout types;

[0104] Issue the practice tasks of difficulty type and timeout type to the students, and send the tips about the difficult steps when the operation steps of the students are the previous steps of the difficult steps during the practice of the students.

[0105] In practical applications, by recording and analyzing each student's operation behavior, we can identify the common difficulties students encounter in specific tasks or knowledge points. Based on the student's operation behavior, we provide personalized teaching suggestions for each student, such as strengthening the practice of certain knowledge points or guiding students to use different learning strategies, that is, strengthening students' practice of difficult and time-out types of knowledge points, and giving tips on difficult steps. Then, we generate a visual learning path and progress curve based on the student's operation behavior each time they practice, so that students can intuitively see their learning progress and improvement direction.

[0106] Step S6: According to the decision path of the decision tree model and the information entropy of each node, the decision basis for the pre-selected transportation plan is obtained, and according to the calculation process of the linear programming algorithm, the selection basis for the optimal transportation plan is obtained, and the comparison result, decision basis and selection basis are output.

[0107] In practical applications, due to the “black box” nature of the neural network model and the lack of intuitive explanation, it is difficult to show it to students in teaching. Students may find it difficult to understand why the model makes a certain decision, and the neural network model has high computational complexity, especially when processing large amounts of data, which may cause the system to respond too long, reducing the timeliness of teaching interaction. Compared to the use of a neural network model, this application uses a decision tree model to evaluate each initial transportation plan, and inputs the characteristic information of each plan into the decision tree model. For example, the plan characteristic information of the initial transportation plan A is: the transportation cost is 1,000 yuan, the transportation time is 3 days, and the cargo safety is the second transportation safety level; the demand characteristic information is: the transportation cost is no more than 1,500 yuan, the transportation time is within 7 days, and the cargo safety is the third transportation safety level. Figure 2 As shown in the figure, if cargo safety is the root node in the decision tree model, and transportation time and transportation cost are the intermediate nodes, the specific constraints in the root node and the intermediate nodes are updated according to the demand characteristic information, and the three constraints that the cargo safety level needs to meet the third transportation safety level, the transportation time needs to be delivered within 7 days, and the transportation cost needs to be no more than 1,500 yuan are obtained. The initial transportation plan A is first evaluated according to the root node to obtain the classification result of the root node, and then the classification results of each intermediate node are obtained in turn. The classification results of all nodes are used as the decision path, and then the information entropy under all nodes is calculated, and the decision path and the information entropy of each node are output as the basis for decision making.

[0108] The comprehensive value of each pre-selected transportation plan is calculated through a linear programming algorithm, and the plan feature information of each pre-selected transportation plan is substituted into the mathematical expression of the required transportation cost, the mathematical expression of the required transportation time, the mathematical expression of the required cargo safety and the objective function to obtain the corresponding calculation results, which are used as the basis for selection.

[0109] The international multimodal transport teaching method of the present invention evaluates the initial transport plan through a decision tree model. The decision tree model has the advantages of being intuitive and easy to explain. It can perform detailed attribute evaluation on the initial transport plan through a series of rule branches, thereby providing students with a clear decision-making basis. The linear programming algorithm performs well in processing high-dimensional data, boundary conditions and classification problems. Through the combination of the decision tree model and the linear programming algorithm, the initial transport plan can be effectively preliminarily evaluated and refined analyzed to achieve higher accuracy and robustness.

[0110] like Figure 3As shown, the embodiment of the present invention further provides an international multimodal transport teaching system, including:

[0111] The acquisition and analysis module 10 is used to obtain the transportation starting point, transportation end point and transportation demand, analyze the transportation demand and obtain demand characteristic information;

[0112] A generating and calculating module 20 is used to generate a plurality of initial transportation plans according to the transportation starting point and the transportation end point, and calculate and obtain the plan characteristic information of each initial transportation plan;

[0113] A scheme evaluation module 30 is used to use the demand characteristic information as a constraint condition of a decision tree model, input the scheme characteristic information of each initial transportation scheme into the decision tree model, and obtain a corresponding pre-selected transportation scheme;

[0114] A scheme selection module 40 is used to select the optimal transportation scheme from the pre-selected transportation schemes using a linear programming algorithm;

[0115] A scheme comparison module 50 is used to determine a student selected scheme from the initial transportation scheme according to the scheme selection instruction issued by the student, and compare the student selected scheme with the optimal transportation scheme to obtain a comparison result;

[0116] According to the output module 60, it is used to obtain the decision basis of the pre-selected transportation plan according to the decision path of the decision tree model and the information entropy of each node, obtain the selection basis of the optimal transportation plan according to the calculation process of the linear programming algorithm, and output the comparison result, decision basis and selection basis.

