Online foreign trade platform one-stop management system

Abstract

The invention discloses an online foreign trade platform one-stop management system, particularly relates to the technical field of cross-border logistics cost prediction, and is used for solving the problem of order total logistics cost prediction deviation caused by neglecting a multi-path coupling effect in an existing system. According to the system, fragmented logistics data such as a batch delivery batch, a multimodal transport node sequence and a tax floating parameter are obtained; constructing a transportation subunit combination network based on the node sequence, and identifying a path combination of shared transfer nodes; calculating a coupling cost increment triggered by the cargo detention time and the tax floating parameter, and generating a nonlinear coupling parameter; quantifying a cost conduction strength coefficient between the transportation subunits according to the parameter; and finally, correcting the initial logistics cost model by the coefficient, and outputting a total order logistics cost prediction value. The nonlinear coupling modeling of the dynamic cost elements in the multimodal transport scene is realized, and the cost prediction precision is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of cross-border logistics cost forecasting, and more specifically, to a one-stop management system for an online foreign trade platform. Background Art

[0002] The globalization of cross-border e-commerce has led foreign trade companies to adopt online, one-stop management systems to integrate core business processes such as order processing, warehousing and logistics, and payment settlement. These systems must integrate with international logistics providers, cross-border payment platforms, and multinational customs data interfaces to achieve comprehensive control over the entire process, from order generation to delivery. Existing technologies have already largely enabled centralized processing of order data and basic logistics status tracking.

[0003] However, when it comes to multi-batch, multi-path cross-border transportation scenarios, the nonlinear coupling effect of dynamic cost elements of different transportation paths has not been quantitatively modeled, resulting in systematic deviations in the system's prediction of the total logistics cost of an order, which in turn leads to distorted corporate profit assessments and operational decision-making errors. The existing technology's isolated processing of massive dynamic variables in fragmented logistics scenarios (such as batch shipments, multimodal transport, and tariff fluctuations) cannot reveal the cost transmission mechanism between transportation sub-units, affecting and restricting the reliability of one-stop system decisions. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides a one-stop management system for an online foreign trade platform to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions: A one-stop management system for an online foreign trade platform, including: The fragment integration module is used to obtain the fragmented logistics data of the current order, including batch shipping batches, multimodal transport node sequences and tariff floating parameters; Dynamic networking module, used to construct a combined network of transport subunits based on the multimodal transport node sequence and identify the transport subunit path combinations that share transfer nodes in the combined network; The coupling quantification module is used to calculate the coupling cost increment triggered by cargo detention time and tariff fluctuation parameters at the shared transfer node for the combination of transport sub-unit paths and batch shipment batches, and generate nonlinear coupling parameters; Intensity quantification module, used to quantify the cost transmission intensity between transport subunits according to nonlinear coupling parameters and generate cost transmission intensity coefficient; The precise correction module is used to correct the initial logistics cost model using the cost transmission intensity coefficient and output the predicted value of the total logistics cost of the order.

[0006] Furthermore, the fragmented logistics data of the current order is obtained. The fragmented logistics data includes batch shipments, multimodal transport node sequences, and tariff fluctuation parameters, including: Receive the transport route information returned by the international logistics service provider interface, parse the transshipment station identification and connection sequence in the transport route information to generate a multimodal transport node sequence; Extract the cargo split list and transport vehicle identification corresponding to the batches of shipments; Link the customs data interface to obtain the tariff floating parameters of each transshipment station identification; Based on the matching of the transport vehicle identification and the transfer terminal identification with the preset terminal scheduling table, the planned detention time of the batch shipment batches at each transfer terminal identification is determined; Integrate multimodal transport node sequences, cargo split lists, transport vehicle identification, tariff floating parameters and planned detention times to generate fragmented logistics data.

[0007] Furthermore, a combined network of transport subunits is constructed based on the multimodal transport node sequence, and a combination of transport subunit paths that share transfer nodes in the combined network is identified, including: Map the transshipment station identifiers in the multimodal transport node sequence to a node set of the combined network; Generate directed edges between nodes according to the connection sequence between transfer station identifiers to form a combined network of transport subunits; Traverse the node set of the combined network and extract the transport sub-unit paths containing the same transfer station identifier as candidate path combinations; The node overlap degree of the transport sub-unit paths in the candidate path combinations is checked, and the path combinations with the number of overlapping transfer station identifiers exceeding the threshold are retained as the transport sub-unit path combinations of shared transfer nodes.

[0008] Furthermore, for the transport subunit path combination and batch shipment, the coupling cost increment triggered by cargo detention time and tariff fluctuation parameters at the shared transshipment node is calculated, and nonlinear coupling parameters are generated, including: Extract the cargo batches with the same transshipment station ID in the cargo split list corresponding to the batches shipped; The planned detention time of the batched goods at the shared transshipment node is added together to generate the total detention time of the node; Filter the tariff floating parameters corresponding to the overlapping transfer station identifiers in the transport sub-unit path combination; The product of the total node detention time and the selected tariff floating parameter is calculated as the coupling cost increment; The nonlinear coupling parameters are generated by summing all coupling cost increments at shared transfer nodes.

