A port data analysis, prediction and management method based on cloud platform

Abstract

The present invention relates to the technical field of port management, and in particular to a port data analysis, prediction and management method based on a cloud platform, comprising: obtaining historical port data on a cloud platform and classifying them; training all port data to obtain an analysis and prediction model, and determining the influence coefficient of each single variable; obtaining application data of ships to be port-entered, simulating and predicting the expected arrival date of the ships according to the analysis and prediction model, determining the ship berthing type according to the ship cargo data to determine the ship berthing area; determining the management type of the ship in the port according to the information of the ships to be port-entered; determining the ship management method according to the ship management type in the port; determining whether to update the analysis and prediction model according to the expected arrival date of the ship and the actual arrival date of the ship; the present invention manages the port by intelligently identifying the matching of the ship unloading and the cargo to be loaded, and predicting the arrival time of the ship according to the model, thereby optimizing the port operation process and improving the port operation efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of port management, and in particular to a port data analysis, prediction and management method based on a cloud platform. Background Art

[0002] With the development of my country's economy and the expansion of foreign trade, China has become the world's largest cargo trading country, and its port throughput ranks first in the world. Ports have become an important link between inland hinterland and ocean transportation. How to effectively improve the level of port management, reduce resource waste, and realize the intelligence of terminal operations is an urgent problem to be solved. The traditional port operation management model mainly relies on manual experience for dispatching and commanding. This model has great limitations and is prone to human errors or inefficiency. Therefore, in order to improve the level and efficiency of port operations, it is necessary to establish an intelligent management system that can automatically analyze and predict the port operation status so as to discover and solve problems in a timely manner.

[0003] Chinese Patent Publication No.: CN117634926A discloses a comprehensive port operation prediction method and system based on big data analysis, including: obtaining target port business information, performing throughput prediction based on the target port business information, and obtaining throughput prediction information; obtaining port cargo information, analyzing cargo accumulation through the port cargo information, and planning ship loading and unloading according to the analysis results; constructing a digital map of the target port, and planning transportation routes and formulating equipment scheduling plans based on the digital map of the target port; obtaining throughput prediction information, performing revenue prediction and resource utilization analysis for the target port, and recommending operation directions; obtaining operation direction recommendation information, performing situation analysis on the target port, and performing pricing analysis on the recommended operation directions; It can be seen that the comprehensive port operation prediction method and system based on big data analysis have the following problems: it is impossible to mobilize ships according to different cargo loading types, resulting in low space utilization of port operations. Summary of the invention

[0004] To this end, the present invention provides a port data analysis, prediction and management method based on a cloud platform to overcome the problem in the prior art that ships cannot be mobilized according to different cargo loading types, resulting in low space utilization of port operations.

[0005] To achieve the above objectives, the present invention provides a port data analysis, prediction and management method based on a cloud platform, comprising:

[0006] Step S1, obtaining historical port data on the cloud platform, and classifying all port data according to the data source of the port data, the port data type includes ship cargo data, ship operation data and environmental data;

[0007] Step S2, determining a single variable in the port data, training all the port data to obtain an analysis and prediction model, and recording the influence of each single variable on the simulation results of the analysis and prediction model, and determining the influence coefficient of each single variable;

[0008] Step S3, obtaining the application data of the vessel to be port-entered, simulating and predicting the expected arrival date of the vessel according to the analysis and prediction model, and determining the berthing type of the vessel according to the cargo data of the vessel to determine the berthing area of ​​the vessel;

[0009] Step S4, determining the management type of the ship at the port according to the berthing area of ​​the ship to be port-entered, the expected arrival date of the ship, the environmental data after leaving the port, and the cargo data of the ship;

[0010] Step S5, determining the ship management method according to the management type of the ship in the port, including:

[0011] Determine the type and condition of the cargo to be loaded, determine whether the cargo can be loaded directly, and determine whether the vessel can load the cargo normally or adjust the vessel's berthing area based on the determination result;

[0012] or, determining an adjusted berthing area for a vessel according to the vessel cargo data and the amount of congestion in each vessel berthing area;

[0013] Step S6, determining whether to update the analysis and prediction model according to the expected arrival date of the ship and the actual arrival date of the ship.

