Water transportation statistical model construction method and device based on ship operation big data

Abstract

The application provides a water transportation statistical model construction method and device based on ship operation big data, comprising: obtaining ship operation big data, preprocessing the ship operation big data; based on the ship in and out port report, data is connected to form a ship report port data; based on the ship report port data, a ship transportation whole chain model is constructed; based on the ship transportation whole chain model, a waterway transportation quantity model system is constructed from multiple angles, and the waterway transportation condition is determined based on the waterway transportation quantity model system; wherein the angles include freight volume, turnover volume, flow direction and throughput. Through deep fusion of multi-source data, the complete ship transportation trip is restored, and the end-to-end waterway transportation quantity model system is established. The waterway transportation quantity model system can comprehensively calculate key indicators such as freight volume, cargo turnover volume, cargo flow direction and port throughput, and lays a solid technical and model foundation for fine statistical analysis of waterway transportation conditions.

Description

Technical Field

[0001] This invention relates to the field of waterway transport volume statistics technology, and in particular to a method and apparatus for constructing a waterway transport statistical model based on big data of ship operations. Background Technology

[0002] Currently, existing methods for statistically analyzing waterway freight volume focus on transport vessels. They primarily employ sampling surveys combined with fluctuation estimations and enterprise reporting, involving manual, hierarchical reporting and aggregation to ultimately generate waterway cargo transport volume data based on the vessel's place of registration. While this method has reflected the overall trend of waterway freight volume since its implementation, it also has certain limitations: I. Insufficient Objectivity and Efficiency of Statistical Methods. Given the abundant data provided by current port entry and exit reports and the AIS (Automatic Identification System), existing methods fail to effectively utilize these big data resources. They still mainly rely on manual methods and enterprise reporting for statistics, which are highly subjective, have limited accuracy, and are inefficient.

[0003] Second, the statistical process suffers from low automation and high reliance on manual review. The existing process includes numerous review steps requiring manual intervention, which not only increases labor costs but also impacts the overall efficiency and scalability of statistical work.

[0004] Third, the utilization rate of maritime data resources is low. While maritime port entry and exit reports, AIS and other data resources are becoming increasingly abundant, statistical work has failed to embrace big data methods and still relies on subjective estimation of fluctuation coefficients, resulting in significant shortcomings in the scientific validity and timeliness of statistical results.

[0005] Fourth, the statistical dimensions are relatively singular, making it difficult to comprehensively and effectively reflect the overall situation of waterway freight transportation. Summary of the Invention

[0006] In view of this, the purpose of this invention is to provide a method and apparatus for constructing a water transport statistical model based on big data of ship operations, so as to construct a new statistical method for waterway transportation.

[0007] In a first aspect, embodiments of the present invention provide a method for constructing a water transport statistical model based on big data of ship operations. The method includes: acquiring big data of ship operations and preprocessing the big data of ship operations; wherein, the big data of ship operations includes: ship entry and exit reports and ship automatic identification system data; concatenating the data based on the ship entry and exit reports to form ship port reporting data; constructing a full-chain model of ship transportation based on the ship port reporting data; wherein, the ship port reporting data is recorded in the order of first leaving the port and then entering the port, and the ship port reporting data is sorted according to the ship identification number and reporting time based on the ship entry and exit reports; the full-chain model of ship transportation is used to characterize the number of transport trips of a ship from full loading to empty unloading, the initial position of the full-chain model of ship transportation is the initial loading behavior of the ship identified in the ship port reporting data, and the ending position of the full-chain model of ship transportation is the arrival and empty unloading behavior of the ship identified in the ship port reporting data; constructing a waterway transport volume model system from multiple perspectives based on the full-chain model of ship transportation, and determining the waterway transport status based on the waterway transport volume model system; wherein, the perspectives include: freight volume, turnover volume, flow direction, and throughput.

[0008] In an optional embodiment of this application, the above-mentioned steps for preprocessing big data on ship operations include: cleaning the data from ship arrival and departure reports; and standardizing and parsing the data from the Automatic Identification System (AIS) to convert it into structured relational data.

[0009] In an optional embodiment of this application, the step of forming ship port reporting data by concatenating data based on ship arrival and departure reports includes: sorting ship arrival and departure reports according to ship identification number and reporting time to form ship port reporting data; if there are consecutive departure records or consecutive arrival records in the ship port reporting data, retaining the latest departure record or the latest arrival record in the consecutive departure records or consecutive arrival records, and removing departure records or arrival records from other times.

[0010] In an optional embodiment of this application, the steps of constructing a full-chain model of ship transportation based on ship port reporting data include: identifying the initial loading behavior of a ship from the ship port reporting data, and taking the location where the initial loading behavior occurs as the initial location of the full-chain model of ship transportation; identifying the unloading behavior of a ship upon arrival at port from the ship port reporting data, and taking the location where the unloading behavior occurs as the end location of the full-chain model of ship transportation; and removing abnormal data from the full-chain model of ship transportation.

[0011] In an optional embodiment of this application, the step of identifying the initial loading behavior of a vessel from the vessel's port declaration data includes: identifying the initial loading behavior of a vessel from the vessel's port declaration data based on a preset first judgment condition; wherein the first judgment condition includes: the loading amount is equal to the actual load, and the loading amount is not equal to 0; or, the loading amount is not equal to the actual load, and the difference between the loading amount and the actual load is less than a preset threshold; the step of identifying the unloading behavior of a vessel arriving at port from the vessel's port declaration data includes: identifying the unloading behavior of a vessel arriving at port based on a preset second judgment condition; wherein the second judgment condition includes: the unloading amount is equal to the actual load, and the unloading amount is not equal to 0; or, the unloading amount is not equal to the actual load, and the difference between the unloading amount and the actual load is less than a threshold.

[0012] In optional embodiments of this application, the above method further includes: calculating the data difference rate distribution between loading volume and actual load or unloading volume and actual load by means of kernel density estimation, determining the distribution peak of the difference rate distribution; and determining a threshold based on the distribution peak.