[0009] Furthermore, extracting the cargo batches with the same transshipment terminal identification in the cargo split list corresponding to the batches of shipment includes: grouping the cargoes with the same first six digits of the customs code under the same transshipment terminal identification into the same batch group.

[0010] Furthermore, the cost transmission intensity between transport subunits is quantified based on the nonlinear coupling parameters to generate the cost transmission intensity coefficient, including: Extract the transfer station identification sequence contained in each path in the transport sub-unit path combination of the shared transfer node; Calculate the proportion of the coupling cost increment of each transport sub-unit path at the shared transfer node to the total value of the nonlinear coupling parameter; Calculate the path overlap between transport subunit paths with the same transfer station identification sequence; The conduction intensity factor of each transport subunit path pair is generated based on the product of the ratio of the coupling cost increment to the total value of the nonlinear coupling parameter and the path overlap; The conduction intensity factors of all transport subunit path pairs are normalized to generate the cost conduction intensity coefficient.

[0011] Furthermore, the cost transmission intensity coefficient is used to modify the initial logistics cost model and output the total logistics cost forecast value of the order, including: Obtain the independent transportation costs and independent tariff costs of the batches of shipments output by the initial logistics cost model; Extract the cost transmission intensity coefficient corresponding to the transport subunit path combination; The cost transmission intensity coefficient is multiplied by the coupling cost increment at the shared transfer node to generate the transmission correction amount; Add the conduction correction amount to the independent transportation cost of the batches shipped in batches to generate the corrected transportation cost; Combine the adjusted shipping costs with the independent tariff costs to generate a forecast of the total logistics cost for the order.

[0012] Compared with the prior art, the present invention has the following beneficial effects: This system uses a dynamic networking module to construct a combined network of intermodal transport node sequences and identify path combinations for shared transshipment nodes, enabling explicit modeling of topological coupling relationships between transport subunits. This design overcomes the limitations of traditional logistics systems in independently calculating multiple paths, accurately capturing the interactive effects of cargo detention and tariff fluctuations at shared nodes. Combined with a coupling quantification module, the system links detention duration and tariff parameters as nonlinear coupling parameters, effectively quantifying the dynamic cost increments of multiple batches at transshipment nodes, fundamentally addressing the problem of under-calculation of coupling costs associated with time-varying policies and operational constraints in cross-border logistics.

[0013] The intensity quantification module further generates a cost transmission intensity coefficient, integrating the dual factors of path overlap and economic weight to accurately characterize the cost diffusion intensity of shared nodes across multiple paths. The precision correction module dynamically refines the initial cost model, ultimately outputting a forecast of the total logistics cost for the order. This complete solution forms a closed-loop technical loop of "network modeling-coupled quantification-transmission correction," significantly improving the accuracy of cost forecasts in multimodal transport scenarios. This provides foreign trade companies with a reliable basis for profit assessment and decision-making, overcoming the technical shortcomings of existing systems that isolate fragmented logistics dynamic variables. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a structural diagram of a one-stop management system for an online foreign trade platform according to the present invention. DETAILED DESCRIPTION

[0015] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0016] Example: Figure 1 A structural diagram of a one-stop management system for an online foreign trade platform of the present invention is provided, which includes: The fragment integration module is used to obtain the fragmented logistics data of the current order, including batch shipping batches, multimodal transport node sequences and tariff floating parameters; Dynamic networking module, used to construct a combined network of transport subunits based on the multimodal transport node sequence and identify the transport subunit path combinations that share transfer nodes in the combined network; The coupling quantification module is used to calculate the coupling cost increment triggered by cargo detention time and tariff fluctuation parameters at the shared transfer node for the combination of transport sub-unit paths and batch shipment batches, and generate nonlinear coupling parameters; Intensity quantification module, used to quantify the cost transmission intensity between transport subunits according to nonlinear coupling parameters and generate cost transmission intensity coefficient; The precise correction module is used to correct the initial logistics cost model using the cost transmission intensity coefficient and output the predicted value of the total logistics cost of the order.