[0014] Further, the ship cargo data includes ship unloading data, ship loading data and cargo type data;

[0015] The vessel operation data includes the departure date of the vessel, the sailing speed of the vessel, the expected arrival date of the vessel, the actual arrival date of the vessel and the congestion of the waterway;

[0016] The environmental data includes corresponding meteorological data and sea surface data.

[0017] Further, the process of determining the single variable includes,

[0018] Determine the simulation results of each influencing variable in the analysis and prediction model;

[0019] Calculating the correlation coefficients between the influencing variables of the simulation results;

[0020] Determine whether the influencing variables are the same single variable based on the size relationship between the difference between the correlation coefficients and the preset correlation standard;

[0021] If the difference between the correlation coefficients is less than or equal to the preset correlation standard, it is determined that the influencing variables are the same single variable.

[0022] Furthermore, in step S2, determining the influence coefficient of each single variable includes:

[0023] Calculate the simulation results of each single variable in the analysis and prediction model;

[0024] Calculate the simulation result difference between the variable difference of each single variable and the corresponding simulation result;

[0025] The influence coefficient of each single variable is determined based on the difference between the variable and the simulation result.

[0026] Further, in the step S3, the ship berthing area includes a fast berthing area, a regular berthing area and a special berthing area;

[0027] The expected arrival date of the ship is determined according to formula (1):

[0028] ,

[0029] In formula (1), N is the expected arrival date of the ship, N0 is the standard arrival date of the ship, n is the total number of influencing variables, αi is the single variable influence coefficient corresponding to the i-th influencing variable, ki0 is the standard deviation of the i-th influencing variable, and ki is the actual difference of the i-th influencing variable.

[0030] Furthermore, in step S4, determining the management type of the vessel in the port includes:

[0031] Determine whether the vessel meets normal docking and loading and unloading conditions on the expected arrival date of the vessel;

[0032] If the vessel meets the normal berthing and loading and unloading conditions, the management type of the vessel at the port is the regular berthing and loading and unloading management type;

[0033] If the vessel does not meet the normal berthing and loading and unloading conditions, the management type of the vessel at the port is the abnormal berthing and loading and unloading management type;

[0034] Among them, the normal docking and loading and unloading conditions are that the docking area is in the fast docking area or the regular docking area, there is no congestion in the docking areas on that day, and the ship cargo and the cargo to be loaded in the port are in normal condition.

[0035] Further, selecting a ship management method according to the management type of the ship in the port includes:

[0036] If the management type of the vessel at the port is a regular berthing and loading and unloading management type, the selected vessel management method is to determine the type of cargo and cargo conditions to be carried, determine whether the cargo can be directly loaded, and determine whether the vessel can load cargo normally or adjust the berthing area of ​​the vessel according to the determination result;

[0037] If the management type of the ship at the port is the abnormal berthing and loading and unloading management type, the selected ship management method is to determine the adjusted berthing area of ​​the ship according to the ship cargo data and the congestion amount of each ship berthing area.

[0038] Further, determining whether cargo can be loaded directly, and determining the vessel to load cargo normally or adjust the vessel's berthing area based on the determination result, including:

[0039] Data identifying the type of cargo the vessel is unloading and the type of cargo the vessel is to load;

[0040] If the cargo type data of the ship's unloading is the same as the cargo type data of the ship to be loaded, it is determined that the cargo can be loaded directly, and the ship is loaded with cargo normally;

[0041] If the cargo type data of the ship unloading is different from that of the ship to be loaded, it is determined that the cargo cannot be loaded directly, and the ship's berthing area needs to be changed after the ship unloads.

[0042] Further, the adjustment berthing area of ​​the vessel is determined to include,

[0043] Determine the amount of congestion in each vessel docking area;

[0044] The congestion allowable adjustment amount is obtained by subtracting the congestion amount corresponding to the non-special parking area from the free amount in the non-special parking area;

[0045] Determining, according to the congestion allowable adjustment amount, whether to adjust the vessel in a non-docking area or to a special docking area;

[0046] If the congestion allowable adjustment amount is a positive number, it is determined that the ship needs to be adjusted to a special berthing area; if the congestion allowable adjustment amount is a non-positive number, it is determined that the ship needs to be adjusted in a non-berthing area.