[0013] In optional embodiments of this application, the steps of constructing a waterway transport volume model system from multiple perspectives based on the full chain model of shipping include: constructing a waterway freight volume model, a waterway turnover model, a throughput model, a cargo flow direction model, and a regional waterway transport volume model based on the full chain model of shipping.

[0014] In an optional embodiment of this application, the step of constructing a cargo flow direction model based on the full chain model of shipping includes: identifying sets of shipping chains entering and leaving various regions based on the full chain model of shipping; constructing a cargo flow direction model based on multiple sets of analytical dimensions of the shipping chains; wherein the analytical dimensions include: outflow model, inflow model and internal flow model.

[0015] In an optional embodiment of this application, the step of constructing a regional waterway transport volume model based on the full chain model of shipping includes: selecting transport records whose starting or ending point in the shipping chain is located within the target area based on the full chain model of shipping; aggregating and calculating the transport records according to multiple dimensions to generate a regional waterway transport volume model; wherein, the dimensions include: cargo type, ship type and time period.

[0016] Secondly, embodiments of the present invention also provide a water transport statistical model construction device based on ship operation big data. The device includes: a ship operation big data preprocessing module for acquiring and preprocessing ship operation big data; wherein the ship operation big data includes: ship arrival and departure reports and ship automatic identification system (AIS) data; a ship transport full-chain model construction module for concatenating data based on ship arrival and departure reports to form ship port reporting data; and constructing a ship transport full-chain model based on the ship port reporting data; wherein the ship port reporting data is recorded in the order of departure followed by arrival, and the ship port reporting data is based on... The vessel entry and exit reports are sorted by vessel identification number and reporting time; the vessel transportation chain model is used to characterize the number of transport trips a vessel undertakes from full loading to empty unloading. The initial position of the vessel transportation chain model is the initial loading behavior of the vessel identified in the vessel port report data, and the ending position of the vessel transportation chain model is the arrival and empty unloading behavior of the vessel identified in the vessel port report data; the waterway transportation volume model system construction module is used to construct a waterway transportation volume model system from multiple perspectives based on the vessel transportation chain model, and to determine the waterway transportation status based on the waterway transportation volume model system; among which, the perspectives include: freight volume, turnover volume, flow direction, and throughput.

[0017] The embodiments of the present invention bring the following beneficial effects: This invention provides a method and apparatus for constructing a water transport statistical model based on big data of ship operations. By deeply integrating multi-source data, it reconstructs complete ship transport trips and establishes an end-to-end waterway transport volume model system. This waterway transport volume model system can comprehensively measure key indicators such as freight volume, cargo turnover, cargo flow direction, and port throughput, laying a solid technical and model foundation for the refined statistics and analysis of waterway transport.

[0018] Other features and advantages of this disclosure will be set forth in the following description, or some features and advantages may be inferred from the description or determined without doubt, or may be learned by practicing the techniques described above.

[0019] To make the above-mentioned objects, features and advantages of this disclosure more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0020] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0021] Figure 1A flowchart illustrating a method for constructing a water transport statistical model based on big data of ship operations, provided in an embodiment of the present invention; Figure 2 A flowchart illustrating another method for constructing a water transport statistical model based on big data of ship operations, provided in an embodiment of the present invention; Figure 3 A schematic diagram illustrating a method for constructing a water transport statistical model based on big data of ship operations, provided in an embodiment of the present invention; Figure 4 This is a schematic diagram illustrating the preprocessing of raw AIS data of a ship, provided by an embodiment of the present invention. Figure 5 This is a schematic diagram of ship port declaration data provided in an embodiment of the present invention; Figure 6 A schematic diagram of a kernel density estimation curve provided in an embodiment of the present invention; Figure 7 A schematic diagram of a water transport statistical model construction device based on big data of ship operations provided in an embodiment of the present invention; Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] Currently, existing methods for statistically analyzing waterway transport volume focus on transport vessels. They primarily employ sampling surveys combined with fluctuation estimations and enterprise reporting, involving manual, hierarchical reporting and aggregation to ultimately generate waterway cargo transport volume data based on the vessel's place of registration. While this method has reflected the overall trend of waterway transport volume since its implementation, it also suffers from limitations such as insufficient objectivity and efficiency, low automation of the statistical process, and relatively singular statistical dimensions.

[0024] Based on this, this invention provides a method and apparatus for constructing a water transport statistical model based on big data of ship operations. Specifically, it proposes a method for analyzing the ship transport chain based on big data of ship operations, such as ship entry and exit reports, and constructs a new statistical method for waterway transport based on this. The core process of the water transport statistical model construction method based on big data of ship operations provided in this embodiment is divided into three main parts: the first part is to preprocess the big data of ship operations, such as ship entry and exit reports, to lay a high-quality data foundation for subsequent analysis; the second part is to construct a complete ship transport chain model based on the preprocessed data; and the third part is to further construct a systematic waterway transport volume model system based on this.

[0025] To facilitate understanding of this embodiment, a detailed description of the water transport statistical model construction method based on big data of ship operation disclosed in this embodiment of the invention will be provided first.

[0026] Example 1: This invention provides a method for constructing a water transport statistical model based on big data of ship operations. See [link to relevant documentation]. Figure 1 The flowchart shown illustrates a method for constructing a water transport statistical model based on big data from ship operations. This method includes the following steps: Step S102: Obtain ship operation big data and preprocess the ship operation big data; wherein, the ship operation big data includes: ship entry and exit reports and ship automatic identification system data.

[0027] This embodiment allows for the preprocessing of big data on ship operations. Statistical analysis of waterway transport volume is primarily based on big data on ship operations, such as ship arrival and departure reports and AIS data. Specifically, ship arrival and departure reports contain information from across the country and require cleaning and filtering to remove duplicate or redundant data. Raw ship AIS data consists of specially formatted ship AIS statements, which need to be parsed and converted into relational data to lay the foundation for subsequent data fusion and association.

