Horizontal transportation automatic management system for combined transportation of iron and water

Abstract

The invention relates to the field of intelligent management, and discloses a horizontal transportation automatic management system for combined transportation, which comprises a data acquisition module, a knowledge graph construction module, a transportation planning module, a transportation monitoring module, a transportation deviation identification module and a transportation regulation and control module, the railway station position information, the waterway port attribute information, the transportation tool state information and the cargo feature information are gathered into initial data of the railway-iron combined transportation network; constructing a knowledge graph; performing constraint reasoning in the knowledge graph to obtain an initial transportation plan of the transportation demand; state data of the combined transportation network after the initial transportation plan is executed are monitored in real time; evaluating a current transportation state and a plan deviation of the railway-iron combined transportation network, and traversing key influence entities and relationships in the initial transportation plan; dynamically adjusting the initial transportation plan, and automatically scheduling the molten iron combined transportation network; according to the invention, the efficiency of automatic management of horizontal transportation for combined transportation of iron and water can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent management, and in particular to a horizontal transportation automation management system for water-rail intermodal transport. BACKGROUND

[0002] In the field of horizontal transportation management of water-rail intermodal transport, the existing technology generally has the problem of insufficient data integration capability. Railway station location information, waterway port attribute information, transportation tool state information and cargo characteristic information are often stored in independent systems or databases, lacking a unified collection and fusion mechanism, and unable to form initial data that can comprehensively reflect the overall state of the water-rail intermodal transport network. This data fragmentation makes it difficult to obtain complete and accurate basic support when making transportation plans, resulting in a reliance on experience in the planning process and an inability to fully combine multi-dimensional constraint conditions, thereby affecting the rationality and adaptability of the transportation plan.

[0003] At the same time, the existing technology has obvious shortcomings in dynamic management and deviation control during transportation. On the one hand, real-time state monitoring of the water-rail intermodal transport network after executing the transportation plan is not comprehensive, making it difficult to synchronously obtain key data such as transportation tool location, cargo state and node operation conditions; on the other hand, even if deviations are found between the transportation state and the plan, there is a lack of deviation evaluation capability based on network entity and relationship association analysis, making it impossible to accurately identify the chain effect of the deviation on the transportation network and the key influencing factors, resulting in delayed transportation plan adjustment and lack of targeted adjustment scheme, ultimately leading to low transportation efficiency and difficulty in meeting the management needs of water-rail intermodal transport horizontal transportation automation and efficiency. SUMMARY

[0004] The present application provides a horizontal transportation automation management system for water-rail intermodal transport to solve the problems raised in the background art.

[0005] To achieve the above-mentioned purpose, the present application provides a horizontal transportation automation management system for water-rail intermodal transport, characterized in that the system comprises a data acquisition module, a knowledge graph construction module, a transportation planning module, a transportation monitoring module, a transportation deviation identification module and a transportation control module, wherein:

[0006] The data acquisition module is used to collect railway station location information, waterway port attribute information, transportation tool state information and cargo characteristic information into initial data of the water-rail intermodal transport network;

[0007] The knowledge graph construction module is configured to construct a knowledge graph of the rail-water intermodal network by taking railway stations and waterway ports in the initial data as entity nodes and taking connection relationships, transportation capacity constraints and time-dependent attributes in the initial data as relationship edges.

[0008] The transportation planning module is configured to perform constraint reasoning in the knowledge graph according to transportation demand to obtain an initial transportation plan of the transportation demand.

[0009] The transportation monitoring module is configured to monitor state data of the rail-water intermodal network in real time after the initial transportation plan is executed.

[0010] The transportation deviation identification module is configured to evaluate a current transportation state of the rail-water intermodal network and a plan deviation based on the state data, and traverse key influence entities and relationships in the initial transportation plan according to the current transportation state and the plan deviation.

[0011] The transportation regulation module is configured to dynamically adjust the initial transportation plan based on the key influence entities and relationships, and send automatic scheduling of the rail-water intermodal network according to the initial transportation plan after the dynamic adjustment.

[0012] In a preferred embodiment, when the data collection module executes the collection of railway station location information, waterway port attribute information, transportation tool state information and cargo characteristic information into initial data of the rail-water intermodal network, it is specifically configured to:

[0013] connect the railway station location information to obtain a station data set of the rail-water intermodal network;

[0014] extract waterway port attribute information from the waterway port data to obtain a port data set of the rail-water intermodal network;

[0015] obtain a tool state data set of the rail-water intermodal network from the transportation tool monitoring platform;

[0016] retrieve cargo characteristic information from the cargo management database to obtain a cargo data set of the rail-water intermodal network;

[0017] fuse the station data set, the port data set, the tool state data set and the cargo data set to obtain the initial data of the rail-water intermodal network.

[0018] In a preferred implementation, when performing constructing the knowledge graph of the rail-water intermodal network by taking the railway stations and waterway ports in the initial data as entity nodes, and taking the connection relationships, transportation capacity constraints and time-dependent attributes in the initial data as relationship edges, the knowledge graph construction module is specifically configured to:

[0019] extracting the railway station entities and waterway port entities in the initial data, assigning a unique identifier to each of the entities and adding a type attribute, to obtain a set of entity nodes of the rail-water intermodal network;

[0020] identifying the connection relationships between the entity nodes in the initial data, parsing the transportation capacity constraint parameters and time-dependent attribute parameters corresponding to the connection relationships, to obtain a set of relationship edges of the rail-water intermodal network;

[0021] performing graph structure assembly on the set of entity nodes and the relationship edges, to obtain a topological structure of the rail-water intermodal network;

[0022] associating the transportation capacity constraint parameters and the time-dependent attribute parameters to the corresponding relationship edges in the topological structure, to obtain a knowledge graph model of the rail-water intermodal network;

[0023] storing the knowledge graph model to a graph database, and establishing a quick query channel for entities and relationships through an index interface of the graph database, to obtain the knowledge graph of the rail-water intermodal network.

[0024] In a preferred implementation, when performing associating the transportation capacity constraint parameters and the time-dependent attribute parameters to the corresponding relationship edges in the topological structure, to obtain the knowledge graph model of the rail-water intermodal network, the knowledge graph construction module is specifically configured to:

[0025] extracting the maximum passing capacity and current load rate features in the transportation capacity constraint parameters, to obtain a set of capacity constraint attributes of the rail-water intermodal network;

[0026] extracting the planned transportation time and real-time delay features in the time-dependent attribute parameters, to obtain a set of time attributes of the rail-water intermodal network;

[0027] binding the transportation capacity constraint information of the set of capacity constraint attributes to the corresponding relationship edges in the topological structure, to obtain a first attribute edge of the rail-water intermodal network;

[0028] binding the time-dependent information in the set of time attributes to the corresponding relationship edges in the topological structure, to obtain a second attribute edge of the rail-water intermodal network;

[0029] The first attribute edge and the second attribute edge are image encapsulated to obtain a knowledge graph model of the intermodal network.

[0030] In a preferred embodiment, the transport planning module, when performing constraint reasoning in the knowledge graph according to the transport demand to obtain an initial transport plan of the transport demand, is specifically configured to:

[0031] The starting point position, the ending point position, the cargo type, and the expected arrival time feature in the transport demand are parsed to generate a structured transport demand of the transport demand.

[0032] In the knowledge graph, the starting point position and the ending point position are taken as query nodes, and all connected paths are found through a graph traversal operation to obtain a candidate path set of the transport demand.

[0033] The matching of the transport capacity constraint attribute carried on each path relationship edge in the candidate path set and the cargo type is verified to obtain a capacity-compliant path meeting the transport capacity requirement.

[0034] The consistency of the time-dependent attribute carried on the relationship edge in the capacity-compliant path and the expected arrival time is verified to obtain a time-compliant path meeting the time requirement.

[0035] The path with the optimal comprehensive transport cost is selected from the time-compliant path as the initial transport plan of the transport demand.

[0036] In a preferred embodiment, the transport monitoring module, when performing real-time monitoring of the state data of the intermodal network after the initial transport plan is executed, is specifically configured to:

[0037] Tool position data of a transport tool in the initial transport plan is obtained through a positioning device.

[0038] Running parameters of the transport tool are collected through a device state sensor to obtain tool state data of the transport tool.

[0039] The current state of the cargo in the transport plan is obtained to obtain state data of the cargo.

[0040] Node state data of a port-station operation collection node is obtained through an operation of the port-station operation collection node.

[0041] The tool position data, the tool state data, the state data, and the node state data are integrated to obtain the state data of the intermodal network.