[0047] Further, the step of determining whether to update the analysis and prediction model according to the expected arrival date of the ship and the actual arrival date of the ship includes:

[0048] Determine the error rate of simulation results of analytical prediction models;

[0049] Determine eligibility criteria based on historical data combined with simulation results from analytical forecasting models;

[0050] Comparing the simulation result error rate with the qualified standard to determine whether the prediction result of the analysis and prediction model is qualified;

[0051] Among them, if the simulation result error rate is less than or equal to the qualification standard, the prediction result of the analysis and prediction model is judged to be qualified and there is no need to update the analysis and prediction model; if the simulation result error rate is greater than the qualification standard, the prediction result of the analysis and prediction model is judged to be unqualified and there is a need to update the analysis and prediction model.

[0052] Compared with the prior art, the beneficial effect of the present invention lies in that the present invention obtains a prediction model through training by finely analyzing the ship cargo data, ship operation data and environmental data, and by defining the ship berthing area and matching the ship unloading with the type of cargo to be loaded, intelligently decides whether the ship can directly load cargo or needs to adjust the berthing area. In addition, by calculating the congestion and vacancy of each berthing area, the berthing position of the ship is dynamically adjusted to alleviate congestion and improve loading and unloading efficiency. The present invention can automatically evaluate and update the analysis and prediction model according to the error between the expected arrival date and the actual arrival date of the ship to ensure the accuracy and reliability of the prediction. The present invention improves the operating efficiency of the port, reduces resource waste, and enhances the adaptability and response speed to various operating conditions.

[0053] Furthermore, in the present invention, obtaining key information such as ship cargo data, ship operation data and environmental data is crucial for port operations. Analysis based on these data can not only improve the port's operating efficiency and the rationality of resource allocation, but also enhance the accuracy of predictions on ship arrivals and cargo loading and unloading, thereby optimizing port loading and unloading work and reducing congestion.

[0054] Furthermore, the present invention classifies the influencing variables in an inductive manner and classifies the variables with similar influence levels into the same single variable, which can simplify the complexity of the model and reduce the computational burden of the model. In addition, by identifying and merging variables with similar influences, the instability of the model caused by small fluctuations of individual variables can be reduced, thereby improving the efficiency of decision-making and the accuracy of prediction.

[0055] Furthermore, in the present invention, by comparing the types of cargo unloaded by the ship and the cargo to be loaded, it is possible to efficiently determine whether cargo loading can be carried out directly, thereby optimizing the port operation process. When the cargo types are the same, the ship can quickly carry out loading and unloading operations, reducing waiting time and berthing costs, and improving the port's operating efficiency. On the contrary, when the cargo types are different, timely adjustment of the ship's berthing area can ensure that the cargo is loaded and unloaded quickly and safely, avoiding cross contamination and operational delays, while reducing unnecessary occupation of port resources, which not only improves the flexibility and response speed of loading and unloading operations, but also helps to maintain the order and safety of the port.

[0056] Furthermore, in the present invention, by reasonably adjusting the berthing area of ​​the ship, the berthing space can be used more reasonably and the overall operating efficiency of the port can be improved; by calculating the congestion and vacancy, the congestion of the port can be effectively managed and alleviated, ensuring the smooth entry and exit of the ship and also reducing the accident rate in the port operation. In addition, the waiting time of the ship and the congestion can be reduced, which can reduce fuel consumption and emissions, improve loading and unloading efficiency and reduce the occurrence of accidents, which can reduce the operating cost of the port. The present invention can better predict and plan future berthing needs by monitoring the expected arrival volume of ships and the standard congested volume of ships at the port on a daily basis.