[0017] Obtain the fragmented logistics data of the current order. The fragmented logistics data includes batch shipments, multimodal transport node sequences, and tariff floating parameters. The specific implementation is as follows: The transport route information returned by the international logistics service provider interface includes a list of node codes arranged in the actual transport order and the connection relationships between nodes. The node codes use the internationally accepted three-letter airport code format. By parsing the node codes and their order in the transport route information, an ordered sequence of transfer terminal identifiers is generated to form a multimodal transport node sequence. The parsing process includes two core operations: first, duplicate node codes are removed to avoid path redundancy; second, the node connection logic is verified to ensure that it complies with the principle of geographic accessibility. For example, if the original transport route information data is "PVG-SIN-SIN-FRA", the parsing generates an ordered sequence including the Shanghai Pudong International Airport identifier "PVG", the Singapore Changi Airport identifier "SIN", and the Frankfurt Airport identifier "FRA". Duplicate "SIN" identifiers are merged. Extract the cargo disassembly list and transport vehicle identification corresponding to each batch of shipments. The cargo disassembly list records the details of each batch of cargo in a structured table, including the unique cargo number, the cargo category corresponding to the International Customs Harmonized System code, weight data in kilograms, and volume data in cubic meters. The transport vehicle identification consists of the transport vehicle type code and the vehicle number. The transport vehicle type code uses a single-letter coding scheme, where "V" represents a ship, "A" represents an aircraft, and "T" represents a truck. The vehicle number follows the container numbering rules promulgated by the International Organization for Standardization. For example, the transport vehicle identification corresponding to the batch number B003 is recorded as "V_MSKU742189", indicating the vessel type and Maersk Line container number 742189.

[0018] Use the associated customs data interface to obtain the tariff fluctuation parameters for each transshipment terminal identifier. This involves sending a structured query request to the customs data interface. This request includes the transshipment terminal identifier field and the customs code field from the cargo unbundling list. The tariff fluctuation parameters returned by the interface contain two data items: a base tax rate as a percentage, and a floating coefficient, which represents the adjustment based on policy changes over the past twelve months. For example, when querying the transshipment terminal identifier "FRA" corresponding to the commodity code "87032341," the returned parameters are "base tax rate 6.5%, floating coefficient ±1.8%." The preset terminal dispatch table is matched based on the transport vehicle identification and the transfer terminal identification. The terminal dispatch table is a relational database table that stores three sets of key parameters: the standard processing time parameter records the processing time (in hours) of different types of transport vehicles under standard working conditions, the current load factor parameter updates the real-time operating load rate of the terminal every hour, and the cargo type correction parameter maps the cargo category and processing priority weight; the matching process performs three steps: the first step is to retrieve the standard processing time base value according to the transport vehicle type code, for example, the identification "A" corresponds to the aircraft type standard processing time of 2.5 hours; the second step is to obtain the current load factor, which is obtained through the field The station operation system collects data through a real-time interface. The third step is to determine the priority weights based on the cargo categories in the cargo split list, with the weight of dangerous goods set to 1.25, the weight of cold chain goods to 1.15, and the weight of general cargo to 1.0. Finally, the planned detention time is calculated through multiplication: Planned detention time = standard processing time × (1 + current load factor) × priority weight. For example, if the standard processing time of the transport vehicle identified as "A_CXA205" at the "FRA" terminal is 2.5 hours, the current load factor is 0.3, and the cargo priority weight is 1.15, the calculated detention time is 2.5 × 1.3 × 1.15 = 3.7375 hours.

[0019] The fragmented logistics data is generated by integrating the multimodal transport node sequence, cargo split list, transport vehicle identification, tariff fluctuation parameters, and planned detention time. The integration process establishes the associated mapping relationship between five groups of data: the multimodal transport node sequence is used as the main chain structure to store the ordered terminal identification list; the tariff fluctuation parameters are used to establish a key-value mapping table according to the transshipment terminal identification, such as the mapping table entry ["FRA"→"6.5%±1.8%"]; the planned detention time is bound to the transport vehicle identification to form a time parameter set, such as the mapping entry ["V_MSKU742189@FRA"→"8.2 hours"]; the cargo split list is grouped by batch shipment batch number to store detailed data; and the transport vehicle identification set independently stores basic vehicle information. The resulting fragmented logistics data is stored in the JSON-LD structured format and contains five data blocks: the node sequence block stores the ordered list ["PVG","SIN","FRA"], the tariff block stores the dictionary {"PVG":"5.2%±0.7%","SIN":"0%±0%","FRA":"6.5%±1.8%"}, the timeliness block stores the dictionary {"B003@PVG":"4.5 hours","B003@SIN":"3.2 hours","B003@FRA":"8.2 hours"}, the split list block stores the batch index {"B003":[{"Cargo number":"C-88921","Item":"87032341","Weight":"4200 kg","Volume":"18.7 m³"}]}, and the tool identification block stores the vehicle index {"B003":"V_MSKU742189"}. A data integrity check mechanism is set up for each data block, and a cyclic redundancy check algorithm is used to generate a check code. The version number is updated in the "year-month-day-serial number" format.

[0020] Using a hash table to map nodes and depth-first search traversal reduces space complexity to O(n) compared to traditional adjacency matrix storage. Depth-first search reduces invalid accesses when detecting path branches compared to breadth-first search. The hash table implements node duplication detection with O(1) complexity, avoiding network bloat caused by redundant nodes. The recursive nature of depth-first search naturally adapts to path branch records, ensuring that shared nodes are identified without omission.