[0028] Step S104: Connect the data based on the ship entry and exit reports to form ship port reporting data; construct a full-chain model of ship transportation based on the ship port reporting data.

[0029] Among them, the ship port reporting data is recorded in the order of departure first and arrival second. The ship port reporting data is sorted by ship identification number and reporting time based on the ship arrival and departure reports. The ship transportation full chain model is used to characterize the transportation trips of the ship from full loading to empty unloading. The initial position of the ship transportation full chain model is the initial loading behavior of the ship identified in the ship port reporting data, and the ending position of the ship transportation full chain model is the arrival and empty unloading behavior of the ship identified in the ship port reporting data.

[0030] This embodiment can construct a complete ship transportation chain model. By serializing ship port reporting data to form ship operation nodes from departure to arrival, a complete ship transportation chain model is formed based on judgment rules, fully reconstructing the ship's transportation trips from full loading to empty unloading.

[0031] Step S106: Construct a waterway transport volume model system from multiple perspectives based on the full chain model of shipping, and determine the waterway transport status based on the waterway transport volume model system; among which, the perspectives include: freight volume, turnover volume, flow direction and throughput.

[0032] In this embodiment, a waterway transportation volume model system can be constructed. Based on the full-chain model of shipping transportation, models are built from different perspectives such as freight volume, turnover, flow direction, and throughput to form a waterway transportation volume model system.

[0033] This invention provides a method for constructing a water transport statistical model based on big data of ship operations. By deeply integrating multi-source data, it reconstructs complete ship transport trips and establishes an end-to-end waterway transport volume model system. This waterway transport volume model system can comprehensively measure key indicators such as freight volume, cargo turnover, cargo flow direction, and port throughput, laying a solid technical and model foundation for the refined statistics and analysis of waterway transport.

[0034] Example 2: This embodiment provides another method for constructing a water transport statistical model based on big data from ship operations. This method is implemented based on the above embodiment. See [link to previous embodiment]. Figure 2 The flowchart shown is another method for constructing a water transport statistical model based on big data of ship operations. This method includes the following steps: Step S202: Acquire big data on ship operations; clean the data from ship arrival and departure reports; standardize and parse the data from the Automatic Identification System (AIS) and convert it into structured relational data.

[0035] See Figure 3 The diagram shows a method for constructing a water transport statistical model based on big data of ship operations. In this embodiment, data preprocessing can be performed first.

[0036] 1. Preprocessing of ship entry and exit reports: The entry and exit report data records the entry and exit of ships and loading and unloading information. The data volume is large and comprehensively records the ship navigation trajectory and cargo loading and unloading dynamics. It has the characteristics of large data scale and wide information dimensions. However, due to system errors in the data collection process and differences in manual entry, there are quality problems such as time logic conflicts, duplicate and redundant records, and missing key fields in the original dataset.

[0037] This embodiment can employ a layered processing architecture to implement data cleaning: (1) In the basic cleaning layer, invalid records such as abnormal timestamps and duplicate ship identifiers are identified and removed by the rule engine; (2) In the data enhancement layer, logical verification and normalization are performed on key numerical fields such as loading and unloading volume and actual loading volume; (3) In the standard output layer, construct a nationally unified data view containing standardized fields such as ship name, port of registry code, reporting agency, loading and unloading volume, port entry and exit identification, cargo classification, actual cargo volume, and reporting time.

[0038] After this stage of in-depth processing, the final vessel arrival and departure reports can include fields such as vessel name, port of registry code, reporting agency, cargo loading / unloading volume, arrival / departure identifier, cargo type, actual cargo load, and report time. This not only eliminates noise interference in the raw data but also, through field standardization and structural optimization, establishes a complete and reliable data support system for subsequent reconstruction of the shipping chain and accurate calculation of cargo volume. This data foundation effectively ensures the quality of the transformation from raw data to business insights, laying a solid foundation for end-to-end analysis.

[0039] 2. AIS data preprocessing: See Figure 4 The diagram illustrates a preprocessing method for raw ship AIS data. This method standardizes and parses the raw AIS data, transforming it into structured relational data. The process specifically includes: (1) Message extraction: Identify and extract the encapsulated original AIS message from the original data stream of AIS data.

[0040] (2) Format conversion: Convert the extracted message information, which exists in the form of ASCII code string, into a binary data stream that can be directly parsed.

[0041] (3) Information Decoding and Structuring: Based on the AIS communication protocol standard, the binary data segments are mapped bit by bit to the corresponding information fields to complete the parsing of key ship information (such as ship name, MMSI (Maritime Mobile Service Identity), latitude and longitude, speed, etc.), and finally generate standard structured data records.

[0042] Step S204: Connect the data based on the ship entry and exit reports to form ship port reporting data; construct a full-chain model of ship transportation based on the ship port reporting data.

[0043] In some embodiments, vessel arrival and departure reports can be sorted by vessel identification number and reporting time to form vessel port reporting data; if there are consecutive departure records or consecutive arrival records in the vessel port reporting data, the latest departure record or the latest arrival record is retained in the consecutive departure records or consecutive arrival records, and departure records or arrival records from other times are removed.

[0044] This embodiment sorts the port entry and exit report data of vessels by vessel identification number and report time, and strings together the vessel's departure and arrival records to obtain the vessel's entry and exit status over a certain time dimension, i.e., "departure-arrival-departure...". If the port report data is not in the order of "departure-arrival-departure-arrival...", consecutive duplicate departure or arrival records are removed. For example, if it is "departure-departure-arrival", then the principle of prioritizing the latest time is applied, retaining the latest departure record data and removing departure records from other times, ultimately only retaining "departure-arrival". This ensures that the port report data of each vessel maintains the logical consistency of "one departure and one arrival", laying the foundation for the construction of the shipping transport chain. (See also...) Figure 5 The diagram shown is a schematic representation of ship port declaration data.