[0057] Furthermore, in the present invention, the simulation results of the analysis and prediction model are dynamically updated according to the actual situation, so as to ensure that the model always maintains a high prediction accuracy. By setting the qualification standards, the risks of the prediction model can be effectively managed to ensure that the prediction results are within an acceptable error range. In addition, the setting of the qualification standards can be adjusted according to the port's throughput and actual needs, and can adapt to a wider range of operating environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 This is a flow chart of a port data analysis, prediction and management method based on a cloud platform according to an embodiment of the present invention;

[0059] Figure 2 A flow chart for determining a single variable for an embodiment of the present invention;

[0060] Figure 3 A logic diagram for determining the management type of a ship in a port according to an embodiment of the present invention;

[0061] Figure 4 A logic diagram for determining whether cargo loading can be performed directly according to an embodiment of the present invention. DETAILED DESCRIPTION

[0062] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0063] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0064] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0065] See also Figure 1 As shown, it is a flow chart of a port data analysis, prediction and management method based on a cloud platform according to an embodiment of the present invention; the present invention provides a port data analysis, prediction and management method based on a cloud platform, including

[0066] Step S1, obtaining historical port data on the cloud platform, and classifying all port data according to the data source of the port data, the port data type includes ship cargo data, ship operation data and environmental data;

[0067] Step S2, determining a single variable in the port data, training all the port data to obtain an analysis and prediction model, recording the influence of each single variable on the simulation results of the analysis and prediction model, and determining the influence coefficient of each single variable;

[0068] Step S3, obtaining the application data of the vessel to be port-entered, simulating and predicting the expected arrival date of the vessel according to the analysis and prediction model, and determining the berthing type of the vessel according to the cargo data of the vessel to determine the berthing area of ​​the vessel;

[0069] Step S4, determining the management type of the ship at the port according to the berthing area of ​​the ship to be port-entered, the expected arrival date of the ship, the environmental data after leaving the port, and the cargo data of the ship;

[0070] Step S5, determining the ship management method according to the management type of the ship in the port, including:

[0071] Determine the type and condition of the cargo to be loaded, determine whether the cargo can be loaded directly, and determine whether the vessel can load the cargo normally or adjust the vessel's berthing area based on the determination result;

[0072] or, determining an adjusted berthing area for a vessel according to the vessel cargo data and the amount of congestion in each vessel berthing area;

[0073] Step S6, determining whether to update the analysis and prediction model according to the expected arrival date of the ship and the actual arrival date of the ship.

[0074] In implementation, there is no specific limitation on the process of training all port data to obtain the analysis and prediction model, as long as it can output the ship arrival date through the training of port data. The analysis and prediction model can be implemented through any training method such as linear regression, decision tree, random forest, support vector machine, neural network, etc. This is the existing technology and is not specifically limited.

[0075] The present invention obtains a prediction model through training by finely analyzing the ship cargo data, ship operation data and environmental data, and by defining the ship berthing area and intelligently deciding whether the ship can directly load cargo or needs to adjust the berthing area according to the matching situation of the ship unloading and the type of cargo to be loaded. In addition, by calculating the congestion and vacancy of each berthing area, the berthing position of the ship is dynamically adjusted to alleviate congestion and improve loading and unloading efficiency. The present invention can automatically evaluate and update the analysis and prediction model according to the error between the expected arrival date and the actual arrival date of the ship to ensure the accuracy and reliability of the prediction. The present invention improves the operating efficiency of the port, reduces resource waste, and enhances the adaptability and response speed to various operating conditions.

[0076] Specifically, the ship cargo data includes ship unloading data, ship loading data and cargo type data;

[0077] The vessel operation data includes the departure date of the vessel, the sailing speed of the vessel, the expected arrival date of the vessel, the actual arrival date of the vessel and the congestion of the waterway;

[0078] The environmental data includes corresponding meteorological data and sea surface data.

[0079] In implementation, meteorological data include wind speed and direction, temperature and humidity, and visibility; sponge data include wave height (high waves may cause ship instability or increase navigation difficulty, affecting the navigation safety of the ship), ocean current strength and direction (ocean currents may affect the ship's navigation route and speed, affect the ship's fuel consumption and sailing time), and sea ice area and thickness.

[0080] In the present invention, obtaining key information such as ship cargo data, ship operation data and environmental data is crucial for port operations. Analysis based on these data can not only improve the port's operating efficiency and the rationality of resource allocation, but also enhance the accuracy of predictions on ship arrivals and cargo loading and unloading, thereby optimizing port loading and unloading work and reducing congestion.