[0042] In a preferred implementation, the transport deviation identification module, when performing the evaluation of the current transport state and the planned deviation of the intermodal network based on the state data, is specifically configured to:

[0043] map the state data to the corresponding entity nodes and relationship edges in the knowledge graph to obtain a state-enhanced knowledge graph;

[0044] In the state-enhanced knowledge graph, traverse the sequence of entity nodes involved in the initial transport plan;

[0045] extract the actual state attributes and planned attributes of the sequence of entity nodes to obtain node-level deviation data of the initial transport plan;

[0046] based on the transport capacity constraint attributes and time-dependent attributes carried on the relationship edges, identify the path validity changes in the initial transport plan caused by state changes to obtain path-level deviation data of the initial transport plan;

[0047] track the chain effects of deviations on individual nodes and edges on the entire transport network through the relationship topology of the knowledge graph to obtain key influence propagation paths of the initial transport plan;

[0048] perform multi-level correlation analysis on the node-level deviation data, the path-level deviation data, and the key influence propagation paths to obtain the current transport state and the planned deviation of the initial transport plan.

[0049] In a preferred implementation, the transport deviation identification module, when performing the evaluation of the current transport state and the planned deviation of the initial transport plan based on the state data, is specifically configured to:

[0050] generate node-path correlation results of the initial transport plan according to the influence degree of individual node deviations in the node-level deviation data and the path-level deviation data on the path connectivity in the path-level deviation data;

[0051] generate path-network correlation results of the initial transport plan according to the influence range of path-level deviations on the network stability in the key influence propagation paths in the path-level deviation data and the key influence propagation paths;

[0052] superimpose the node-path correlation results and the path-network correlation results to obtain a deviation severity level of the initial transport plan;

[0053] Compiling the deviation severity level and the key influence propagation path, a current transportation state of the initial transportation plan and a plan deviation are obtained.

[0054] In a preferred embodiment, when performing the traversal of the key influence entities and relationships in the initial transportation plan according to the current transportation state and the plan deviation, the transportation deviation identification module is specifically configured to:

[0055] Based on the deviation severity level corresponding to the current transportation state and the plan deviation, an affected entity node is identified in the knowledge graph, and an initial influence entity set of the initial transportation plan is obtained.

[0056] According to the transportation capacity constraint attribute carried on the relationship edge in the knowledge graph, an entity node with a transportation bottleneck feature in the initial influence entity set is screened, and a capacity influence entity subset of the initial transportation plan is obtained.

[0057] According to the time dependency attribute carried on the relationship edge in the knowledge graph, an entity node with a time criticality feature in the initial influence entity set is screened, and a time influence entity subset of the initial transportation plan is obtained.

[0058] The relationship topology of the capacity influence entity subset and the time influence entity subset is traced back in the knowledge graph, a relationship edge directly connected with the entity node and affecting transportation continuity is identified, and a key influence relationship set of the initial transportation plan is obtained.

[0059] The key influence entities and relationships of the initial transportation plan in the capacity influence entity subset, the time influence entity subset and the key influence relationship set are determined according to importance ranking.

[0060] In a preferred embodiment, when performing the dynamic adjustment of the initial transportation plan based on the key influence entities and relationships, the transportation regulation module is specifically configured to:

[0061] For the key influence entities, a replaceable transportation node is queried in the knowledge graph, and an alternative node set of the initial transportation plan is obtained.

[0062] For the relationship, a replacement path satisfying the transportation capacity constraint and the time dependency attribute is found in the knowledge graph, and an alternative path set of the initial transportation plan is obtained.

[0063] Based on the alternative node set and the alternative path set, the initial transportation plan is updated, and a dynamically adjusted initial transportation plan is generated.

[0064] Compared with the prior art, the present application has the following beneficial effects:

[0065] 1. The horizontal transportation automation management system for water-rail intermodal transport according to the present application, through the data acquisition module, multi-dimensional information such as railway site location, waterway port attribute, transportation tool state and cargo characteristics is collected and fused to form complete water-rail intermodal transport network initial data; then through the knowledge graph construction module, network topology is built with entity nodes and relationship edges, and key parameters such as transportation capacity constraint and time dependent attribute are associated to provide accurate and comprehensive basic support for transportation planning. Based on the knowledge graph, constraint reasoning is performed to generate an initial transportation plan, which can fully meet the location, cargo type and time requirements in the transportation demand, significantly improve the scientificity and adaptability of the initial transportation plan, and lay a solid foundation for subsequent efficient transportation.

[0066] 2. The system according to the present application, through the transportation monitoring module, network state data such as transportation tool position, running parameter, cargo state and node operation condition is acquired in real time to ensure comprehensive perception of the transportation process; the transportation deviation identification module can analyze the deviation between the current transportation state and the plan in multiple levels, accurately locate key influence entities and relationships, and avoid the expansion of the chain influence of the deviation on the transportation network; the transportation regulation module dynamically adjusts the transportation plan based on key influencing factors and realizes automatic scheduling, effectively guarantees the continuity and stability of the transportation process, and greatly improves the efficiency of the horizontal transportation automation management for water-rail intermodal transport, and enhances the real-time and flexibility of the management process. BRIEF DESCRIPTION OF DRAWINGS

[0067] Figure 1 The system architecture diagram of the horizontal transportation automation management system for water-rail intermodal transport according to an embodiment of the present application is provided.

[0068] The implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0069] In order to make the purpose, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments belong to part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0070] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. The singular forms "said" and "this" used in the embodiments of the present application and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise. "Multiple" generally includes at least two.

[0071] Depending on the context, the word "if" or "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0072] Furthermore, the timing of the steps in the following method embodiments is merely an example and not a strict limitation.

[0073] In practice, the server-side equipment deployed in a horizontal transport automation management system for rail-water intermodal transport may consist of one or more devices. This system can be implemented as a business instance, a virtual machine, or hardware devices. For example, it can be implemented as a business instance deployed on one or more devices in a cloud node. Simply put, it can be understood as software deployed on a cloud node, providing a horizontal transport automation management system for rail-water intermodal transport to various user terminals. Alternatively, it can be implemented as a virtual machine deployed on one or more devices in a cloud node, with application software installed to manage various user terminals. Or, it can be implemented as a server composed of numerous identical or different types of hardware devices, with one or more hardware devices configured to provide a horizontal transport automation management system for rail-water intermodal transport to various user terminals.

[0074] In terms of implementation, a horizontal transport automation management system for rail-water intermodal transport and its user terminal are mutually compatible. Specifically, if the horizontal transport automation management system for rail-water intermodal transport is implemented as an application installed on a cloud service platform, then the user terminal acts as a client establishing a communication connection with that application; or if the horizontal transport automation management system for rail-water intermodal transport is implemented as a website, then the user terminal acts as a webpage; or if the horizontal transport automation management system for rail-water intermodal transport is implemented as a cloud service platform, then the user terminal acts as a mini-program within an instant messaging application.

[0075] like Figure 1 The diagram shown is a system architecture diagram of a horizontal transport automation management system for rail-water intermodal transport provided by an embodiment of the present invention.

[0076] The horizontal transportation automation management system 100 for the iron-water combined transportation can be arranged in a cloud server, and in the implementation form, can be used as one or more service devices, or can be installed as an application on the cloud (such as a server of a mobile service operator, a server cluster, etc.), or can be developed as a website. According to the functions implemented, the horizontal transportation automation management system 100 for the iron-water combined transportation can include a data collection module 101, a knowledge graph construction module 102, a transportation planning module 103, a transportation monitoring module 104, a transportation deviation identification module 105, and a transportation regulation module 106. The modules in the present application can also be referred to as units, which refer to a series of computer program segments that can be executed by an electronic device processor and can complete a fixed function, and are stored in the memory of the electronic device.

[0077] In the embodiment of the present application, each of the above modules can be independently implemented and called by other modules. Here, calling can be understood as that a module can connect a plurality of modules of another type and provide corresponding services for the plurality of modules connected thereby. In the horizontal transportation automation management system for the iron-water combined transportation provided in the embodiment of the present application, without modifying the program code, the application range of the horizontal transportation automation management system architecture for the iron-water combined transportation can be adjusted in the form of increasing modules and direct calling, cluster horizontal expansion is realized, so as to achieve the purpose of quickly and flexibly expanding the horizontal transportation automation management system for the iron-water combined transportation. In actual application, the above modules can be arranged in the same device or different devices, or can be arranged in a virtual device, such as a service instance in a cloud server.

[0078] The following will be described in combination with specific embodiments, respectively for each component and specific work flow of the horizontal transportation automation management system for the iron-water combined transportation:

[0079] The data collection module 101 is used to collect railway station location information, waterway port attribute information, transportation tool state information, and cargo characteristic information as initial data of the iron-water combined transportation network.

[0080] In the embodiment of the present application, when the data collection module performs the collection of the railway station location information, the waterway port attribute information, the transportation tool state information, and the cargo characteristic information as the initial data of the iron-water combined transportation network, it is specifically used for:

[0081] Connecting topology is performed on the railway station location information to obtain a station data set of the iron-water combined transportation network;

[0082] The waterway port attribute information is extracted from the waterway port data to obtain a port data set of the iron-water combined transportation network;

[0083] Obtaining the transport tool state information from the transport tool monitoring platform to obtain a tool state data set of the rail-water combined transport network;

[0084] Retrieving the cargo feature information from the cargo management database to obtain a cargo data set of the rail-water combined transport network;

[0085] Data fusion is performed on the station data set, the port data set, the tool state data set and the cargo data set to obtain initial data of the rail-water combined transport network.