[0045] In some embodiments, the initial loading behavior of a vessel can be identified from the vessel's port reporting data based on a preset first judgment condition; wherein the first judgment condition includes: the loading amount is equal to the actual load, and the loading amount is not equal to 0; or, the loading amount is not equal to the actual load, and the difference between the loading amount and the actual load is less than a preset threshold; the unloading behavior of a vessel arriving at port can be identified from the vessel's port reporting data based on a preset second judgment condition; wherein the second judgment condition includes: the unloading amount is equal to the actual load, and the unloading amount is not equal to 0; or, the unloading amount is not equal to the actual load, and the difference between the unloading amount and the actual load is less than a threshold.

[0046] like Figure 3 As shown, this embodiment can determine the initial loading behavior of a ship.

[0047] Based on complete data on ship arrivals and departures, by logically determining the cargo loading and actual load in each departure report, it is possible to ascertain whether a ship is loading cargo at a certain location while empty, thus determining the initial position of the chain. The initial loading behavior of a ship is mainly determined by two conditions: (1) Under the condition that the loaded quantity equals the actual load: ;in, For the cargo volume of the ship, This refers to the actual load capacity of the vessel.

[0048] Based on the logic of ensuring that ships "go out and come in", if the amount of cargo loaded at the departure port = the actual cargo volume at the departure port ≠ 0, that is, the amount of cargo loaded at this point is consistent with the amount of cargo actually transported at the departure port, then it is considered that the empty ship loading and transportation starts from this point, and thus the initial position of the chain is determined.

[0049] (2) Under the condition that the loaded quantity is not equal to the actual load: ;in, For the cargo volume of the ship, This refers to the actual load capacity of the vessel, and the cargo volume loaded on the vessel is not equal to the actual load capacity.

[0050] Based on the logic of ensuring that a ship "goes out and comes in", if the cargo volume loaded at the port and the actual load at the port are not equal and the difference percentage is within 4% (i.e., the threshold is 4%), it is considered that the empty ship loading and transportation started from that location.

[0051] like Figure 3 As shown, this embodiment can also determine the unloading behavior of ships upon arrival at port.

[0052] Based on the initial loading of the ship, by logically judging the unloading volume and actual load of the ship each time it enters the port, it is possible to determine whether the ship has unloaded at a certain place, thus completing the closed loop of the ship transportation chain and determining the end position of the chain. The behavior of a ship unloading at the port is mainly judged by three conditions.

[0053] (1) Under the condition that the unloading volume equals the actual load volume: ;in, For the volume of cargo unloaded by the ship, This refers to the actual load capacity of the vessel.

[0054] Based on the premise of ensuring "one out and one in", if the amount of cargo unloaded upon arrival at the port = the actual amount of cargo loaded upon arrival at the port ≠ 0, that is, the amount of cargo actually brought in by the ship is the same as the amount of cargo unloaded at the port, then the ship is considered to have unloaded empty at that location.

[0055] (2) Under the condition that the unloading volume is not equal to the actual load volume: ;in, For the volume of cargo unloaded by the ship, This refers to the actual load capacity of the vessel, and the unloaded cargo volume is not equal to the actual load capacity.

[0056] Based on ensuring that a vessel "goes out and comes in", if the unloading volume upon arrival at the port is not equal to the actual load upon arrival, and the difference percentage is within 4% (i.e., the threshold is 4%), it is considered that the vessel has unloaded empty at that location.

[0057] Based on the ship's departure loading conditions and arrival unloading conditions, the ship's overall navigation trajectory is linked together to form ship transportation chain data.

[0058] In some embodiments, the data difference rate distribution between loading volume and actual load or unloading volume and actual load can be calculated by kernel density estimation, and the peak value of the difference rate distribution can be determined; a threshold value can be determined based on the peak value.

[0059] Due to systematic errors in ship port reporting and inherent cargo losses during shipping, discrepancies may arise between reported load and actual loading / unloading volumes. This discrepancy is easily overlooked when determining whether a ship is loading or unloading empty, leading to data errors. Therefore, the variance rate needs to be taken into account. Kernel density estimation can be used to calculate the distribution of the variance rate and find its peak value. ;in, It is the difference rate between the ship's loading / unloading volume and the actual load, that is, the difference between the ship's outbound loading volume and the actual load, and the difference between the ship's inbound unloading volume and the actual load. V is the reported loading or unloading volume at port. This refers to the actual load capacity of the vessel, and both the declared loading / unloading volume and the actual load capacity are... .

[0060] The difference rate of all valid records constitutes the sample set. , where n is the total sample size. The kernel density estimation method is used to calculate the probability density distribution of the sample. At any point, the kernel density estimate is: ;in, Let x be the probability density estimate at point x. The kernel function, typically the Gaussian kernel function, is chosen. h is a broadband parameter that controls the smoothness of the density curve.

[0061] The choice of bandwidth h directly affects the kernel density estimation results. Too small a bandwidth will lead to an overly sharp density curve (overfitting), while too large a bandwidth will result in excessive smoothing and masking of detailed features. In this embodiment, the optimal bandwidth can be determined using the Silverman rule (an empirical rule for selecting bandwidth in kernel density estimation): ;in, is the sample standard deviation, IQR is the sample interquartile range, and n is the total sample size.

[0062] This rule can adaptively adjust the bandwidth according to the dispersion of the data, preserving important distribution characteristics while keeping the curve smooth.

[0063] See also Figure 6 The diagram shown illustrates a kernel density estimation curve. (The diagram shows the process of obtaining the kernel density estimation curve.) Afterwards, the peak value of the distribution can be obtained. Analysis of the difference rate between the reported loading / unloading volume and the actual cargo volume reveals that the difference is distributed at two extremes: one is mainly concentrated above 0.8, which is a normal phenomenon during ship loading / unloading; the other is below 0.04, where the difference is too small to be caused by ship loading / unloading behavior, and is considered to be due to errors in the system reporting process or cargo damage during transportation. Therefore, a 4% error margin should be added when judging ship loading behavior to avoid mistakenly deleting normal reported port data; that is, the threshold can be 4%.