[0081] See also Figure 2 As shown, it is a flow chart of determining a single variable according to an embodiment of the present invention. The process of determining the single variable includes:

[0082] Determine the simulation results of each influencing variable in the analysis and prediction model;

[0083] Calculating the correlation coefficients between the influencing variables of the simulation results;

[0084] Determine whether the influencing variables are the same single variable based on the size relationship between the difference between the correlation coefficients and the preset correlation standard;

[0085] If the difference between the correlation coefficients is less than or equal to the preset correlation standard, it is determined that the influencing variables are the same single variable.

[0086] In implementation, in the process of determining the influence of each influencing variable on the arrival date of the ship, when determining the simulation results of each influencing variable in the analysis and prediction model, other influencing variables are regarded as constants and have the same value, so as to avoid the influence of other variables on the simulation results;

[0087] The correlation coefficient between the simulation results and the corresponding influencing variables is the Pearson correlation coefficient. Influencing variables with similar correlation coefficients can be judged as having similar influences on the expected arrival date of the ship. In this case, the two types of variables are regarded as the same single variable;

[0088] In implementation, the absolute value of the difference between correlation coefficients is compared with the preset correlation standard. The preset correlation standard is selected within the interval [0.05, 0.1]. If the difference between the correlation coefficients corresponding to the influencing variables is less than or equal to the preset correlation standard, then these influencing variables are determined to be the same single variable.

[0089] The process of using the Pearson correlation coefficient to determine the single variable that affects the arrival date of the ship, classifying the way in which the influencing variables are summarized, and classifying the variables with similar influence levels into the same single variable can simplify the complexity of the model and reduce the computational burden of the model. In addition, by identifying and merging variables with similar influences, the instability of the model caused by small fluctuations in individual variables can be reduced, which improves the efficiency of decision-making and the accuracy of predictions.

[0090] Specifically, in step S2, determining the influence coefficient of each single variable includes:

[0091] Calculate the simulation results of each single variable in the analysis and prediction model;

[0092] Calculate the simulation result difference between the variable difference of each single variable and the corresponding simulation result;

[0093] The influence coefficient of each single variable is determined based on the difference between the variable and the simulation result.

[0094] In implementation, the influence coefficient of a single variable is determined by comparing the difference between the simulation results and the difference between the single variable (wherein the difference is the absolute value after the difference, calculated without dimension). The weight is determined by comparing the impact of the change in the variable on the model output, thereby accurately evaluating the impact of each variable on the simulation results. This can enhance the flexibility and adaptability of the model and improve the accuracy of the model in predicting the expected arrival date of the ship.

[0095] See also Figure 3 As shown, it is a logic diagram for determining the management type of a ship in a port according to an embodiment of the present invention. In the step S3, the ship berthing area includes a fast berthing area, a regular berthing area and a special berthing area;

[0096] The expected arrival date of the ship is determined according to formula (1):

[0097] ,

[0098] In formula (1), N is the expected arrival date of the ship, N0 is the standard arrival date of the ship, n is the total number of influencing variables, αi is the single variable influence coefficient corresponding to the i-th influencing variable, ki0 is the standard deviation of the i-th influencing variable, and ki is the actual difference of the i-th influencing variable.

[0099] In implementation, ki0 takes the value of 1, which is the influence of a change in the influencing variable on the expected arrival date of the ship through the corresponding influence coefficient. The value of ki can be positive or negative, reflecting that a single variable has positive and negative effects on navigation, such as sailing with the wind and sailing against the wind.

[0100] The standard arrival date of the ship can be directly obtained in the analysis and prediction model. The expected arrival date of the ship when there is no variable influence is the standard arrival date of the ship. It can be understood that the ship unloading data, ship loading data and cargo type data determine the time when the ship needs to arrive at or leave the port to avoid the deterioration of fresh goods. The meteorological factors, sea surface factors and channel congestion during the voyage of the ship have a greater impact on the navigation process.