[0086] Specifically, all railway station location information is called from the station database of the railway operation management system, including the unique identification of each station, latitude and longitude, the name and number of the line to which the station belongs; the connection relationship of the station is determined in combination with the line laying and operation scheduling records to form a station data set of the rail-water combined transport network.

[0087] Further, the waterway port data is obtained from the port information system, and the following attributes need to be extracted: port name, regional latitude and longitude, official level, total number of berths, maximum berthing tonnage of single berth, number of types of loading and unloading equipment, and types of cargo that can be received; the corresponding information is screened and irrelevant content is removed to form a structured data set, which is the port data set.

[0088] Further, the core transport tools of the rail-water combined transport are freight trains and ships, and real-time state data is called at fixed intervals through a special interface of the monitoring platform, including the unique number of the tool, the current position, the speed, the actual load and volume, the power system parameters, and the braking system status; when the data is missing or abnormal, the technical personnel are contacted for verification and supplement, and the structured data set formed after classification and arrangement is the tool state data set.

[0089] Further, the following cargo features need to be retrieved: waybill number, cargo name, type, weight, volume, packaging method, special transportation requirements, origin, destination, and current transportation stage; the cargo information meeting the conditions is extracted by searching in the cargo management database according to the "waybill associated with rail-water combined transport"; when the special transportation requirements field is empty, the management department is contacted for supplement, and the structured data set formed is the cargo data set.

[0090] Further, the data fusion rules are first formulated: in terms of association rules, the origin and destination of the cargo are matched with the railway station name of the station data and the port name of the port data respectively to ensure that the transportation path corresponds to specific nodes; the current transportation stage of the cargo is matched with the tool number of the tool state data to determine the carrying tool. In terms of integration rules, the same timestamp is added based on the collection of unified time, and the position coordinate formats of the station, the port and the tool are calibrated according to the geographical coverage range.

[0091] Further, the site and port data are then integrated according to geographic coordinates to form a basic data set containing all transport nodes; the tool state data is matched with the basic data set to mark the current node or inter-node position of the tool; the cargo data is matched with the basic data set and the tool state data respectively to supplement the corresponding node and tool information of the cargo. Finally, the data format is unified to form a complete structured data set, which is the initial data of the rail-water combined transport network.

[0092] In general, the connection topology of the railway site position information is to first call all the position information of the railway sites from the site database of the railway operation management system, including the unique identification, latitude and longitude, and the railway line to which the site belongs, then determine the connection relationship between the sites according to the actual laying situation and operation scheduling record of the railway line, and finally arrange the site information with the connection relationship according to a specific structure to obtain the site data set of the rail-water combined transport network.

[0093] In general, the extraction of waterway port attribute information from the waterway port data is to first obtain complete waterway port data from the port information system of the port management department, covering basic archives, operation records, equipment information, etc., then determine the port attribute categories to be extracted and select the corresponding information therefrom, eliminate irrelevant contents, and finally arrange the selected attribute information according to a specified format to obtain the port data set of the rail-water combined transport network.

[0094] In general, the acquisition of transport tool state information from the transport tool monitoring platform is to first determine the range of transport tools such as freight trains and freight ships involved in the rail-water combined transport, then call the real-time state data of these tools at fixed intervals through the special data interface of the monitoring platform, check and supplement the data without errors, arrange the state information according to a specific structure, and obtain the tool state data set of the rail-water combined transport network.

[0095] In general, the retrieval of cargo characteristic information from the cargo management database is to first determine the cargo characteristic items to be retrieved, then perform retrieval in the cargo management database with the condition of "waybill associated transport mode containing rail-water combined transport", extract the cargo information meeting the condition and check and supplement the completeness, and finally arrange the complete cargo characteristic information according to a specified format to obtain the cargo data set of the rail-water combined transport network.

[0096] In general, the data fusion of the four types of data sets is to first develop association and integration rules, then spatially integrate the site and port data sets, match the tool state data set with the basic data set, and match the cargo data set with the relevant data sets, finally unify the format of all matched data and delete duplicate records, and finally obtain the initial data of the rail-water combined transport network.

[0097] The knowledge graph construction module 102 is configured to take railway stations and waterway ports in the initial data as entity nodes, take connection relationships, transportation capacity constraints and time-dependent attributes in the initial data as relationship edges, and construct a knowledge graph of the rail-water combined transport network.

[0098] In the embodiment of the present application, when the knowledge graph construction module is used to construct a knowledge graph of the rail-water combined transport network by taking railway stations and waterway ports in the initial data as entity nodes and taking connection relationships, transportation capacity constraints and time-dependent attributes in the initial data as relationship edges, the knowledge graph construction module is specifically configured to:

[0099] extract railway station entities and waterway port entities in the initial data, assign a unique identifier to each of the entities and add a type attribute, and obtain a set of entity nodes of the rail-water combined transport network;

[0100] identify connection relationships between the entity nodes in the initial data, analyze transportation capacity constraint parameters and time-dependent attribute parameters corresponding to the connection relationships, and obtain a set of relationship edges of the rail-water combined transport network;

[0101] assemble the set of entity nodes and the set of relationship edges into a graph structure, and obtain a topological structure of the rail-water combined transport network;

[0102] associate the transportation capacity constraint parameters and the time-dependent attribute parameters to corresponding relationship edges in the topological structure, and obtain a knowledge graph model of the rail-water combined transport network;

[0103] store the knowledge graph model in a graph database, establish a quick query channel for entities and relationships through an index interface of the graph database, and obtain a knowledge graph of the rail-water combined transport network.

[0104] When the knowledge graph construction module is used to associate the transportation capacity constraint parameters and the time-dependent attribute parameters to corresponding relationship edges in the topological structure, and obtain a knowledge graph model of the rail-water combined transport network, the knowledge graph construction module is specifically configured to:

[0105] extract maximum passing capacity and current load rate features in the transportation capacity constraint parameters, and obtain a set of capacity constraint attributes of the rail-water combined transport network;

[0106] extract planned transportation time and real-time delay features in the time-dependent attribute parameters, and obtain a set of time attributes of the rail-water combined transport network;

[0107] bind transportation capacity constraint information of the set of capacity constraint attributes to corresponding relationship edges in the topological structure, and obtain a first attribute edge of the rail-water combined transport network;

[0108] binding time dependent information in the time attribute set with corresponding relation edges in the topology structure to obtain a second attribute edge of the combined railway-waterway transportation network;

[0109] performing image packaging on the first attribute edge and the second attribute edge to obtain a knowledge graph model of the combined railway-waterway transportation network.

[0110] Specifically, first, all information related to railway stations and waterway ports in the initial data is sorted out, including the names, cities, and operation numbers of railway stations, and the names, port areas, and berth numbers of waterway ports. These information is verified one by one to exclude duplicate records and invalid information, and to determine each valid railway station entity and waterway port entity. A type attribute is added to each entity, and all railway station entities and waterway port entities that have completed unique identifier allocation and type attribute addition are integrated to form an entity node set of the combined railway-waterway transportation network.

[0111] Further, the association content between the entity nodes recorded in the initial data is viewed, including the cargo transfer records between railway stations and waterway ports and fixed transportation line annotations, and according to these contents, it is determined which entity nodes have a connection relationship. For each pair of entity nodes determined to have a connection relationship, the transportation capacity related content corresponding to the connection relationship is extracted from the initial data, and these contents are sorted into transportation capacity constraint parameters; at the same time, the time related content corresponding to the connection relationship is extracted from the initial data, and these contents are sorted into time dependent attribute parameters. For each pair of entity nodes having a connection relationship and having parsed transportation capacity constraint parameters and time dependent attribute parameters, a relation edge containing the two entity nodes, corresponding transportation capacity constraint parameters, and time dependent attribute parameters is constructed, and all constructed relation edges are integrated to form a relation edge set of the combined railway-waterway transportation network.

[0112] Further, a graph structure is determined to be constructed in the form of node-edge association, each entity node in the entity node set is taken as an independent node in the graph structure, and these nodes are labeled in the graph structure in turn, and the labeling content includes the name and unique identifier of the entity node. According to each relation edge in the relation edge set, the two entity nodes connected by the relation edge are found, the labeling positions of the two entity nodes in the graph structure are found, and the two nodes are connected by a line segment, and the connection relationship type corresponding to the relation edge is labeled on the line segment. After completing the labeling of all entity nodes and the connection of all relation edges, it is checked whether there is a problem in the graph structure, and after correcting all problems, a graph structure containing all entity nodes and correct connection relationships between entity nodes is formed, which is the topology structure of the combined railway-waterway transportation network.