[0117] As an optional implementation, the solution evaluation module includes the following units not shown in the figure:

[0118] A training set acquisition unit, used to acquire a training set formed by feature information of a plurality of sample transportation schemes, wherein each sample transportation scheme has a corresponding scheme attribute, and each feature information of the sample transportation scheme corresponds to a different feature attribute;

[0119] A model generation unit is used to generate a decision tree model using the training set, iteratively calculate the information gain of each feature attribute under each node, divide the training data set contained in the current node according to the feature attribute corresponding to the maximum information gain, and split multiple child nodes based on the current node until the solution attribute can be determined according to the training data set contained in the node.

[0120] As an optional implementation, the solution evaluation module further includes the following units not shown in the figure:

[0121] A feedback receiving unit, used to receive feedback information from users, wherein the feedback information includes: applicability evaluation and rationality evaluation;

[0122] A sample updating unit, configured to use the transportation demand and the pre-selected transportation scheme as new sample transportation schemes, and update the corresponding sample transportation schemes in the training set according to the applicability evaluation and the rationality evaluation;

[0123] The model updating unit is used to update the decision tree model by adopting the updated sample transportation plan.

[0124] As an optional implementation, the demand characteristic information includes: required transportation cost, required transportation time and required cargo safety, and the solution selection module includes the following units not shown in the figure, including:

[0125] A conversion unit, used to convert the required transportation cost, the required transportation time and the required cargo safety into corresponding mathematical expressions respectively;

[0126] A construction unit for constructing an objective function based on transportation cost, transportation time and cargo safety;

[0127] The calculation unit is used to calculate the comprehensive value of each pre-selected transportation plan according to the mathematical expression of the required transportation cost, the mathematical expression of the required transportation time, the mathematical expression of the required cargo safety and the objective function, and take the pre-selected transportation plan with the best comprehensive value as the optimal transportation plan.

[0128] As an optional implementation, the generation calculation module includes the following units not shown in the figure, including:

[0129] A map acquisition unit, used for acquiring map data;

[0130] A route planning unit, configured to determine a route starting point and a route end point in the map data according to the transport starting point and the transport end point, and perform route planning through a route planning algorithm to obtain a plurality of transport routes;

[0131] The tool matching unit is used to assign multiple transportation tools to each transportation route to obtain the corresponding initial transportation plan.

[0132] As an optional implementation, the international multimodal transport teaching system further includes:

[0133] A data receiving module, configured to receive simulated burst data, wherein the simulated burst data is generated based on at least one of an actual case and external real-time data;

[0134] A node acquisition module is used to acquire a transportation node, where the transportation node is the current transportation location after simulated transportation according to the optimal transportation plan;

[0135] The scheme generating module is used to regenerate a plurality of initial transport schemes according to the simulated burst data, transport nodes and transport destinations.

[0136] As an optional implementation, the international multimodal transport teaching system further includes:

[0137] A recording module is used to record students' operation behaviors in real time, including: solution selection, operation steps, error frequency and completion time;

[0138] A judgment module, used for treating the operation step whose error frequency is greater than the frequency threshold as a difficult step, treating the type of solution selected whose error frequency is greater than the frequency threshold as a difficult type, and treating the type of solution selected whose completion time is greater than the time threshold as a timeout type;

[0139] The publishing module is used to publish difficult and time-out types of practice tasks to the students, and send prompts about difficult steps during the students' practice.

[0140] Each module of the international multimodal transport teaching system can be implemented in whole or in part by software, hardware and their combination. Each module and unit can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to each module.

[0141] like Figure 4 As shown, an embodiment of the present invention discloses a computer device, including a memory and a processor, wherein the memory stores a computer program;

[0142] The computer device may be a server, and its internal structure diagram may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the international multimodal transport teaching method described in the above embodiments is implemented.

[0143] Those skilled in the art will understand that Figure 4The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0144] The embodiment of the present invention further discloses a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute the international multimodal transport teaching method described in the above embodiments.

[0145] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0146] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. International multimodal transport teaching method, characterized by: include: Obtaining the transportation starting point, transportation end point and transportation demand, analyzing the transportation demand and obtaining demand characteristic information; Generate multiple initial transportation plans according to the transportation starting point and transportation end point, and calculate the plan characteristic information of each initial transportation plan; The demand characteristic information is used as a constraint condition of a decision tree model, and the solution characteristic information of each initial transportation solution is input into the decision tree model to obtain a corresponding pre-selected transportation solution; Selecting the best transportation plan from the pre-selected transportation plans by using a linear programming algorithm; Determine a student selection plan from the initial transportation plan according to the plan selection instruction issued by the student, compare the student selection plan with the optimal transportation plan, and obtain a comparison result; According to the decision path of the decision tree model and the information entropy of each node, the decision basis for the pre-selected transportation plan is obtained, and according to the calculation process of the linear programming algorithm, the selection basis for the optimal transportation plan is obtained, and the comparison result, decision basis and selection basis are output.