[0064] This embodiment can also remove invalid and incorrectly entered data. In addition to data within the normal chain, some invalid or incorrectly entered data will be removed to check the consistency of the data logic. If a situation occurs where "loading volume is 0, but unloading volume is greater than 0," it is judged as abnormal data. This type of data is usually caused by port staff repeatedly reporting to the port or filling in invalid records, and it will also be removed. Ultimately, a complete transportation chain is formed from the ship loading cargo (port of origin) to unloading cargo (port of destination), fully demonstrating the overall situation of ship transportation.

[0065] Step S206: Construct a waterway transport volume model system from multiple perspectives based on the full chain model of shipping, and determine the waterway transport status based on the waterway transport volume model system; among which, the perspectives include: freight volume, turnover volume, flow direction and throughput.

[0066] In some embodiments, a waterway freight volume model, a waterway turnover model, a throughput model, a cargo flow direction model, and a regional waterway freight volume model can be constructed based on the full-chain model of shipping.

[0067] I. For example Figure 3 As shown, a waterway freight volume model can be constructed: The freight volume model aims to accurately calculate the freight volume of each ship in each trip based on the complete shipping chain. The core of the model follows the "first-load, first-unload" principle, that is, the cargo loaded by the ship at different ports is unloaded in the subsequent ports in the order of loading.

[0068] Suppose a complete transport trip T contains n segments, each segment consisting of a departure event. He Jin Gang Incident Composition (i=1,2,3……n,j=1,2,3……n), let Let be the loading volume at the i-th port. Let be the unloading volume of the j-th port.

[0069] The core of the model is to construct a matrix showing the correspondence between loading and unloading volumes. , indicating how much of the cargo loaded at port i is unloaded at port j.

[0070] 1. Calculate cargo transportation data for shipments with no unloading in between: There is no unloading during the intermediate process; the goods are unloaded all at the last port of entry, allowing for accurate identification of the outbound loading volume and the inbound unloading volume. That is:

[0071] At this point, the total unloading volume satisfies:

[0072] 2. Calculate cargo transportation data involving loading and unloading: (1) If the type of goods does not change during a trip, the calculation is performed according to the principle of "first-in, first-out". When there is unloading in the middle process, the recursive relationship is constructed according to the principle of "first-in, first-out".

[0073] (2) The type of goods changes during a trip. Analyzing the changes in the types of goods, it was found that most of the changes were irregular and it was not possible to determine the loading and unloading of the same type of goods. Therefore, the calculation was carried out according to the general principle of "first-in, first-out".

[0074] II. Figure 3 As shown, a waterway transport turnover model can be constructed: The turnover rate is calculated by multiplying the freight volume by the distance. The freight volume can be calculated using a freight volume model, while the distance between ports is calculated using AIS data. Currently, the freight volume model has been built, and the distance between ports needs to be calculated. First, based on AIS ship trajectory data, the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) clustering algorithm is used to reconstruct ship navigation behavior. After associating ship port reporting data and AIS data through MMSI numbers, a series of instantaneous trajectory points of the ship are obtained. The DBSCAN clustering algorithm is used to cluster these trajectory points, identifying the ship's stopping and navigation behaviors, obtaining the latitude and longitude range of the stopping area, and calculating the navigation distance based on the trajectory between two stopping areas.

[0075] Based on the ship cargo volume model, the cargo volume of operating ships is calculated, and then the operating ship turnover is obtained by multiplying the cargo volume by the transport distance between ports. Let the waterway transport distance between port p and port q be... (Unit: km), then the total turnover W is:

[0076] in, The port pairs calculated in the transport volume model Freight volume between them.

[0077] III. Figure 3 As shown, a throughput model can be constructed: 1. Construct a port-province mapping relationship: To ensure the accurate collection and statistical analysis of vessel entry and exit data, this study first comprehensively reviewed the correspondence between maritime administration agencies, ports, and cities based on a provincial vessel entry and exit report database. This database records key information about vessel entry and exit, including reporting agencies, vessel identification numbers, entry and exit times, cargo types, and deadweight tonnage, serving as the foundational data source for constructing throughput models.

[0078] In the data preprocessing stage, the focus is on establishing a mapping dictionary between ship reporting agency codes and port geographical locations. Since reporting agency codes are directly associated with maritime administration agencies, and different agencies may correspond to different terminals or operating areas within the same port area, fine-grained matching is required to ensure that data can be accurately aggregated to the specific port, city, or even province level. The construction of the mapping dictionary follows these principles: (1) Code standardization: All reporting agency codes are processed in a unified format to ensure data consistency; (2) Multi-level matching: Establish a three-level correspondence of “reporting agency code - port (city) - province” to ensure that data can be summarized according to different administrative levels; (3) Outlier handling: For ambiguous or missing mapping relationships, manual verification and correction are carried out in combination with port geographical location, historical data and business logic.

[0079] Based on the provincial vessel entry and exit port report database, a comprehensive review was conducted of the correspondence between maritime administration agencies and ports and cities in the entry and exit port report data. A mapping dictionary between vessel reporting agency codes and port geographical locations was also established, mapping the reporting agency codes to specific ports, cities, and provinces, as shown in Table 1.

[0080] Table 1

[0081] 2. Model Building: Based on the mapping relationship of port terminals and inbound / outbound report data, and following the principle of "one record per vessel," the inbound and outbound volumes of vessels in various ports of a province are summarized. Simultaneously, cargo volume within the ports is excluded to obtain a cargo throughput model for the province's ports. Specifically: (1) Data association and aggregation: The reporting agency codes in the ship entry and exit port report data are matched to the corresponding ports, cities and provinces through the constructed mapping dictionary to realize the geographical location aggregation of the data.

[0082] (2) Statistics on port arrival and departure: According to the port dimension, the port arrival report records and departure report records of all vessels are summarized separately. Each record corresponds to a complete port arrival or departure behavior, and the departure loading volume and arrival unloading volume are calculated separately.