[0021] Based on the multimodal transport node sequence, a combined network of transport subunits is constructed, and the combination of transport subunit paths that share transfer nodes in the combined network is identified. The specific implementation is as follows: When constructing a combined network of transport subunits based on an intermodal node sequence, the transshipment terminal identifiers in the intermodal node sequence are first mapped to a node set in the combined network. This mapping process uses a dictionary data structure to establish a unique mapping between terminal identifiers and network nodes. Each transshipment terminal identifier is added to the set as an independent node, and the node attributes record the terminal's geographic coordinates and region code. The geographic coordinates are obtained by matching the International Air Transport Association's airport database and are formatted in decimal degrees. For example, the transshipment terminal identifier "PVG" is mapped to the node {Identifier: "PVG", Coordinates: "31.1434,121.805", Region: "CN"}. After the node set is generated, duplicate identifier detection is performed. When duplicate terminal identifiers are detected, only the first node is retained and marked as a shared node. For example, the second "SIN" in the sequence ["SIN", "FRA", "SIN"] is identified as a redundant node. Directed edges between nodes are generated based on the connection sequence between transshipment terminal identifiers. The rule for generating directed edges is to establish a one-way connection from the origin to the destination between two adjacent nodes in the node sequence. Edge attributes include a transport mode code and a distance value. The transport mode code is automatically derived based on the geographic relationship between the nodes. The derivation rule is as follows: if the two nodes are in the same region code, the value is "T" (truck transport); if they are separated by an ocean, the value is "V" (ship transport); and if they are separated by a land border, the value is "R" (rail transport). The distance value is calculated by calculating the spherical distance between the node coordinates using the Haversine formula, with distance units uniformly in kilometers. For example, a directed edge {origin: "PVG", destination: "SIN", mode: "V", distance: 3827} is generated between nodes PVG (31.1434, 121.805) and SIN (1.3502, 103.994). The resulting combined network of transport subunits consists of two basic elements: a node set and an edge set. The network structure is stored as an adjacency list.

[0022] The node set of the combined network is traversed, and transport subunit paths containing the same transfer terminal identifier are extracted as candidate path combinations. The traversal process uses a breadth-first search algorithm to scan the network starting from the network entry node, recording the complete path to each node during the scan. When a node is detected to be traversed by multiple paths, the candidate path combination extraction operation is triggered. The specific implementation involves four steps: the first step is to initialize the path queue and add the starting node to the queue; the second step is to iteratively pop the head node of the queue and explore its adjacent nodes to generate new paths; the third step is to mark the node as a shared node when a new path reaches a visited node; the fourth step is to extract all complete paths containing the shared node to form a candidate path combination. For example, if the network contains three paths: Path 1: PVG → SIN → FRA, Path 2: PVG → SIN → DXB, and Path 3: HKG → SIN → FRA, the candidate combination {Path 1, Path 2, Path 3} is extracted at the shared node "SIN". The candidate path combination is stored as a list of path objects, each of which contains a path number, a node sequence, and a set of edge attributes.

[0023] The node overlap of the transport subunit paths in the candidate path combination is checked. The node overlap is defined as the ratio of the number of transfer station identifiers contained in both paths to the average number of nodes in the paths. The check process performs three calculation steps: the first step is to calculate the total number of nodes for each path in the path pair, for example, path A contains 4 nodes and path B contains 5 nodes; the second step is to calculate the number of intersections of the node identifiers of the two paths, that is, the number of occurrences of the same station identifier; the third step is to calculate the overlap index: overlap = number of intersection nodes / [(number of nodes in path A + number of nodes in path B) / 2], the value range of this index is 0 to 2; for example, when the number of intersection nodes of path A and path B is 2 and the average number of nodes is (4+5) / 2=4.5, the overlap = 2 / 4.5≈0.444. The path combinations with the number of overlapping transfer station identifications exceeding the threshold are retained as the transport sub-unit path combinations of shared transfer nodes. The threshold is set with a fixed benchmark value of 0.4, which is determined based on the analysis of historical logistics data: the average overlap of 1,000 groups of actual transport path combinations is 0.38, and 0.4 is taken as the screening threshold; when the overlap of all path pairs in the candidate path combination is greater than or equal to 0.4, the entire combination is retained. For example, in the aforementioned candidate combination, the overlap of path 1-path 2 is 0.444>0.4, the overlap of path 1-path 3 is 0.444>0.4, and the overlap of path 2-path 3 is 0.222<0.4, and the entire combination is eliminated due to the existence of substandard path pairs.

[0024] The final output of the transport subunit path combinations for shared transit nodes is stored in a graph database format. Node entities contain identification fields and shared tag fields, while edge entities contain path number references. Three data files are generated during data persistence: a node file stores all shared node information, an edge file stores path connectivity, and a path file stores complete path attributes. For example, a validated and retained path combination has the following storage structure: a node file records ["SIN", "FRA"], an edge file records {"Path ID: TR789": ["PVG→SIN", "SIN→FRA"]}, and a path file records {"TR789": {Node sequence: ["PVG", "SIN", "FRA"], distance: [3827, 10253]}}.