[0101] Cargo type data includes fresh cargo, non-fresh cargo and special cargo. Fresh cargo such as fruits and vegetables, meat or seafood need to be guaranteed fresh. Non-fresh cargo such as frozen food, dry goods and processed food have a long shelf life. Special cargo such as dangerous goods, live goods and controlled items need special quarantine. Before the ship departs, the type of cargo unloaded by the ship and the type of cargo loaded by the ship are confirmed. The cargo type and pre-classification mark are in the ship cargo data and will not be repeated here.

[0102] Among them, the berthing type of ships transporting fresh goods is the fast berthing type, the berthing type of ships transporting non-fresh goods is the regular berthing type, and the berthing type of ships transporting special goods is the special berthing type;

[0103] The docking area for a fast docking type vessel is a fast docking area, the docking area for a regular docking type vessel is a regular docking area, and the docking area for a special docking type vessel is a special docking area.

[0104] Specifically, in step S4, determining the management type of the ship in the port includes:

[0105] Determine whether the vessel meets normal docking and loading and unloading conditions on the expected arrival date of the vessel;

[0106] If the vessel meets the normal berthing and loading and unloading conditions, the management type of the vessel at the port is the regular berthing and loading and unloading management type;

[0107] If the vessel does not meet the normal berthing and loading and unloading conditions, the management type of the vessel at the port is the abnormal berthing and loading and unloading management type;

[0108] Among them, the normal docking and loading and unloading conditions are that the docking area is in the fast docking area or the regular docking area, there is no congestion in the docking areas on that day, and the ship cargo and the cargo to be loaded in the port are in normal condition.

[0109] In practice, whether congestion occurs in the berthing area can be determined based on the number of ships expected to arrive at the port each day and the standard number of ships that are congested; when the number of ships expected to arrive at the port each day is less than or equal to the standard number of ships that are congested, it is determined that there is no congestion;

[0110] The standard congested vessel quantity is determined within a range of 0.9 to 1.1 times the maximum number of vessels that appear under the condition of normal operation of each berthing area in the historical port data; whether the goods are in a normal state is determined based on the transportation standards of various goods in the goods type data, such as whether fresh goods are spoiled, whether non-fresh goods are damaged in packaging due to improper storage, etc., which does not meet the conditions for unloading of the goods at the destination. This is a prior art and will not be elaborated on here.

[0111] Specifically, the ship management method selected according to the type of management of the ship in the port includes:

[0112] If the management type of the vessel at the port is a regular berthing and loading and unloading management type, the selected vessel management method is to determine the type of cargo and cargo conditions to be carried, determine whether the cargo can be directly loaded, and determine whether the vessel can load cargo normally or adjust the berthing area of ​​the vessel according to the determination result;

[0113] If the management type of the ship at the port is the abnormal berthing and loading and unloading management type, the selected ship management method is to determine the adjusted berthing area of ​​the ship according to the ship cargo data and the congestion amount of each ship berthing area.

[0114] See also Figure 4 As shown, it is a logic diagram of determining whether cargo loading can be directly performed in an embodiment of the present invention, determining whether cargo loading can be directly performed, and determining whether the ship can load cargo normally or adjust the ship's berthing area according to the determination result, including:

[0115] Data identifying the type of cargo the vessel is unloading and the type of cargo the vessel is to load;

[0116] If the cargo type data of the ship's unloading is the same as the cargo type data of the ship to be loaded, it is determined that the cargo can be loaded directly, and the ship is loaded with cargo normally;

[0117] If the cargo type data of the ship unloading is different from that of the ship to be loaded, it is determined that the cargo cannot be loaded directly, and the ship's berthing area needs to be changed after the ship unloads.

[0118] In implementation, for example, if a ship is in a fast docking area, and the type of cargo unloaded by the ship is fresh cargo, and the type of cargo to be loaded by the ship is also fresh cargo according to the application data of the ship, there is no need to adjust the docking area of ​​the ship, and cargo can be loaded directly in the current area;

[0119] If the ship is in a regular berthing area, and according to the ship's application data, the type of cargo to be loaded on the ship is fresh cargo, and the types of cargo before and after are different, the unloaded ship will be docked in a fast berthing area.