[0113] Further, based on the topological structure of the water-iron combined transport network as a framework, each relationship edge in the topological structure is checked one by one, the transport capacity constraint parameters and the time-dependent attribute parameters corresponding to the relationship edge are found from the relationship edge set, and the transport capacity constraint parameters and the time-dependent attribute parameters are directly marked on the corresponding relationship edge line segment in the topological structure, so as to ensure that each relationship edge is accurately accompanied by its corresponding transport capacity constraint parameters and time-dependent attribute parameters. After completing the marking of all parameters, the knowledge graph model of the water-iron combined transport network is formed. Select a graph database, and according to the data storage specification of the graph database, the entity nodes, relationship edges and the parameters attached to the relationship edges in the knowledge graph model are entered into the graph database one by one, and the data is checked in real time during the entering process to ensure that there is no omission or error. After the entering is completed, an index is established for the unique identifier of the entity node using the index interface provided by the graph database, so that when the unique identifier of a certain entity node is input, the entity node can be quickly located; at the same time, an index is established for the connection relationship type of the relationship edge, so that when a certain connection relationship type is queried, all relationship edges of the type can be quickly found. Through the establishment of the two types of indexes, a quick query channel of entities and relationships is formed, and finally the knowledge graph of the water-iron combined transport network is obtained.

[0114] Specifically, from the obtained transport capacity constraint parameters, information related to the maximum passing capacity is selected one by one, specifically the maximum freight transport volume of each relationship edge corresponding to the water-iron combined transport line in unit time, the maximum number of freight units that can be accommodated in a single transport, etc. These information are directly taken from the explicit records of the transport line capacity in the initial data; then the current load rate feature is extracted from the transport capacity constraint parameters; subsequently, the maximum passing capacity information of each relationship edge is paired with the current load rate feature information to form the capacity constraint attribute items in units of relationship edges, and after all the items are integrated, the capacity constraint attribute set of the water-iron combined transport network is obtained.

[0115] Further, from the time-dependent attribute parameters, the planned transport time information corresponding to each relationship edge is first located, which is the standard duration of the freight from the starting entity node to the target entity node connected by the relationship edge recorded in the initial data, and it is required to ensure that the duration is a fixed planned value explicitly marked in the initial data; then the real-time delay feature is extracted from the time-dependent attribute parameters, specifically the difference between the actual transport time and the planned transport time of each relationship edge corresponding to the line in the last regular operation cycle, and the specific operation date when the difference occurs, each delay data corresponds to a corresponding transport record; then the planned transport time information of each relationship edge is combined with the real-time delay feature information to form the time attribute items in units of relationship edges, and after all the items are integrated, the time attribute set of the water-iron combined transport network is obtained.

[0116] Further, first, each relationship edge in the iron-water combined transport network topology structure is uniquely identified and marked, and the marking content is the combination of the starting and target entity node names corresponding to the relationship edge, which is used as a unique identifier to ensure that each identifier is unique and can clearly distinguish different relationship edges in the topology structure. Then, find the capacity constraint attribute item corresponding to each relationship edge identifier from the capacity constraint attribute set, and add the maximum transport capacity, current load rate and other transport capacity constraints contained in the item to the corresponding relationship edge in the topology structure in the form of text annotation, which is close to the relationship edge and does not block other elements, ensuring that the information is clear and visible. After all the relationship edges have completed the transport capacity constraint information annotation, the relationship edges with capacity constraint information in the topology structure are the first attribute edges of the iron-water combined transport network.

[0117] Further, first, confirm the unique identifier of each relationship edge in the iron-water combined transport network topology structure; then, select the time attribute item matching the identifier of each relationship edge from the time attribute set, and add the planned transport time, real-time delay and other time-dependent information contained in the item to the other side of the corresponding relationship edge in the topology structure in the form of text annotation, which is on the other side of the relationship edge from the annotated transport capacity constraint information to avoid information overlap; after all the relationship edges have been accurately annotated with time-dependent information, the relationship edges with time-dependent information in the topology structure are the second attribute edges of the iron-water combined transport network.

[0118] Further, first, integrate the topology structure elements corresponding to the first attribute edges and the second attribute edges into the same image drawing area, and ensure that the entity node positions of the two types of attribute edges are completely overlapped and the relationship edge lines are overlapped with each other to form a unified and conflict-free graph structure basis; then, format the transport capacity constraint information and time-dependent information on each overlapped relationship edge; finally, save the entire graph structure in image format, select an industry- common and high-definition format, and check and ensure that all entity nodes, relationship edges and annotation information in the image are clear and distinguishable, without blurring, missing or misplacement. The integrated, formatted, added legend and saved image graph structure is the knowledge graph model of the iron-water combined transport network.

[0119] Overall, extracting railway station entities and waterway port entities from the initial data is the core of this step, and in this process, each extracted entity is assigned a unique identifier, and each entity is added with a corresponding type attribute. Through these operations, the entity node set of the iron-water combined transport network is finally integrated.

[0120] Overall, this step focuses on the time-dependent attribute parameters, and extracts the planned transportation time and real-time delay features. After systematic analysis and summary of these two types of features, the time attribute set of the iron-water combined transport network is obtained.

[0121] Overall, this step mainly integrates the obtained entity node set and relationship edge, and reasonably assembles the entity node and relationship edge according to the construction logic of the graph structure, ensures the accurate matching of the entity node and the corresponding relationship edge, and finally forms the topology structure of the iron-water combined transport network.

[0122] Overall, this step takes the topology structure of the iron-water combined transport network as the basic framework, and accurately associates the previously parsed transportation capacity constraint parameters and time-dependent attribute parameters with the corresponding relationship edges in the topology structure, so that each relationship edge is attached with corresponding parameter information. Through this association operation, the knowledge graph model of the iron-water combined transport network is finally obtained.

[0123] Overall, this step first stores the constructed knowledge graph model into the designated graph database, and then uses the index interface provided by the graph database to establish a special fast query channel for entities and relationships. Through a series of storage and index establishment operations, the complete knowledge graph of the iron-water combined transport network is finally obtained.

[0124] Overall, this step focuses on the transportation capacity constraint parameters, and extracts the maximum passing capacity and current load rate features. Through accurate screening and integration of these two types of features, the capacity constraint attribute set of the iron-water combined transport network is finally formed.

[0125] Overall, this step focuses on the time-dependent attribute parameters, and extracts the planned transportation time and real-time delay features. After systematic analysis and summary of these two types of features, the time attribute set of the iron-water combined transport network is obtained.

[0126] Overall, this step takes the topology structure as the basis, and accurately binds the transportation capacity constraint information contained in the capacity constraint attribute set with the corresponding relationship edges in the topology structure. After the binding is completed, the relationship edges in the topology structure that are attached with transportation capacity constraint information are the first attribute edges of the iron-water combined transport network.

[0127] Overall, this step relies on the topology structure to bind the time-dependent information contained in the time attribute set with the corresponding relationship edges in the topology structure. After the binding is completed, the relationship edges in the topology structure that are attached with time-dependent information are the second attribute edges of the iron-water combined transport network.

[0128] In general, this step is specially image packaging processing for the acquired first attribute edge and second attribute edge, and the two types of attribute edges are formed into a unified image structure through packaging. The image structure after completing the packaging is the knowledge graph model of the intermodal network.

[0129] The transportation planning module 103 is configured to perform constraint reasoning in the knowledge graph according to the transportation demand, to obtain an initial transportation plan of the transportation demand.

[0130] In the embodiment of the present application, when the transportation planning module performs constraint reasoning in the knowledge graph according to the transportation demand to obtain an initial transportation plan of the transportation demand, it is specifically configured to:

[0131] Analyzing the starting point position, the end point position, the cargo type and the expected arrival time feature in the transportation demand, to generate a structured transportation demand of the transportation demand;

[0132] In the knowledge graph, taking the starting point position and the end point position as query nodes, finding all connected paths through graph traversal operation to obtain a candidate path set of the transportation demand;

[0133] Verifying the matching of the transportation capacity constraint attribute carried on each path relationship edge in the candidate path set with the cargo type, to obtain a capacity-compliant path meeting the transportation capacity requirement;

[0134] Verifying the consistency of the time-dependent attribute carried on the relationship edge in the capacity-compliant path with the expected arrival time, to obtain a time-compliant path meeting the time requirement;

[0135] Selecting a path with the optimal comprehensive transportation cost from the time-compliant path as the initial transportation plan of the transportation demand.

[0136] Specifically, the starting point position information is extracted from the transportation demand description. According to the keywords such as “railway station”, “train station” or “port” and “harbor” in the demand, it is determined whether the starting point is a railway station or a waterway port entity and the specific name is recorded. Then, the end point position information is extracted according to the same keyword identification method, the end point type is determined and the name is recorded. Then, the cargo type information is extracted, and the product category is recorded according to the clear labeling of the demand. Then, the expected arrival time feature is extracted, and the specific arrival date and time required by the demand is recorded. Finally, the extracted starting point position, end point position, cargo type and expected arrival time are arranged in the fixed format of “starting point-end point-cargo type-expected arrival time”, to form a structured information set, which is the structured transportation demand of the transportation demand.