[0025] By grouping the first six digits of customs codes (at the HS classification level), the number of calculation groups is reduced compared to grouping by complete codes. The first six digits correspond to broad commodity categories (e.g., Class 87 vehicles), ensuring relevance while avoiding fragmentation caused by over-segmentation. A 0.4 threshold, determined by regression from historical data, improves grouping accuracy compared to manual experience.

[0026] For the combination of transport subunit paths and batch shipments, the coupling cost increment triggered by cargo detention time and tariff fluctuation parameters at the shared transfer node is calculated, and nonlinear coupling parameters are generated. The specific implementation is as follows: When calculating the incremental coupling cost for transport subunit route combinations and batch shipments, the system first extracts cargo batches with the same transshipment terminal ID from the cargo disassembly list corresponding to the batch. This extraction process then performs a field match based on the cargo details of the batches. The matching rule is that cargo with the same first six digits of the customs code for the same transshipment terminal ID are grouped together. For example, batch B003, under the transshipment terminal ID "FRA," contains two types of cargo: cargo numbers C-88921 (customs code 87032341) and C-88922 (customs code 87032342). Because they share the same first six digits, "870323," these cargoes are combined into batch group G-01. The batch group data structure records the total weight and volume of the cargo within the group. The total weight is measured in kilograms, and the total volume is measured in cubic meters. This is calculated by summing the weight and volume of all cargo within the group.

[0027] The planned detention times of batched cargo at shared transshipment nodes are added together to generate the total node detention time. This process uses arithmetic accumulation: the planned detention times of each batch within the batch group, identified at the terminal, are directly added together to generate the total detention time. The planned detention time is uniformly expressed in hours. For example, if batch group G-01 includes batches B003 (detention time 3.7 hours) and B007 (detention time 4.2 hours), the total node detention time is 3.7 + 4.2 = 7.9 hours. If the same batch appears multiple times at a shared node, only the first detention time is used to avoid duplicate calculations.

[0028] Filter the tariff parameters corresponding to overlapping transshipment terminal identifiers within a transport subunit route combination. This filter query uses the tariff parameter mapping table generated in the previous step based on the shared transshipment node identifier. Extracted parameters include a base tax rate and a floating coefficient. For example, the shared node identifier "FRA" corresponds to the tariff floating parameters {"Base Tax Rate": 6.5,"Floating Coefficient": 1.8}, expressed in percentages. The tariff parameter filter is limited to the terminal identifiers actually traversed by the current transport subunit route combination; parameters for unrelated terminal identifiers are not included in the calculation.

[0029] The coupling cost increment is calculated as the product of the total node detention time and the selected tariff fluctuation parameter. This calculation is performed in two steps: First, the tariff fluctuation parameter is converted into the actual impact value. The actual impact value = base tariff rate × (1 + floating coefficient / 100). For example, if the base tariff rate is 6.5 and the floating coefficient is 1.8, the actual impact value = 6.5 × (1 + 0.018) = 6.617. Second, the coupling cost increment is calculated as: total node detention time × actual impact value, in hours·percentage. For example, if the total detention time is 7.9 hours and the actual impact value is 6.617, the coupling cost increment is 7.9 × 6.617 ≈ 52.2743 hours·percentage. The result is rounded to two decimal places.

[0030] The nonlinear coupling parameter is generated by summing all coupling cost increments at shared transfer nodes. This summation is performed across all shared transfer nodes included in the current transport subunit path combination, accumulating the coupling cost increments for each node. For example, if the combination contains three shared nodes, "FRA," "SIN," and "DXB," with increments of 52.27, 38.45, and 29.83, respectively, the nonlinear coupling parameter is 52.27 + 38.45 + 29.83 = 120.55 hours·percent. The final parameter is stored as a floating-point value, along with a list of the shared node identifiers and their corresponding increment components.

[0031] A three-dimensional tensor is used to store the intensity coefficient (path A × path B × node), increasing the node dimension compared to a two-dimensional matrix. The tensor structure explicitly represents the triple relationship between "path pair" and "node," avoiding the black-box calculations of traditional graph convolutional networks. The dual-factor fusion (economic proportion + physical overlap) ensures that the transmission intensity combines economic logic with topological characteristics.

[0032] The cost transmission intensity between transport subunits is quantified based on the nonlinear coupling parameters to generate the cost transmission intensity coefficient. The specific implementation is as follows: When quantifying the intensity of cost transmission between transport subunits based on nonlinear coupling parameters, we first extract the sequence of transshipment terminal identifiers contained in each path in the transport subunit path combination that shares a transshipment node. This extraction operation is based on the transport subunit path combination data generated in the previous step, traversing the node sequence attributes of each path to obtain a complete, ordered list of terminal identifiers. For example, path number TR789 contains the sequence ["PVG","SIN","FRA"], path number TR845 contains the sequence ["PVG","SIN","DXB"], and path number TR912 contains the sequence ["HKG","SIN","FRA"]. The sequence data is stored as a path dictionary structure, with the key being the path number and the value being the list of terminal identifiers.