2. The international multimodal transport teaching method according to claim 1, characterized in that: The method for generating the decision tree model comprises: Obtaining a training set formed by feature information of a plurality of sample transportation plans, wherein each sample transportation plan has a corresponding plan attribute, and each feature information of the sample transportation plan corresponds to a different feature attribute; The training set is used to generate a decision tree model, the information gain of each feature attribute under each node is iteratively calculated, the training data set contained in the current node is divided according to the feature attribute corresponding to the maximum information gain, and multiple child nodes are split based on the current node until the solution attribute can be determined according to the training data set contained in the node.

3. The international multimodal transport teaching method according to claim 2, characterized in that: The method for generating the decision tree model further includes: Receiving user feedback information, the feedback information including: applicability evaluation and rationality evaluation; Taking the transportation demand and the pre-selected transportation plan as new sample transportation plans, and updating the corresponding sample transportation plans in the training set according to the applicability evaluation and rationality evaluation; Update the decision tree model using the updated sample transportation plan.

4. The international multimodal transport teaching method according to claim 1, characterized in that: The demand characteristic information includes: required transportation cost, required transportation time and required cargo safety. The linear programming algorithm is used to select the optimal transportation plan from the pre-selected transportation plans, including: Convert the required transportation cost, required transportation time and required cargo security into corresponding mathematical expressions respectively; Construct an objective function based on transportation cost, transportation time, and cargo security; The comprehensive value of each pre-selected transportation plan is calculated according to the mathematical expression of the required transportation cost, the mathematical expression of the required transportation time, the mathematical expression of the required cargo safety and the objective function, and the pre-selected transportation plan with the best comprehensive value is taken as the optimal transportation plan.

5. The international multimodal transport teaching method according to any one of claims 1 to 4, characterized in that: The generating of a plurality of initial transportation plans according to the transportation starting point and the transportation end point comprises: Get map data; Determine a route starting point and a route end point in the map data according to the transport starting point and the transport end point, perform route planning by a route planning algorithm, and obtain a plurality of transport routes; Assign multiple means of transportation to each transportation route and obtain the corresponding initial transportation plan.

6. The international multimodal transport teaching method according to any one of claims 1 to 4, characterized in that: Students are also taught about emergencies, which include the following steps: receiving simulated burst data, wherein the simulated burst data is generated according to at least one of an actual case and external real-time data; Obtaining a transportation node, where the transportation node is a current transportation location after simulated transportation according to an optimal transportation plan; A plurality of initial transport plans are regenerated according to the simulated burst data, transport nodes and transport destinations.

7. The international multimodal transport teaching method according to any one of claims 1 to 4, characterized in that: Corresponding practice tasks and prompts are also generated according to the students' operation behaviors, which specifically include the following steps: Record students' operation behaviors in real time, including: solution selection, operation steps, error frequency and completion time; The operation steps with an error frequency greater than a frequency threshold are regarded as difficult steps, the types of schemes selected with an error frequency greater than the frequency threshold are regarded as difficult types, and the types of schemes selected with a completion time greater than a time threshold are regarded as timeout types; Issue the students with practice tasks of difficulty type and timeout type, and send prompts about difficult steps during the students' practice.

8. International multimodal transport teaching system, characterized by: include: An acquisition and analysis module is used to obtain a transportation starting point, a transportation end point and transportation demand, analyze the transportation demand and obtain demand characteristic information; A generating and calculating module is used to generate a plurality of initial transportation plans according to the transportation starting point and the transportation end point, and calculate and obtain the plan characteristic information of each initial transportation plan; A scheme evaluation module, used to use the demand characteristic information as a constraint condition of a decision tree model, input the scheme characteristic information of each initial transportation scheme into the decision tree model, and obtain a corresponding pre-selected transportation scheme; A scheme selection module, used for selecting the optimal transportation scheme from the pre-selected transportation schemes by using a linear programming algorithm; A scheme comparison module is used to determine a student selection scheme from the initial transportation scheme according to the scheme selection instruction issued by the student, and compare the student selection scheme with the optimal transportation scheme to obtain a comparison result; According to the output module, it is used to obtain the decision basis of the pre-selected transportation plan according to the decision path of the decision tree model and the information entropy of each node, obtain the selection basis of the optimal transportation plan according to the calculation process of the linear programming algorithm, and output the comparison result, decision basis and selection basis.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images