[0137] Further, in the knowledge graph of the iron water intermodal network, first, the corresponding entity nodes are located according to the starting point and the ending point name of the structured transportation demand; starting from the starting point node, all the adjacent entity nodes are visited in turn along the relationship edges directly connected to the node, and the path segment of the starting point node and the current adjacent node is recorded every time an adjacent node is visited; continue to start from the adjacent node, visit the next entity node along all untraversed relationship edges, and record the new path segment at the same time, until the ending entity node is reached, at which time all the path segments are integrated into a complete path and saved; if an entity node that has been visited is encountered during traversal, the branch traversal is immediately stopped to avoid a circular path; repeat the above-mentioned traversal starting from the starting point node until all possible branch traversals are completed, all saved complete connected paths are collected, and a candidate path set of the transportation demand is formed.

[0138] Further, one path in the candidate path set is taken out, and the transportation capacity constraint attribute carried by each relationship edge of the path is checked one by one; according to the type of the structured transportation demand, the transportation capacity requirement of the goods is determined; first, it is checked whether the maximum passing capacity of each relationship edge in the path meets the transportation capacity requirement of the goods, and then the remaining transportation capacity is calculated according to the current load rate to confirm that the remaining transportation capacity can completely accommodate the transportation capacity of the goods; if all the relationship edges of the path meet the requirement, it is indicated that the transportation capacity constraint attribute matches the type of the goods, and the path is marked as a capacity-compliant path; all the paths in the candidate path set are verified one by one in the above-mentioned manner, all the paths marked as capacity-compliant paths are collected, and the capacity-compliant paths that meet the transportation capacity requirement are obtained.

[0139] Further, a capacity-compliant path is selected, and the time-dependent attribute carried by each relationship edge of the path is extracted one by one; the total expected transportation time of the path is calculated: first, the planned transportation time of each relationship edge is accumulated to obtain the basic transportation time; then, according to the real-time delay characteristics of each relationship edge, if the fixed delay time of the edge in the latest operation cycle is displayed, the fixed delay time is added to the planned transportation time of the corresponding edge, and the delay time of all edges is summed to obtain the total delay time; the basic transportation time and the total delay time are added to obtain the total expected arrival time of the path from the starting point to the ending point; the total expected arrival time is compared with the expected arrival time of the structured transportation demand, if the total expected arrival time is earlier than or equal to the expected arrival time, it is indicated that the time-dependent attribute is consistent with the expected arrival time, and the path is marked as a time-compliant path; all the capacity-compliant paths are verified according to the above-mentioned process, all the paths marked as time-compliant paths are collected, and the time-compliant paths that meet the time requirement are obtained.

[0140] Further, it is determined that the comprehensive transportation cost is composed of transportation fees and transfer fees, the transportation fees are calculated according to the transportation unit price of each section of the time-compliant path, and the transfer fees are calculated for the transfer nodes in the path, each transfer node including the goods loading and unloading fees and temporary storage fees, the single transfer fee of the node is calculated according to the charging standard, and the total transfer fee is obtained by summarizing the fees of all transfer nodes; the total transportation fee and the total transfer fee of each time-compliant path are added to obtain the comprehensive transportation cost thereof; and the comprehensive transportation costs of all time-compliant paths are compared to screen out the path with the lowest cost as the initial transportation plan of the transportation demand.

[0141] In general, this step focuses on the core information analysis of the transportation demand, and the scattered demand information is arranged in a structured form by extracting and sorting the start point position, end point position, goods type and expected arrival time characteristics in the demand, and finally a structured transportation demand of the transportation demand is generated.

[0142] In general, this step takes the start point position and end point position in the structured transportation demand as the query node in the knowledge graph, and all possible connected paths from the start point node to the end point node are checked one by one by means of graph traversal operation, and the path set is formed after the paths are summarized.

[0143] In general, this step checks whether the transportation capacity constraint properties carried by each relationship edge on each path in the candidate path set match the goods type one by one, and all paths meeting the transportation capacity requirements are retained after verification and screening to obtain the capacity-compliant paths.

[0144] In general, this step verifies the time dimension around the capacity-compliant paths, confirms whether the time-dependent properties carried by each relationship edge on each path are consistent with the expected arrival time one by one, and screens out the paths meeting the time requirements after verification to obtain the time-compliant paths.

[0145] In general, this step compares and analyzes the comprehensive transportation costs of the time-compliant paths, selects the path with the optimal comprehensive transportation cost, and determines the path as the initial transportation plan of the transportation demand.

[0146] The transportation monitoring module 104 is configured to monitor the state data of the rail-water combined transportation network in real time after the initial transportation plan is executed.

[0147] In the embodiment of the present application, when the transportation monitoring module monitors the state data of the rail-water combined transportation network in real time after the initial transportation plan is executed, it is specifically configured to:

[0148] Obtain tool position data of the transportation tool in the initial transportation plan by means of the positioning device.

[0149] Collecting operation parameters of the transport tool through a device state sensor to obtain tool state data of the transport tool;

[0150] Obtaining current state of the goods in the transport plan to obtain state data of the goods;

[0151] Obtaining node state data of the port-station operation collection node through operation of the port-station operation collection node;

[0152] Integrating the tool position data, the tool state data, the state data and the node state data to obtain state data of the intermodal network.

[0153] Specifically, a GPS positioning device is pre-installed on a transport tool for executing an initial transport plan, latitude and longitude information is automatically collected at a fixed interval and a time stamp is recorded, the latitude and longitude information is associated with a transport tool identifier to ensure accurate matching of positioning data with a specific tool, abnormal data with signal interference is removed, and finally a data set containing a transport tool identifier, a collection time stamp and latitude and longitude is formed, which is tool position data of the transport tool for the initial transport plan.

[0154] Further, corresponding device state sensors are installed on key running components according to the type of the transport tool, operation parameters such as engine temperature, driving / sailing speed and remaining fuel quantity are collected in real time, and the parameters are bound with the transport tool identifier and the time stamp, and finally a data set containing the transport tool identifier, the time stamp and various operation parameters is formed, which is tool state data.

[0155] Further, temperature and humidity sensors and vibration sensors are installed on the outside of the goods package or in the container involved in the initial plan, pressure sensors are additionally installed on goods prone to damage, environmental temperature and humidity, goods vibration frequency and intensity, and pressure values of goods prone to damage are collected respectively, workers manually inspect at fixed inspection points during loading and unloading and transportation, and record the appearance integrity and loading firmness of the goods, sensor data and inspection records are integrated according to the goods batch identifier and the time stamp, repeated and contradictory records are removed, and finally a data set containing the goods batch identifier, the time stamp and various state information is formed, which is goods state data.

[0156] Further, loading and unloading equipment, storage areas and operation dispatching stations in intermodal ports and railway stations are determined as operation collection nodes, operation state sensors are installed on the loading and unloading equipment to collect data such as start-stop state, operation efficiency and fault alarm, inventory sensors are arranged in the storage areas to record the quantity, storage location and inventory turnover rate of the goods, and workers at the dispatching stations record the time of issuing operation instructions, execution progress and completion, device sensor data, inventory data and dispatching records are summarized according to the operation node identifier and the time stamp to ensure complete recording of operation of the operation node, and finally a data set containing the operation node identifier, the time stamp and operation data is formed, which is port-station node state data.

[0157] Further, a data integration uniform format is determined, containing a record timestamp, an initial transportation plan number, a transportation tool identifier, a cargo batch identifier, and a work node identifier; tool position, tool state, cargo state, and node state data are matched according to the format; missing data is supplemented through a supplement mechanism; the supplemented complete data is sorted according to the timestamp to ensure time sequence continuity, and finally a complete data set containing various integrated data is formed, which is the state data of the rail-water combined transportation network. In general, this step focuses on the transportation tools executing the initial transportation plan, and through the positioning device deployed on the transportation tools, position information of the transportation tools is collected in real time. After the position information is associated with the transportation tool identifier and the collection time, tool position data of the transportation tools executing the initial transportation plan is obtained.

[0158] In general, this step focuses on the transportation tools executing the initial transportation plan, and a device state sensor is installed on a key operating component of the transportation tools. Through the sensor, operating parameters such as engine temperature, driving speed, and remaining fuel quantity of the transportation tools are collected in real time. After the operating parameters are classified, arranged, and bound with the transportation tool identifier, tool state data of the transportation tools is obtained.

[0159] In general, this step focuses on the cargo in the transportation plan, and through a sensor installed on the cargo packaging or loading container, environmental temperature and humidity and vibration conditions of the cargo are collected. In combination with appearance inspection records of manual inspection at fixed checkpoints, after the information is integrated and associated with the cargo batch identifier, state data of the cargo is obtained.

[0160] In general, this step takes the work collection nodes of the port and the railway station as the object, and through a sensor installed on the loading and unloading equipment and the storage area, equipment operating state and inventory information are collected. In combination with instruction execution conditions recorded by the work dispatching table, after the operation related data is summarized and bound with the work node identifier, node state data of the port-station work collection node is obtained.

[0161] In general, this step matches and integrates the tool position data, the tool state data, the cargo state data, and the node state data according to a uniform timestamp and a transportation plan number, supplements missing data, and arranges in time sequence, to obtain the state data of the rail-water combined transportation network.