[0033] Calculate the ratio of the incremental coupling cost at the shared transit node to the total value of the nonlinear coupling parameter for each transport subunit route. This calculation process involves three steps: First, obtain the incremental coupling cost at each shared node calculated in the previous step. For example, the incremental cost at the shared node "SIN" is 38.45 hours·percentage. Second, obtain the total value of the nonlinear coupling parameter, for example, 120.55 hours·percentage. Third, calculate the ratio: ratio = incremental cost at the node for that route / total value of the nonlinear coupling parameter. For example, if the incremental cost at the "SIN" node for route TR789 is 12.8 hours·percentage, then ratio = 12.8 / 120.55 ≈ 0.106. The ratio calculation is rounded to four decimal places.

[0034] Calculate the path overlap between transport subunit routes with the same transfer terminal identifier sequence. This calculation uses a node sequence comparison method: First, obtain the transfer terminal identifier sequences of two routes. Second, calculate the number of nodes where the sequences intersect. Third, calculate the overlap = number of intersection nodes / [(length of sequence A + length of sequence B) / 2]. For example, if route TR789 has a sequence length of 3 and route TR845 has a sequence length of 3, and the number of intersection nodes is ["PVG","SIN"], then the overlap is 2 / [(3 + 3) / 2] = 2 / 3 ≈ 0.6667. When comparing multiple routes, calculate the pairwise overlap for each route pair.

[0035] The transmission intensity factor (TIF) for each transport subunit path pair is generated based on the product of the ratio of the coupling cost increment to the total value of the nonlinear coupling parameter and the path overlap. The generation rule is: for each transport subunit path pair, the average ratio of the two paths at the shared node is taken and multiplied by the path overlap of the path pair. The specific formula is: TIF = (Ratio_PathA + Ratio_PathB) / 2 × Path overlap. For example, if the ratios of paths TR789 and TR845 at node "SIN" are 0.106 and 0.128, respectively, and the overlap is 0.6667, then the TIF is (0.106 + 0.128) / 2 × 0.6667 ≈ 0.117 × 0.6667 ≈ 0.078. The calculation results are stored as a path pair matrix, with the rows and columns representing the path numbers.

[0036] The cost transmission intensity coefficients are generated by normalizing the transmission intensity factors of all transport subunit path pairs. This normalization is performed using the sum-proportional method: The sum of the transmission intensity factors for all path pairs is calculated. The second step is to divide each factor by the sum to obtain the normalized value. For example, if the transmission intensity factors for three path pairs are 0.078, 0.095, and 0.102, respectively, and their sum is 0.275, the normalized coefficients are 0.078 / 0.275≈0.2836, 0.095 / 0.275≈0.3455, and 0.102 / 0.275≈0.3709, respectively. The resulting coefficients are stored as a three-dimensional tensor data structure, with the first dimension representing the starting path number, the second dimension representing the target path number, and the third dimension representing the shared node identifier. For example, the tensor element T[TR789][TR845]["SIN"]=0.2836.

[0037] The cost transmission intensity coefficient is used to modify the initial logistics cost model and output the total logistics cost forecast value of the order. The specific implementation is as follows: When obtaining the independent transportation costs and independent tariff costs for each batch of shipments output by the initial logistics cost model, the initial logistics cost model is calculated based on standard logistics pricing rules. Independent transportation costs include transportation vehicle rental fees, fuel consumption costs, and labor costs, all expressed in US dollars. Independent tariff costs are calculated by multiplying the dutiable value of the goods by the tariff rate, also expressed in US dollars. For example, the independent transportation cost for batch B003 is US$4,200, and the independent tariff cost is US$780. This acquisition process is achieved by querying the model output database, where database records store cost data fields indexed by batch number.

[0038] Extracting the cost transmission intensity coefficient for a transport subunit path combination is performed based on the three-dimensional tensor data structure generated in the previous step. Specifically, the tensor data is retrieved based on the identifier of the transport subunit path combination being processed. For example, when processing path combination PC-01, the tensor element T[TR789][TR845]["SIN"]=0.2836 is extracted. The intensity coefficient data is stored in a distributed in-memory database, and the combination identifier is used as the primary key for queries.