[0083] (3) Port Internal Transport Exclusion: To avoid double counting, the model needs to identify and exclude transport activities within the port. The specific method is as follows: If the port entry reporting agency and the port departure reporting agency of a certain vessel belong to the same port, and the time interval between the two reports is short and the sailing distance is limited, then it is determined to be port internal shifting or short-distance transshipment, and is not included in the port throughput.

[0084] IV. Figure 3 As shown, a cargo flow direction model can be constructed: In some embodiments, a set of shipping chains entering and leaving various regions can be identified based on a full-chain shipping model; a cargo flow direction model can be constructed based on the shipping chains from multiple analytical dimensions; wherein the analytical dimensions include: outflow model, inflow model and internal flow model.

[0085] To accurately depict the characteristics and spatial distribution patterns of inland waterway freight transport between regions, a freight flow direction model is constructed based on a specific region (hereinafter referred to as "the region") and the shipping transport chain. This model, through in-depth analysis and multi-dimensional convergence of shipping behavior, systematically reveals the freight exchange between the region and external regions, as well as within the region itself. The specific construction process is as follows: 1. Identification and extraction of the transportation chain entering and leaving a certain location: Based on the established port-province mapping dictionary, by identifying whether the origin (departure) and destination (arrival) ports of each transport chain belong to a certain location, all shipping activities related to that location are accurately screened, forming a set of shipping chains entering and leaving that location, i.e., shipping chains that depart from, arrive at, and pass through that location. This step is fundamental to distinguishing between internal and external transport, ensuring the accuracy of flow direction analysis, and also provides standardized data units for subsequent cargo aggregation.

[0086] 2. Construction of a multidimensional cargo flow model: Based on the transportation chain set extracted in the first step, the model constructs the following three core analytical dimensions according to the technical route of "flow direction division - cargo type aggregation - OD generation": (1) Outflow model (a certain place → another province): This model aggregates all shipping chains originating from a specific port and destined for ports in other provinces. Starting with "departure from a specific port" and ending with "entry into a port in another province," the model categorizes and summarizes cargo volumes by type, forming an OD matrix showing the cargo volume distribution by type from the specified port to each province. This model visually reflects the port's outward reach, main cargo flow directions, and advantageous outbound cargo types.

[0087] (2) Inflow model (outside province → a certain place): This model aggregates all shipping chains originating from ports in other provinces and destined for a specific location. Starting with "outbound from another province" and ending with "inbound from a specific location," the model aggregates freight volumes by cargo type, generating a cargo volume matrix for each province leading to a specific location. This model reveals the economic hinterland of a given location, its external resource input patterns, and the main types of imported goods it relies on.

[0088] (3) Internal flow model (within a certain place): By eliminating non-essential transportation activities such as intra-port shifting and short-distance transshipment, the model aggregates shipping chains where both the origin and destination ports are located within a specific area. The model aggregates cargo volumes between port areas by cargo type, constructing an intra-regional port-to-origin (OD) cargo volume matrix. This model reflects the economic connections, cargo transshipment, and regional circulation characteristics among the port areas within a given area.

[0089] V. For example Figure 3 As shown, a regional waterway transport volume model can be constructed: In some embodiments, transportation records with at least one end of the shipping chain located within the target area can be screened based on the full-chain model of shipping; the transportation records are aggregated and calculated according to multiple dimensions to generate a regional waterway transportation volume model; wherein, the dimensions include: cargo type, vessel type and time period.

[0090] Regional transport volume refers to the total freight volume originating from a port of call within the target region. Based on a full-chain model of shipping, the system filters out all transport records where at least one end of the shipping chain is located within this region, using this as the statistical boundary. By aggregating and calculating these records according to different dimensions (such as cargo type, vessel type, time period, etc.), a regional waterway transport volume model is finally generated.

[0091] Let the target region be R, and its port set be... Each complete transport trip t contains a series of departure events. He Jin Gang Incident The number of passes is t (i=1,2,3……n,j=1,2,3……n).

[0092] The condition for including trip t in the statistics is: there exists i such that or This means that at least one port of departure or arrival belongs to the region.

[0093] The total transport volume in region R is equal to the sum of the loads of all included trips: ;in, This indicates that the trip was included in the statistics. The sum of the number of trips. Let be the loading quantity of the i-th trip in trip t.

[0094] The method provided in this invention can construct a full-chain model of ship transportation based on big data of ship operations and through systematic data preprocessing, accurately reconstructing the actual transportation path of ships. Furthermore, it establishes a waterway transportation model system including core indicators such as cargo volume, turnover, and throughput, providing reliable data support for monitoring and statistical analysis of waterway transportation conditions using big data.

[0095] The method provided in the embodiments of the present invention has the following main advantages: 1. High degree of data fusion and improved analysis efficiency: Fully explore the information value in the big data of ship operation, improve the efficiency of data use through innovative data fusion and analysis methods, and effectively reduce statistical errors caused by a single data source.

[0096] 2. High degree of automation and improved statistical accuracy: It fully relies on big data technology for analysis and calculation, reduces the dependence on manual statistics, helps control data deviation, and improves management efficiency.

[0097] 3. Diverse analytical dimensions. Based on big data from ship operations, multi-dimensional analytical approaches and diverse analytical perspectives can be developed, effectively improving the validity and comprehensiveness of statistical analysis.

[0098] In summary, the method provided in this embodiment of the invention, based on big data of ship operations, constructs a ship transportation chain model, realizing a complete reconstruction of the entire transportation process from empty ship loading to unloading at port. Building upon this foundation, a waterway transportation model methodology system is further constructed, encompassing freight volume, turnover, throughput, and cargo flow direction, effectively improving the accuracy and consistency of statistical results and providing systematic model support for multi-dimensional statistical analysis and application of inland waterway shipping.