[0162] The transportation deviation identification module 105 is configured to evaluate a current transportation state of the rail-water combined transportation network and a plan deviation based on the state data, and traverse key influence entities and relationships in the initial transportation plan according to the current transportation state and the plan deviation.

[0163] In the embodiment of the present application, when the transport deviation identification module performs the evaluation of the current transport state and the planned deviation of the intermodal network based on the state data, it is specifically used for:

[0164] mapping the state data to the corresponding entity nodes and relationship edges in the knowledge graph to obtain a state-enhanced knowledge graph;

[0165] In the state-enhanced knowledge graph, the entity node sequence involved in the initial transport plan is traversed;

[0166] The actual state attributes and the planned attributes of the entity node sequence are extracted to obtain node-level deviation data of the initial transport plan;

[0167] Based on the transport capacity constraint attributes and the time-dependent attributes carried on the relationship edges, the path validity changes caused by state changes in the initial transport plan are identified to obtain path-level deviation data of the initial transport plan;

[0168] Through the relationship topology structure of the knowledge graph, the chain effect of the deviation of a single node and edge on the entire transport network is tracked to obtain a key influence propagation path of the initial transport plan;

[0169] The node-level deviation data, the path-level deviation data, and the key influence propagation path are subjected to multi-level correlation analysis to obtain the current transport state and the planned deviation of the initial transport plan.

[0170] When the transport deviation identification module performs the multi-level correlation analysis of the node-level deviation data, the path-level deviation data, and the key influence propagation path to obtain the current transport state and the planned deviation of the initial transport plan, it is specifically used for:

[0171] According to the influence degree of a single node deviation in the node-level deviation data and the path-level deviation data on the path connectivity in the path-level deviation data, a node-path correlation result of the initial transport plan is generated;

[0172] According to the influence range of the path-level deviation on the network stability in the key influence propagation path, a path-network correlation result of the initial transport plan is generated;

[0173] The node-path correlation result and the path-network correlation result are subjected to influence superposition to obtain a deviation severity level of the initial transport plan;

[0174] The deviation severity level and the key influence propagation path are compiled to obtain the current transport state and the planned deviation of the initial transport plan.

[0175] Specifically, the attribute classification of entity nodes and relationship edges in the water-railway combined transport network knowledge graph is determined, including location, operation efficiency, transportation capacity, and transportation time, etc. The water-railway combined transport network state data is matched according to the attribute classification: the tool location data corresponds to the transportation tool node attribute, the node state data corresponds to the port-station node attribute, the tool and cargo state data correspond to the transportation tool and cargo associated node attribute, and the transportation line related state data corresponds to the relationship edge attribute. After matching, the state enhanced knowledge graph containing real-time state information is formed.

[0176] Further, the entity node names and connection sequence involved in the initial transportation plan are extracted to determine the node sequence. The starting node of the sequence is found in the state enhanced knowledge graph, and the next node is located and accessed in turn according to the planned connection sequence, and the node identifier and connection relationship are recorded to ensure that all entity nodes in the plan are covered without missing any transfer or terminal nodes.

[0177] The planned attributes of each entity node are extracted from the initial transportation plan document, and the actual state attributes are extracted from the corresponding node attribute column of the state enhanced knowledge graph. The two are compared one by one and the differences are recorded, and the data entries are sorted according to "node name-planned attribute-actual state attribute-deviation content", and the node level deviation data of the initial transportation plan is obtained after summarizing.

[0178] Further, the current transportation capacity constraints and time dependent attributes of the relationship edges involved in the initial plan are extracted from the state enhanced knowledge graph, and the planned attributes of the corresponding relationship edges are obtained from the initial plan. It is judged whether the current attributes of each relationship edge satisfy the path validity: if the remaining through capacity is lower than the plan and cannot carry the transportation demand, or the actual transportation time is longer than the plan and affects the overall timely completion, it is determined that the path segment validity has changed. The comparison results are sorted according to "relationship edge identifier-planned attribute-current attribute-validity change", and the path level deviation data is obtained after summarizing.

[0179] Further, the entity nodes with deviations are screened out from the node level deviation data, and the relationship edges with deviations are screened out from the path level deviation data. Based on the topology structure of the knowledge graph, the other nodes and edges associated with the deviation nodes / edges are viewed, the influence range of the deviation is tracked along the associated relationship, and the transmission process is recorded. Among all the tracking paths, the path with the widest influence range and the greatest impact on the overall transportation efficiency is selected as the key influence propagation path of the initial plan.

[0180] Further, the node level and path level deviation data are associated, the causal relationship between entity node deviation and relationship edge deviation is analyzed and recorded. The associated deviation data and the key influence propagation path are combined to analyze the deviation transmission rule, and the propagation range and influence degree are determined. Based on the comprehensive analysis results, the deviation position, content, propagation path and overall influence are sorted to form the current transportation state and plan deviation report of the initial plan.

[0181] Specifically, the node deviation type and content are extracted from the node level deviation data, and the node name and deviation performance involved are determined. All paths containing the node are found from the path level deviation data, and the link that depends on the node for path connectivity is determined. The influence degree is determined: "severe influence" if the path is interrupted, "moderate influence" if the efficiency is reduced without interruption, and "mild influence" if the time is slightly prolonged. The entries are sorted according to "node deviation content - associated path identifier - influence degree", and the node-path association results are generated after summarizing.

[0182] Further, the path deviation situation is screened out from the path level deviation data, and the relationship edge identifier and deviation content are recorded. The network branch containing the deviation path is located in the key influence propagation path, and the branch role is determined. The influence range is analyzed: "local influence range" if only a small number of nodes in the branch are affected, "regional influence range" if all nodes in the branch and secondary branches are affected, and "global influence range" if multiple core branches and the overall order are affected. The entries are sorted according to "path deviation content - associated propagation path branch - influence range", and the path-network association results are generated after summarizing.

[0183] Further, the influence degree of the node-path association results is quantified as follows: mild for basic points, moderate for medium points, and severe for high points; the influence range of the path-network association results is quantified as follows: local for basic points, regional for medium points, and global for high points. The corresponding entries in the two types of association results are matched, and the influence degree and range scores are added to obtain the combined superposition score of a single deviation. The combined superposition score is divided into levels: mild for basic + basic, moderate for basic + medium / medium + basic / medium + medium, and severe for medium + high / high + medium / high + high. The combined superposition score and level of all deviation combinations are sorted to form the deviation severity level.

[0184] Further, the deviation information corresponding to each level is extracted from the deviation severity level, and the key influence propagation path is combined to describe the deviation transmission process. The information is integrated according to "deviation severity level - deviation specific content - influence propagation process - overall network influence summary" to form the current transportation state and plan deviation analysis document of the initial plan.

[0185] Overall, this step accurately matches and maps the state data of the pig iron intermodal network to the corresponding entity nodes and relationship edges in the knowledge graph according to attribute categories, so that the knowledge graph contains real-time state information, and finally forms a state-enhanced knowledge graph.

[0186] Overall, this step locates and accesses all entity nodes involved in the initial transportation plan in sequence based on the connection order of the entity nodes determined by the initial transportation plan in the state-enhanced knowledge graph, and completes the traversal of the sequence of entity nodes related to the initial transportation plan.

[0187] Overall, this step extracts the planned attributes of the entity nodes from the initial transportation plan, and then obtains the actual state attributes of the corresponding nodes from the state-enhanced knowledge graph. By comparing the two, the difference information is sorted out, and the node-level deviation data of the initial transportation plan is obtained.

[0188] Overall, this step determines whether the state change causes a change in the effectiveness of the initial transportation plan path based on the transportation capacity constraint attributes and time-dependent attributes carried by the relationship edges in the state-enhanced knowledge graph, and sorts out the effectiveness change results to obtain the path-level deviation data of the initial transportation plan.

[0189] Overall, this step uses the relationship topology of the knowledge graph to track the cascading effects of individual nodes or edges with deviations on the transportation network, filters out the propagation path with the widest influence range and the greatest influence degree, and obtains the key influence propagation path of the initial transportation plan.

[0190] Overall, this step multi-level associates the node-level deviation data, path-level deviation data, and key influence propagation path to analyze the causal relationship and propagation law among deviations, integrates the deviation position, content, and overall influence, and obtains the current transportation state and plan deviation of the initial transportation plan.

[0191] Overall, this step focuses on analyzing the influence degree of single node deviation on the path connectivity in path-level deviation based on node-level deviation data and path-level deviation data, and generates the node-path association result of the initial transportation plan by sorting out this influence relationship and organizing it into standard items.

[0192] Overall, this step focuses on analyzing the influence range of path-level deviation on the network stability in key influence propagation path based on path-level deviation data and key influence propagation path, and generates the path-network association result of the initial transportation plan by organizing the influence range determination results into system items.

[0193] Overall, this step performs influence layer superposition calculation on the generated node-path association result and path-network association result, integrates the influence information of the two, and divides the levels to obtain the deviation severity level of the initial transportation plan.