[0039] The cost transmission intensity coefficient is multiplied by the coupling cost increment at the shared transit node to generate the transmission correction. The calculation process is performed in two steps. The first step is to obtain the coupling cost increment at the shared transit node, which is obtained from the node coupling cost increment record calculated in the previous step. For example, the coupling cost increment at the shared transit node "SIN" is 38.45 hours·percentage. The second step is to perform the multiplication operation: transmission correction = cost transmission intensity coefficient × coupling cost increment. For example, if the intensity coefficient is 0.2836 and the coupling cost increment is 38.45, the transmission correction = 0.2836 × 38.45 ≈ 10.90462 hours·percentage. The result is converted to US dollars using the exchange rate factor of 0.15 USD per hour, which is based on historical logistics cost data. For example, 10.90462 hours·percentage × 0.15 ≈ 1.635693 USD.

[0040] The pass-through correction is added to the individual shipping costs of each batch to generate the adjusted shipping cost. This addition is arithmetic accumulation: Adjusted Shipping Cost = Individual Shipping Cost + Pass-through Correction (in monetary terms). For example, if batch B003 has an individual shipping cost of $4,200 and a pass-through correction of $1.635693, the adjusted shipping cost is $4,200 + $1.635693 = $4,201.635693. If multiple pass-through corrections apply to the same batch, all relevant corrections are added together. For example, if a batch is affected by corrections of $1.63 and $0.87, the total correction is $2.5.

[0041] The adjusted shipping cost and independent tariff cost are combined to generate a total logistics cost forecast for the order. This combination performs an addition calculation: Total Logistics Cost Forecast = Adjusted Shipping Cost + Independent Tariff Cost. For example, if batch B003 has an adjusted shipping cost of $4,201.64 and an independent tariff cost of $780, the forecast is $4,201.64 + $780 = $4,981.64. If an order includes multiple batches, the forecast values for all batches are added together. For example, if an order includes batches B003 (at $4,981.64) and B007 (at $5,210.20), the total forecast is $4,981.64 + $5,210.20 = $10,191.84. The final forecast value is stored as a floating-point value, with associated cost component details for each batch.

[0042] During implementation, the exchange rate coefficient was determined by collecting 10,000 samples of international logistics cost data from the past three years and performing a linear regression analysis of the tariff cost per unit time and the actual monetary cost. The regression coefficient was 0.15 USD / hour·percentage, with a correlation coefficient of 0.92. Regarding the boundary handling of the transmission correction, negative values were forcibly reset to zero to avoid cost deductions, as logistics cost transmission can only result in additional expenses. Data persistence uses a columnar storage format, with each record containing four basic fields: order number, batch number, cost type, and currency value.

[0043] Column-based storage of order cost data (order ID + batch ID + cost type + value) saves storage space compared to row-based storage. Column compression technology optimizes multi-batch query efficiency, reducing the time required for aggregate calculations for batches of 10,000 or more. Component-level associated storage supports cost traceability, accelerating data location during audits.

[0044] In the field of logistics cost forecasting, traditional approaches typically calculate transportation and tariff costs for individual batches based on static rate models, failing to effectively quantify the coupling effects caused by shared nodes in intermodal transport scenarios. This approach explicitly models the coupling between nodes in paths by constructing a combined network of transport subunits. Transshipment terminal identifiers are mapped to network nodes and directed connections are established. Deep traversal is then used to extract shared node path combinations, distinguishing it from conventional independent path calculation methods. In particular, a node overlap verification mechanism, combined with dynamic threshold settings, accurately captures highly correlated path groups. This design overcomes the limitations of existing techniques that separate physical topology from economic parameters. Furthermore, the product of the tariff fluctuation parameter and the detention time is innovatively proposed as the coupling cost increment. A dynamic association between goods and nodes is established based on the first six digits of the customs code, addressing the modeling challenges of coupling the time-varying nature of tariffs with cargo collection and distribution in cross-border logistics. Finally, the transmission intensity coefficient of a three-dimensional tensor structure is integrated with the path economic weight and topological overlap to achieve quantitative correction of nonlinear cost transmission.

[0045] The calculations involved in the embodiments are all dimensionless numerical calculations, and the preset parameters and thresholds in the calculations are set by those skilled in the art according to actual conditions.

[0046] It should be noted that the present invention can be deployed on the device itself to implement embedded applications, and can also be run on a PC or other terminal with a user interface, thereby meeting various hardware environments and usage requirements.

[0047] The above embodiments can be implemented in whole or in part via software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions or computer programs. When loaded or executed on a computer, the processes or functions described in the embodiments of this application are fully or partially performed. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wireless or wired transmission. Wired transmission methods include optical fiber, twisted pair, coaxial cable, etc.; wireless transmission methods include infrared, microwave, etc. The computer-readable storage medium can be any available medium accessible by a computer, or a data storage device such as a server or data center that contains a collection of one or more available media. The available medium can be magnetic media (e.g., floppy disks, hard disks, tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives.

[0048] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and modules described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0049] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms.

[0050] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed across multiple network modules. Some or all of the modules may be selected to achieve the purpose of this embodiment according to actual needs.