[0120] In the present invention, by comparing the types of cargo unloaded by the ship and the cargo to be loaded, it is possible to efficiently determine whether cargo loading can be carried out directly, thereby optimizing the port operation process. When the cargo types are the same, the ship can quickly carry out loading and unloading operations, reducing waiting time and berthing costs, and improving the port's operating efficiency. On the contrary, when the cargo types are different, timely adjustment of the ship's berthing area can ensure that the cargo is loaded and unloaded quickly and safely, avoiding cross contamination and operational delays, while reducing unnecessary occupation of port resources, which not only improves the flexibility and response speed of loading and unloading operations, but also helps to maintain the order and safety of the port.

[0121] Specifically, the adjustment of the berthing area for the vessel is determined to include:

[0122] Determine the amount of congestion in each vessel docking area;

[0123] The congestion allowable adjustment amount is obtained by subtracting the congestion amount corresponding to the non-special parking area from the free amount in the non-special parking area;

[0124] Determining, according to the congestion allowable adjustment amount, whether to adjust the vessel in a non-docking area or to a special docking area;

[0125] If the congestion allowable adjustment amount is a positive number, it is determined that the ship needs to be adjusted to a special berthing area; if the congestion allowable adjustment amount is a non-positive number, it is determined that the ship needs to be adjusted in a non-berthing area.

[0126] In practice, the congestion volume is determined based on the difference between the expected number of ships arriving at the port each day and the standard number of congested ships, or is set based on actual scenarios;

[0127] Before adjusting the ships, the ships that have already docked in the original area will be integrated and the vacant berthing spaces will be adjusted to one area to reduce the uncontrollability of loading and unloading cargoes, improve loading and unloading efficiency, and reduce the occurrence of accidents.

[0128] For example, if there are 10 vacancies in the fast docking area and 5 congestion vacancies in the regular ship docking area, the congested ship needs to be docked in the fast docking area; if there are 10 vacancies in the fast docking area and 12 congestion vacancies in the regular ship docking area, the congested ship needs to be docked in the special docking area.

[0129] In the present invention, by reasonably adjusting the berthing area of ​​the ship, the berthing space can be used more reasonably and the overall operation efficiency of the port can be improved; by calculating the congestion and vacancy, the congestion of the port can be effectively managed and alleviated, ensuring the smooth entry and exit of the ship and also reducing the accident rate in the port operation. In addition, the waiting time of the ship and the congestion can be reduced, which can reduce fuel consumption and emissions, improve loading and unloading efficiency and reduce the occurrence of accidents, which can reduce the operating cost of the port. The present invention can better predict and plan future berthing needs by monitoring the expected arrival volume of ships and the standard congested ship volume of the port on a daily basis.

[0130] Specifically, the steps of determining whether to update the analysis and prediction model according to the expected arrival date of the ship and the actual arrival date of the ship include:

[0131] Determine the error rate of simulation results of analytical prediction models;

[0132] Determine eligibility criteria based on historical data combined with simulation results from analytical forecasting models;

[0133] Comparing the simulation result error rate with the qualified standard to determine whether the prediction result of the analysis and prediction model is qualified;

[0134] Among them, if the simulation result error rate is less than or equal to the qualification standard, the prediction result of the analysis and prediction model is judged to be qualified and there is no need to update the analysis and prediction model; if the simulation result error rate is greater than the qualification standard, the prediction result of the analysis and prediction model is judged to be unqualified and there is a need to update the analysis and prediction model.

[0135] In practice, the expected arrival date of the ship and the actual arrival date of the ship are subtracted, and the error rate of the simulation result is determined by the following formula:

[0136] Simulation result error rate ,

[0137] Among them, m is the error rate of the simulation result, N1 is the actual arrival date of the ship, N is the expected arrival date of the ship, 0<m≤1. It can be understood that the actual arrival date of the ship and the expected arrival date of the ship are not specific dates, but dates from the departure date of the ship. In essence, they are time lengths, and the dates may not be integers.

[0138] In practice, the qualified standard can be determined based on the data range of the error rate corresponding to the error date of 0.5 to 1 day; generally, when the throughput of the port is large, the error date for determining the qualified standard is smaller.