[0194] In general, this step integrates and compiles the classification information of deviation severity level and the transmission process of key influence propagation path, sorts out the specific situation, influence degree and propagation logic of the deviation, and finally obtains the current transportation state of the initial transportation plan and the plan deviation.

[0195] The transportation regulation module 106 is configured to dynamically adjust the initial transportation plan based on the key influence entities and relationships, and send automatic scheduling of the intermodal network according to the dynamically adjusted initial transportation plan.

[0196] In the embodiment of the present application, when the transportation deviation identification module executes the key influence entities and relationships in the initial transportation plan according to the current transportation state and the plan deviation, it is specifically used for:

[0197] Based on the deviation severity level corresponding to the current transportation state and the plan deviation, the affected entity nodes are identified in the knowledge graph to obtain the initial influence entity set of the initial transportation plan;

[0198] According to the transportation capacity constraint attribute carried on the relationship edge in the knowledge graph, the entity nodes with transportation bottleneck characteristics in the initial influence entity set are screened to obtain the capacity influence entity subset of the initial transportation plan;

[0199] According to the time dependence attribute carried on the relationship edge in the knowledge graph, the entity nodes with time criticality characteristics in the initial influence entity set are screened to obtain the time influence entity subset of the initial transportation plan;

[0200] In the knowledge graph, the relationship topology of the capacity influence entity subset and the time influence entity subset is traced back to identify the relationship edge directly connected with the entity node and affecting the transportation continuity to obtain the key influence relationship set of the initial transportation plan;

[0201] The importance of the capacity influence entity subset, the time influence entity subset and the key influence relationship set in the initial transportation plan is determined.

[0202] The transportation regulation module is specifically used for:

[0203] For the key influence entities, the replaceable transportation nodes are queried in the knowledge graph to obtain the alternative node set of the initial transportation plan;

[0204] For the relationship, find alternative paths in the knowledge graph that meet the transportation capacity constraints and time-dependent attributes, obtaining a set of alternative paths for the initial transportation plan;

[0205] Update the initial transportation plan based on the set of alternative nodes and the set of alternative paths, generating a dynamically adjusted initial transportation plan.

[0206] Specifically, the deviation severity level is extracted from the transportation state and plan deviation, and the corresponding deviation entity range is determined. In the knowledge graph, the corresponding level entity nodes are located, and the actual and planned state attributes are compared to confirm that the deviation belongs to the current range, and irrelevant nodes are excluded. The resulting set is the initial impact entity set of the initial transportation plan.

[0207] Further, the transportation capacity constraint attributes of the associated relationship edges of each node in the initial impact entity set are extracted from the knowledge graph, including the maximum transport capacity of the transport line, the current actual load rate, and the remaining transport capacity. Analyze these attributes one by one, if the actual load rate exceeds the plan or the remaining capacity cannot meet the subsequent demand, it is determined that the node has a transport bottleneck feature; filter out such nodes from the initial impact entity set, and the resulting subset is the capacity impact entity subset.

[0208] Further, the time-dependent attributes of the associated relationship edges of each node in the initial impact entity set are extracted, including the planned transit time, the actual transit time, the delay time, and the time impact coefficient of node delay on subsequent nodes. Filter out such nodes from the initial impact entity set, and the resulting subset is the time impact entity subset.

[0209] Further, take the nodes of the capacity and time impact entity subset as the starting point, expand the relationship topology in the knowledge graph, view the "direct associated relationship edge" attributes of each node and obtain all direct connection edges. Analyze each edge one by one, if its transportation capacity constraint does not meet the demand or the time-dependent attribute has a delay, which leads to the inability of the goods to be normally transported, it is determined that the edge affects the continuity of transportation; the resulting set is the key impact relationship set.

[0210] Further, based on the impact range and degree, the importance scores of the elements of the capacity impact entity subset, the time impact entity subset, and the key impact relationship set are determined and graded. Sort and select the highest grade elements in descending order of the score grade, where the front elements of the capacity and time impact entity subset are the key impact entities, and the front elements of the key impact relationship set are the key impact relationships, and the key impact entities and relationships of the initial transportation plan are obtained.

[0211] Specifically, the specific type of the key influence entity, the core function in the initial plan, and the basic attributes to be met are determined. The entity nodes with consistent types and the same core functions are screened in the knowledge graph to ensure that the candidate nodes are in similar transportation areas as the original entity and the transportation capacity constraints are not lower than the original entity. The real-time operation state of the candidate nodes is queried to confirm that there is no failure, no transportation capacity saturation, and the original entity transportation task can be undertaken. The nodes that meet the conditions are arranged in the format of "node type-node name-core function-real-time transportation capacity" to form a set of candidate nodes.

[0212] Further, the original path starting and ending entity nodes corresponding to the key influence relationship are extracted, and the transportation capacity constraints and time dependency requirements that the original path needs to meet are determined. Taking the original path starting and ending points as fixed nodes, all connected paths without key influence relationship edges are traversed in the knowledge graph to cover different combinations of transfer nodes. The transportation capacity constraints of each relationship edge are checked one by one to see if they meet the requirements of the original path, the total estimated transportation time is calculated, and it is confirmed that the time dependency requirements are met; a set of alternative paths is formed.

[0213] Further, the node with the highest functional matching degree and the most sufficient real-time transportation capacity is selected from the set of alternative nodes to replace the corresponding entity in the initial plan, ensuring the connection between the original plan and the transportation link; the path with the shortest total transportation time and the largest transportation capacity redundancy is selected from the set of alternative paths to replace the original path containing the key influence relationship in the initial plan, ensuring that the starting and ending points of the new path are consistent with the original path, and the passing nodes are compatible with the replaced alternative nodes. Check if the updated plan meets the core requirements of goods transportation, including matching of goods types and node / path attributes, and total transportation time meeting the expected arrival time; reorganize the updated nodes, paths, and supporting transportation parameters according to the initial transportation plan format to form the dynamically adjusted initial transportation plan.

[0214] Overall, this step locates and identifies the entity nodes affected by the deviation according to the deviation severity level corresponding to the current transportation state and plan deviation in the knowledge graph. After sorting and integrating these nodes, the initial influence entity set of the initial transportation plan is obtained.

[0215] Overall, this step screens the entity nodes in the initial influence entity set by using the transportation capacity constraint attributes carried by the relationship edges in the knowledge graph, and selects the nodes with transportation bottleneck characteristics to form the capacity influence entity subset of the initial transportation plan.

[0216] Overall, this step uses the time dependency attributes carried by the relationship edges in the knowledge graph to screen the entity nodes with time criticality characteristics from the initial influence entity set to obtain the time influence entity subset of the initial transportation plan.

[0217] In general, this step traces the relationship topology of the ability-influencing entity subset and the time-influencing entity subset in the knowledge graph, identifies the relationship edges directly connected to these entity nodes and affecting the continuity of transportation, and obtains the key influence relationship set of the initial transportation plan by aggregating these relationship edges.

[0218] In general, this step determines the key influence entities and relationships of the initial transportation plan by prioritizing the elements in the ability-influencing entity subset, the time-influencing entity subset, and the key influence relationship set through importance ranking.

[0219] In general, this step focuses on key influence entities and filters alternative transportation nodes that are consistent with the key influence entity type, functionally matched, and can undertake transportation tasks in the knowledge graph of the intermodal network. After sorting these nodes, the candidate node set of the initial transportation plan is obtained.

[0220] In general, this step focuses on key influence relationships and finds alternative paths that do not contain key influence relationship edges and meet transportation capacity constraints and time dependency attributes in the knowledge graph with the starting point and endpoint of the original path as fixed nodes. After aggregating these paths, the candidate path set of the initial transportation plan is obtained.

[0221] In general, this step replaces the key influence entities in the initial transportation plan with suitable candidate nodes and replaces the original path containing key influence relationships with qualified alternative paths based on the candidate node set and the candidate path set. After verification and integration, the dynamically adjusted initial transportation plan is generated.

[0222] It is obvious to those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application.

[0223] The embodiments of the present application can acquire and process related data based on artificial intelligence technology. Artificial intelligence is the use of digital computers or computer-controlled machines to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to obtain optimal results.

[0224] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the present application.

Claims

1. An automated management system for horizontal transport in rail-water intermodal transport, characterized in that, The system includes a data acquisition module, a knowledge graph construction module, a transportation planning module, a transportation monitoring module, a transportation deviation identification module, and a transportation control module, wherein: The data acquisition module is used to collect railway station location information, waterway port attribute information, transportation vehicle status information and cargo characteristic information into initial data for the rail-water intermodal transport network. The knowledge graph construction module is used to construct a knowledge graph of the rail-water intermodal transport network using railway stations and waterway ports in the initial data as entity nodes and connection relationships, transport capacity constraints and time dependency attributes in the initial data as relation edges. The transportation planning module is used to perform constrained reasoning in the knowledge graph based on transportation demand to obtain an initial transportation plan for the transportation demand. The transportation monitoring module is used to monitor the status data of the rail-water intermodal transport network in real time after the initial transportation plan is executed. The transport deviation identification module is used to evaluate the current transport status and plan deviation of the rail-water intermodal transport network based on the status data, and to traverse the key influencing entities and relationships in the initial transport plan according to the current transport status and plan deviation. The transportation control module is used to dynamically adjust the initial transportation plan based on the key influencing entities and relationships, and to send automated scheduling to the rail-water intermodal transport network according to the dynamically adjusted initial transportation plan.