[0099] Example 3: Corresponding to the above method embodiments, this invention provides a device for constructing a water transport statistical model based on big data of ship operations. See [link to relevant documentation]. Figure 7 The diagram shows a structural schematic of a water transport statistical model building device based on ship operation big data. This device includes: The ship operation big data preprocessing module 71 is used to acquire ship operation big data and preprocess it; the ship operation big data includes: ship entry and exit reports and ship automatic identification system data; The ship transportation full-chain model construction module 72 is used to connect data based on ship arrival and departure reports to form ship port reporting data; and to construct a ship transportation full-chain model based on the ship port reporting data. Among them, the ship port reporting data is recorded in the order of departure first and arrival second, and the ship port reporting data is sorted according to the ship identification number and the reporting time based on the ship arrival and departure reports. The ship transportation full-chain model is used to represent the transportation trips of the ship from full loading to empty unloading. The initial position of the ship transportation full-chain model is the initial loading behavior of the ship identified in the ship port reporting data, and the ending position of the ship transportation full-chain model is the arrival and empty unloading behavior of the ship identified in the ship port reporting data. The waterway transport volume model system construction module 73 is used to construct a waterway transport volume model system from multiple perspectives based on the full chain model of ship transport, and to determine the waterway transport status based on the waterway transport volume model system; among which, the perspectives include: freight volume, turnover volume, flow direction and throughput.

[0100] This invention provides a device for constructing a water transport statistical model based on big data of ship operations. By deeply integrating multi-source data, it reconstructs complete ship transport trips and establishes an end-to-end waterway transport volume model system. This waterway transport volume model system can comprehensively measure key indicators such as freight volume, cargo turnover, cargo flow direction, and port throughput, laying a solid technical and model foundation for the refined statistics and analysis of waterway transport.

[0101] The aforementioned ship operation big data preprocessing module is used to clean the data from ship arrival and departure reports; and to standardize and parse the data from the Automatic Identification System (AIS) into structured relational data.

[0102] The aforementioned ship transportation full-chain model construction module is used to generate ship port reporting data by sorting the ship entry and exit reports according to the ship identification number and the reporting time. If there are consecutive exit records or consecutive arrival records in the ship port reporting data, the latest exit record or the latest arrival record is retained in the consecutive exit records or consecutive arrival records, and exit records or arrival records from other times are removed.

[0103] The aforementioned ship transportation full-chain model construction module is used to identify the initial loading behavior of ships from the ship port declaration data, and take the location of the initial loading behavior as the initial position of the ship transportation full-chain model; to identify the unloading behavior of ships upon arrival at port from the ship port declaration data, and take the location of the unloading behavior as the end position of the ship transportation full-chain model; and to remove abnormal data from the ship transportation full-chain model.

[0104] The aforementioned ship transportation full-chain model construction module is used to identify the initial loading behavior of a ship based on a preset first judgment condition from the ship's port declaration data. The first judgment condition includes: the loading volume equals the actual load volume, and the loading volume is not equal to 0; or, the loading volume is not equal to the actual load volume, and the difference between the loading volume and the actual load volume is less than a preset threshold. The aforementioned ship transportation full-chain model construction module is also used to identify the unloading behavior of a ship arriving at port based on a preset second judgment condition from the ship's port declaration data. The second judgment condition includes: the unloading volume equals the actual load volume, and the unloading volume is not equal to 0; or, the unloading volume is not equal to the actual load volume, and the difference between the unloading volume and the actual load volume is less than a threshold.

[0105] The aforementioned ship transportation full-chain model construction module is also used to calculate the data difference rate distribution between loading volume and actual load or unloading volume and actual load through kernel density estimation, determine the distribution peak of the difference rate distribution, and determine the threshold based on the distribution peak.

[0106] The aforementioned waterway transport volume model system construction module is used to construct waterway freight volume models, waterway turnover models, throughput models, cargo flow direction models, and regional waterway transport volume models based on the full chain model of ship transport.

[0107] The aforementioned waterway transport volume model system construction module is used to identify sets of ship transport chains entering and leaving various regions based on the ship transport full chain model; and to construct a cargo flow direction model based on multiple analysis dimensions of the ship transport chains; among which, the analysis dimensions include: outflow model, inflow model and internal flow model.

[0108] The aforementioned waterway transport volume model system construction module is used to screen transport records where at least one end of the shipping chain, either the starting point or the ending point, is located within the target area based on the full chain model of shipping; the transport records are aggregated and calculated according to multiple dimensions to generate a regional waterway transport volume model; the dimensions include: cargo type, vessel type, and time period.

[0109] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described device for constructing a water transport statistical model based on big data of ship operations can be referred to the corresponding process in the aforementioned embodiment of the method for constructing a water transport statistical model based on big data of ship operations, and will not be repeated here.

[0110] Example 4: This invention also provides an electronic device for running the above-described method for constructing a water transport statistical model based on big data from ship operations; see also Figure 8The diagram shows the structure of an electronic device, which includes a memory 100 and a processor 101. The memory 100 stores one or more computer instructions, which are executed by the processor 101 to implement the above-mentioned method for constructing a water transport statistical model based on big data of ship operations.

[0111] Furthermore, Figure 8 The electronic device shown also includes a bus 102 and a communication interface 103, with the processor 101, the communication interface 103 and the memory 100 connected via the bus 102.

[0112] The memory 100 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 103 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc. The bus 102 may be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 8 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0113] Processor 101 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of processor 101 or by instructions in software form. Processor 101 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this invention can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software module can reside in a readily available storage medium in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory 100, and processor 101 reads information from memory 100 and, in conjunction with its hardware, completes the steps of the method described in the foregoing embodiments.

[0114] This invention also provides a computer-readable storage medium storing computer-executable instructions. When these computer-executable instructions are called and executed by a processor, they cause the processor to implement the above-described method for constructing a water transport statistical model based on big data of ship operations. For specific implementation details, please refer to the method embodiments, which will not be repeated here.

[0115] The computer program product of the method and apparatus for constructing a water transport statistical model based on big data of ship operation provided in the embodiments of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods in the preceding method embodiments. For specific implementation, please refer to the method embodiments, which will not be repeated here.