[0051] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0052] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0053] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

[0054] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A one-stop management system for an online foreign trade platform, characterized in that: include: The fragment integration module is used to obtain the fragmented logistics data of the current order, including batch shipping batches, multimodal transport node sequences and tariff floating parameters; Dynamic networking module, used to construct a combined network of transport subunits based on the multimodal transport node sequence and identify the transport subunit path combinations that share transfer nodes in the combined network; The coupling quantification module is used to calculate the coupling cost increment triggered by cargo detention time and tariff fluctuation parameters at the shared transfer node for the combination of transport sub-unit paths and batch shipment batches, and generate nonlinear coupling parameters; Intensity quantification module, used to quantify the cost transmission intensity between transport subunits according to nonlinear coupling parameters and generate cost transmission intensity coefficient; The precise correction module is used to correct the initial logistics cost model using the cost transmission intensity coefficient and output the predicted value of the total logistics cost of the order.

2. The one-stop management system for an online foreign trade platform according to claim 1, characterized in that: Obtain fragmented logistics data for the current order. Fragmented logistics data includes batch shipments, multimodal transport node sequences, and tariff fluctuation parameters, including: Receive the transport route information returned by the international logistics service provider interface, parse the transshipment station identification and connection sequence in the transport route information to generate a multimodal transport node sequence; Extract the cargo split list and transport vehicle identification corresponding to the batches of shipments; Link the customs data interface to obtain the tariff floating parameters of each transshipment station identification; Based on the matching of the transport vehicle identification and the transfer terminal identification with the preset terminal scheduling table, the planned detention time of the batch shipment batches at each transfer terminal identification is determined; Integrate multimodal transport node sequences, cargo split lists, transport vehicle identification, tariff floating parameters and planned detention times to generate fragmented logistics data.

3. The one-stop management system for an online foreign trade platform according to claim 2, characterized in that: A combined network of transport subunits is constructed based on the multimodal transport node sequence, and the transport subunit path combinations that share transfer nodes in the combined network are identified, including: Map the transshipment station identifiers in the multimodal transport node sequence to a node set of the combined network; Generate directed edges between nodes according to the connection sequence between transfer station identifiers to form a combined network of transport subunits; Traverse the node set of the combined network and extract the transport sub-unit paths containing the same transfer station identifier as candidate path combinations; The node overlap degree of the transport sub-unit paths in the candidate path combinations is checked, and the path combinations with the number of overlapping transfer station identifiers exceeding the threshold are retained as the transport sub-unit path combinations of shared transfer nodes.

4. The one-stop management system for an online foreign trade platform according to claim 3, characterized in that: For the combination of transport subunit routes and batch shipments, the coupling cost increment triggered by cargo detention time and tariff fluctuation parameters at the shared transfer node is calculated, and nonlinear coupling parameters are generated, including: Extract the cargo batches with the same transshipment station ID in the cargo split list corresponding to the batches shipped; The planned detention time of the batched goods at the shared transshipment node is added together to generate the total detention time of the node; Filter the tariff floating parameters corresponding to the overlapping transfer station identifiers in the transport sub-unit path combination; The product of the total node detention time and the selected tariff floating parameter is calculated as the coupling cost increment; The nonlinear coupling parameters are generated by summing all coupling cost increments at shared transfer nodes.

5. The one-stop management system for online foreign trade platform according to claim 4, characterized in that: Extracting the cargo batches with the same transshipment terminal identification from the cargo split list corresponding to the batches of shipment includes: grouping the cargoes with the same first six digits of the customs code under the same transshipment terminal identification into the same batch group.

6. The one-stop management system for an online foreign trade platform according to claim 4, characterized in that: The cost transmission intensity between transport subunits is quantified based on the nonlinear coupling parameters to generate the cost transmission intensity coefficient, including: Extract the transfer station identification sequence contained in each path in the transport sub-unit path combination of the shared transfer node; Calculate the proportion of the coupling cost increment of each transport sub-unit path at the shared transfer node to the total value of the nonlinear coupling parameter; Calculate the path overlap between transport subunit paths with the same transfer station identification sequence; The conduction intensity factor of each transport subunit path pair is generated based on the product of the ratio of the coupling cost increment to the total value of the nonlinear coupling parameter and the path overlap; The conduction intensity factors of all transport subunit path pairs are normalized to generate the cost conduction intensity coefficient.

7. The one-stop management system for online foreign trade platform according to claim 6, characterized in that: The cost transmission intensity coefficient is used to modify the initial logistics cost model and output the total logistics cost forecast for the order, including: Obtain the independent transportation costs and independent tariff costs of the batches of shipments output by the initial logistics cost model; Extract the cost transmission intensity coefficient corresponding to the transport subunit path combination; The cost transmission intensity coefficient is multiplied by the coupling cost increment at the shared transfer node to generate the transmission correction amount; Add the conduction correction amount to the independent transportation cost of the batches shipped in batches to generate the corrected transportation cost; Combine the adjusted shipping costs with the independent tariff costs to generate a forecast of the total logistics cost for the order.

Images