2. The automated management system for horizontal transport oriented towards rail-water intermodal transport as described in claim 1, characterized in that, When the data acquisition module compiles railway station location information, waterway port attribute information, transport vehicle status information, and cargo characteristic information into initial data for the rail-water intermodal transport network, it is specifically used for: By connecting the railway station location information into a topology, a set of station data for the rail-water intermodal transport network is obtained; Extract waterway port attribute information from waterway port data to obtain the port data set of the rail-water intermodal transport network; The status information of the transport vehicles is obtained from the transport vehicle monitoring platform to obtain the set of vehicle status data of the rail-water intermodal transport network; Cargo characteristic information is retrieved from the cargo management database to obtain the cargo data set of the rail-water intermodal transport network; The initial data of the rail-water intermodal transport network is obtained by fusing the site data set, the port data set, the tool status data set, and the cargo data set.

3. The automated management system for horizontal transport oriented towards rail-water intermodal transport as described in claim 1, characterized in that, When the knowledge graph construction module constructs a knowledge graph of the rail-water intermodal transport network using railway stations and waterway ports in the initial data as entity nodes and connection relationships, transport capacity constraints, and time dependency attributes in the initial data as relation edges, it is specifically used for: Extract railway station entities and waterway port entities from the initial data, assign a unique identifier to each entity and add a type attribute to obtain the entity node set of the rail-water intermodal transport network; Identify the connection relationships between the entity nodes in the initial data, parse the transport capacity constraint parameters and time dependency attribute parameters corresponding to the connection relationships, and obtain the set of relationship edges of the rail-water intermodal transport network. The topology of the rail-water intermodal transport network is obtained by assembling the set of entity nodes and the relation edges into a graph structure. By associating the transport capacity constraint parameters and the time-dependent attribute parameters with the corresponding relation edges in the topology, a knowledge graph model of the rail-water intermodal transport network is obtained. The knowledge graph model is stored in a graph database, and a fast query channel for entities and relationships is established through the index interface of the graph database to obtain the knowledge graph of the rail-water intermodal transport network.

4. The automated management system for horizontal transport oriented towards rail-water intermodal transport as described in claim 3, characterized in that, When the knowledge graph construction module associates the transport capacity constraint parameters and the time-dependent attribute parameters with the corresponding relation edges in the topology to obtain the knowledge graph model of the rail-water intermodal transport network, it is specifically used for: Extract the maximum throughput capacity and current load rate features from the transport capacity constraint parameters to obtain the capacity constraint attribute set of the rail-water intermodal transport network; Extract the planned transport time and real-time delay features from the time-dependent attribute parameters to obtain the time attribute set of the rail-water intermodal transport network; The transport capacity constraint information of the capacity constraint attribute set is bound to the corresponding relation edge in the topology to obtain the first attribute edge of the rail-water intermodal transport network. The time dependency information in the time attribute set is bound to the corresponding relation edge in the topology to obtain the second attribute edge of the rail-water intermodal transport network; Image encapsulation is performed on the first attribute edge and the second attribute edge to obtain the knowledge graph model of the rail-water intermodal transport network.

5. The automated management system for horizontal transport oriented towards rail-water intermodal transport as described in claim 1, characterized in that, When the transportation planning module performs constrained reasoning based on transportation demand in the knowledge graph to obtain an initial transportation plan for the transportation demand, it is specifically used for: The system analyzes the origin location, destination location, cargo type, and expected arrival time characteristics of the transportation demand to generate a structured transportation demand. In the knowledge graph, the starting point and the ending point are used as query nodes. All connected paths are found through graph traversal operations to obtain a set of candidate paths for the transportation demand. Verify the matching between the transport capacity constraint attribute carried on each path relationship edge in the candidate path set and the cargo type to obtain a capacity-compliant path that meets the transport capacity requirements; Verify the consistency between the time dependency attributes carried on the relation edges in the capability compliance path and the expected arrival time to obtain a time compliance path that meets the time requirements; The route with the optimal overall transportation cost among the time-compliant routes is selected as the initial transportation plan for the transportation demand.

6. The automated management system for horizontal transport oriented towards rail-water intermodal transport as described in claim 1, characterized in that, When the transportation monitoring module performs real-time monitoring of the status data of the rail-water intermodal transport network after the initial transportation plan is executed, it is specifically used for: The location data of the transportation vehicles used in executing the initial transportation plan are obtained through a positioning device; The operating parameters of the transportation vehicle are collected by the equipment status sensors to obtain the vehicle status data; Obtain the current status of the goods in the transportation plan to obtain the status data of the goods; The node status data of the port-station operation data collection nodes is obtained by analyzing their operational status. The tool location data, tool status data, status data, and node status data are integrated to obtain the status data of the rail-water intermodal transport network.

7. The automated management system for horizontal transport oriented towards rail-water intermodal transport as described in claim 1, characterized in that, When the transport deviation identification module performs the evaluation of the current transport status and planned deviation of the rail-water intermodal transport network based on the status data, it is specifically used for: The state data is mapped to the corresponding entity nodes and relation edges in the knowledge graph to obtain a state-enhanced knowledge graph; In the state-enhanced knowledge graph, the sequence of entity nodes involved in the initial transportation plan is traversed; Extract the actual state attributes and planned attributes of the entity node sequence to obtain the node-level deviation data of the initial transportation plan; Based on the transportation capacity constraint attributes and time dependency attributes carried on the relation edges, the changes in path effectiveness caused by state changes in the initial transportation plan are identified, and the path-level deviation data of the initial transportation plan is obtained. By leveraging the relational topology of the knowledge graph, the chain reaction of deviations at individual nodes and edges on the entire transportation network is traced, revealing the key impact propagation paths of the initial transportation plan. A multi-level correlation analysis is performed on the node-level deviation data, the path-level deviation data, and the key impact propagation path to obtain the current transportation status and plan deviation of the initial transportation plan.

8. The automated management system for horizontal transport oriented towards rail-water intermodal transport as described in claim 7, characterized in that, When the transportation deviation identification module performs multi-level correlation analysis on the node-level deviation data, the path-level deviation data, and the key impact propagation path to obtain the current transportation status and deviation of the initial transportation plan, it is specifically used for: Based on the degree of influence of the deviation of a single node in the node-level deviation data and the path-level deviation data on the path connectivity in the path-level deviation data, the node-path association result of the initial transportation plan is generated; Based on the path-level deviation data and the deviation of a single node in the critical impact propagation path, the impact range of the path-level deviation on the network stability in the critical impact propagation path is determined, and the path-network association result of the initial transportation plan is generated. The severity level of the deviation of the initial transportation plan is obtained by superimposing the node-path association results and the path-network association results. The severity level of the deviation and the propagation path of the key impacts are compiled to obtain the current transportation status and deviation of the initial transportation plan.

9. The automated management system for horizontal transport oriented towards rail-water intermodal transport as described in claim 1, characterized in that, When the transportation deviation identification module performs the task of traversing the key influencing entities and relationships in the initial transportation plan based on the current transportation status and the plan deviation, it is specifically used for: Based on the severity level of the deviation corresponding to the current transportation status and the plan deviation, the affected entity nodes are identified in the knowledge graph to obtain the initial set of entities affecting the initial transportation plan. Based on the transportation capacity constraint attributes carried on the relation edges in the knowledge graph, entity nodes with transportation bottleneck characteristics are selected from the initial set of influencing entities to obtain the capacity-influencing entity subset of the initial transportation plan; Based on the time dependency attributes carried on the relation edges in the knowledge graph, entity nodes with time-critical characteristics are filtered from the initial set of influencing entities to obtain the time-influencing entity subset of the initial transportation plan; In the knowledge graph, trace the relationship topology between the subset of entities affected by capability and the subset of entities affected by time, identify the relationship edges that are directly connected to the entity nodes and affect the continuity of transportation, and obtain the set of key influencing relationships for the initial transportation plan; The key entities and relationships affecting the initial transportation plan are determined in the subset of entities affecting the capability, the subset of entities affecting the time, and the set of key impact relationships, ranked by importance.

10. The automated management system for horizontal transport oriented towards rail-water intermodal transport as described in claim 1, characterized in that, When the transportation control module performs dynamic adjustments to the initial transportation plan based on the key influencing entities and relationships, it is specifically used for: For the key influencing entities, alternative transportation nodes are queried in the knowledge graph to obtain a set of candidate nodes for the initial transportation plan; For the aforementioned relationship, alternative paths that satisfy the transportation capacity constraints and time dependency attributes are searched in the knowledge graph to obtain a set of alternative paths for the initial transportation plan; The initial transportation plan is updated based on the set of candidate nodes and the set of candidate paths to generate a dynamically adjusted initial transportation plan.

Images