[0116] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system and / or device described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0117] Furthermore, in the description of the embodiments of the present invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in the present invention based on the specific circumstances.

[0118] If the aforementioned functions are implemented as software functional units 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 this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0119] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0120] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for constructing a water transport statistical model based on big data of ship operations, characterized in that, The method includes: Acquire big data on ship operations and preprocess the big data on ship operations; wherein, the big data on ship operations includes: ship entry and exit reports and data from the Automatic Identification System (AIS); The ship arrival and departure reports are linked together to form ship port reporting data. A ship transportation full-chain model is constructed based on the ship port reporting data. The ship port reporting data is recorded in the order of departure first and arrival second, and is sorted by ship identification number and reporting time based on the ship arrival and departure reports. The ship transportation full-chain model is used to represent the number of transport trips a ship undertakes from full loading to empty unloading. The initial position of the ship transportation full-chain model is the initial loading behavior of the ship identified in the ship port reporting data, and the ending position of the ship transportation full-chain model is the arrival and empty unloading behavior of the ship identified in the ship port reporting data. Based on the aforementioned full-chain model of ship transportation, a waterway transportation volume model system is constructed from multiple perspectives, and the waterway transportation status is determined based on the waterway transportation volume model system; wherein, the perspectives include: freight volume, turnover volume, flow direction and throughput.

2. The method according to claim 1, characterized in that, The steps for preprocessing the aforementioned ship operation big data include: Data cleaning is performed on ship arrival and departure reports; The data from the Automatic Identification System (AIS) is standardized and parsed, and then converted into structured relational data.

3. The method according to claim 1, characterized in that, The steps for forming ship port declaration data by concatenating data based on the aforementioned ship arrival and departure reports include: Based on the vessel entry and exit reports, sorted by vessel identification number and reporting time, vessel port reporting data is generated; If the vessel's port reporting data contains consecutive departure records or consecutive arrival records, the latest departure record or the latest arrival record is retained in the consecutive departure records or consecutive arrival records, and departure records or arrival records with other times other than the latest time are removed.

4. The method according to claim 1, characterized in that, The steps for constructing a full-chain model of ship transportation based on the aforementioned ship port declaration data include: The initial loading behavior of a ship is identified from the ship's port declaration data, and the location where the initial loading behavior occurs is used as the initial location of the ship transportation full chain model. Identify the ship arrival and unloading behavior from the ship port reporting data, and take the location where the ship arrival and unloading behavior occurs as the end position of the ship transportation full chain model; Remove outlier data from the aforementioned ship transportation chain model.

5. The method according to claim 4, characterized in that, The steps for identifying the initial loading behavior of a vessel from the vessel's port declaration data include: The initial loading behavior of a vessel is identified from the vessel's port reporting data based on a preset first judgment condition; wherein, the first judgment condition includes: the loading amount is equal to the actual load, and the loading amount is not equal to 0; or, the loading amount is not equal to the actual load, and the difference between the loading amount and the actual load is less than a preset threshold. The steps for identifying vessel arrival and unloading behavior from the vessel port declaration data include: The unloading behavior of a ship is identified from the ship's port reporting data based on a preset second judgment condition; wherein the second judgment condition includes: the unloading volume is equal to the actual load volume, and the unloading volume is not equal to 0; or, the unloading volume is not equal to the actual load volume, and the difference between the unloading volume and the actual load volume is less than the threshold.

6. The method according to claim 5, characterized in that, The method further includes: The data difference rate distribution between the loaded volume and the actual load, or between the unloaded volume and the actual load, is calculated by kernel density estimation, and the peak value of the difference rate distribution is determined. The threshold is determined based on the distribution peak value.

7. The method according to claim 1, characterized in that, The steps for constructing a waterway transport volume model system from multiple perspectives based on the aforementioned full-chain shipping model include: Based on the aforementioned full-chain model of ship transportation, a waterway freight volume model, a waterway turnover model, a throughput model, a cargo flow direction model, and a regional waterway transportation volume model are constructed.

8. The method according to claim 7, characterized in that, The steps for constructing a cargo flow direction model based on the aforementioned ship transportation full-chain model include: Based on the aforementioned full-chain model of ship transportation, identify the sets of ship transportation chains entering and leaving various regions; Based on the aforementioned shipping chain, a cargo flow direction model is constructed from multiple analytical dimensions; wherein, the analytical dimensions include: outflow model, inflow model, and internal flow model.

9. The method according to claim 7, characterized in that, The steps for constructing a regional waterway transport volume model based on the aforementioned full-chain shipping model include: Based on the aforementioned shipping chain model, shipping records in which at least one end of the shipping chain, either the starting point or the ending point, is located within the target area are selected. The transport records are aggregated and calculated according to multiple dimensions to generate a regional waterway transport volume model; wherein, the dimensions include: cargo type, vessel type and time period.

10. A device for constructing a water transport statistical model based on big data of ship operations, characterized in that, The device includes: The ship operation big data preprocessing module is used to acquire ship operation big data and preprocess the ship operation big data; wherein, the ship operation big data includes: ship entry and exit reports and ship automatic identification system data; The ship transportation full-chain model construction module is used to connect data based on the ship arrival and departure reports to form ship port reporting data; and to construct a ship transportation full-chain model based on the ship port reporting data. The ship port reporting data is recorded in the order of departure first, then arrival, and is sorted according to the ship identification number and reporting time based on the ship arrival and departure reports. The ship transportation full-chain model is used to represent the number of transport trips a ship undertakes from full loading to empty unloading. The initial position of the ship transportation full-chain model is the initial loading behavior of the ship identified in the ship port reporting data, and the ending position of the ship transportation full-chain model is the arrival and empty unloading behavior of the ship identified in the ship port reporting data. The waterway transport volume model system construction module is used to construct a waterway transport volume model system from multiple perspectives based on the ship transport full-chain model, and to determine the waterway transport status based on the waterway transport volume model system; wherein, the perspectives include: freight volume, turnover volume, flow direction and throughput.

Images