[0139] The updated prediction model is trained to redefine single variables and their weights in order to supplement the port data corresponding to the large error rate of the simulation results.

[0140] In the present invention, the simulation results of the analysis and prediction model are dynamically updated according to the actual situation, so as to ensure that the model always maintains a high prediction accuracy. By setting the qualification standards, the risks of the prediction model can be effectively managed to ensure that the prediction results are within an acceptable error range. In addition, the setting of the qualification standards can be adjusted according to the throughput and actual needs of the port, and can adapt to a wider range of operating environments.

[0141] If the port data analysis, prediction and management method based on the cloud platform of the present invention is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or 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, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0142] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0143] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A port data analysis, prediction and management method based on a cloud platform, characterized in that: include: Step S1, obtaining historical port data on the cloud platform, and classifying all port data according to the data source of the port data, the port data type includes ship cargo data, ship operation data and environmental data; Step S2, determining a single variable in the port data, training all the port data to obtain an analysis and prediction model, recording the influence of each single variable on the simulation results of the analysis and prediction model, and determining the influence coefficient of each single variable; Step S3, obtaining the application data of the vessel to be docked, predicting the expected arrival date of the vessel based on the analysis and prediction model, determining the type of vessel to be docked based on the cargo data of the vessel and determining the docking area of ​​the vessel, In step S3, the expected arrival date of the ship is determined according to formula (1): In formula (1), N is the expected arrival date of the ship, N0 is the standard arrival date of the ship, n is the total number of influencing variables, αi is the single variable influence coefficient corresponding to the i-th influencing variable, ki0 is the standard deviation of the i-th influencing variable, and ki is the actual difference of the i-th influencing variable; Step S4, determining the management type of the ship at the port according to the berthing area of ​​the ship to be port-entered, the expected arrival date of the ship, the environmental data after leaving the port, and the cargo data of the ship; Step S5, determining the ship management method according to the management type of the ship in the port, including: Determine the type and condition of the cargo to be loaded, determine whether the cargo can be loaded directly, and determine whether the vessel can load the cargo normally or adjust the vessel's berthing area based on the determination result; or, determining an adjusted berthing area for a vessel according to the vessel cargo data and the amount of congestion in each vessel berthing area; Step S6, determine whether to update the analysis and prediction model according to the expected arrival date of the ship and the actual arrival date of the ship. If the error rate of the simulation result is greater than the qualified standard, the prediction result of the analysis and prediction model is determined to be unqualified, and the analysis and prediction model needs to be updated. The updated prediction model is to supplement the port data corresponding to the simulation result error rate greater than the qualified standard for training and re-determine the single variable and the corresponding influence coefficient, wherein the qualified standard is determined by taking the value of the error date within the data range of the error rate corresponding to 0.5 to 1 day, and the simulation result error rate is determined by the following formula: Simulation result error rate , Where m is the error rate of the simulation result, N1 is the actual arrival date of the ship, N is the expected arrival date of the ship, 0<m≤1; The vessel cargo data includes vessel unloading data, vessel cargo loading data and cargo type data; The vessel operation data includes the departure date of the vessel, the sailing speed of the vessel, the expected arrival date of the vessel, the actual arrival date of the vessel and the congestion of the waterway; The environmental data includes corresponding meteorological data and sea surface data; In step S4, determining the management type of the vessel in the port includes: Determine whether the vessel meets normal docking and loading and unloading conditions on the expected arrival date of the vessel; If the vessel meets the normal berthing and loading and unloading conditions, the management type of the vessel at the port is the regular berthing and loading and unloading management type; If the vessel does not meet the normal berthing and loading and unloading conditions, the management type of the vessel at the port is the abnormal berthing and loading and unloading management type; The normal berthing and loading and unloading conditions are that the berthing area is in the fast berthing area or the regular berthing area, and there is no congestion in the berthing areas on that day, and the ship cargo and the cargo to be loaded at the port are in normal condition; The vessel management method selected according to the type of management of the vessel in the port includes: If the management type of the vessel at the port is a regular berthing and loading and unloading management type, the selected vessel management method is to determine the type of cargo and cargo conditions to be carried, determine whether the cargo can be directly loaded, and determine whether the vessel can load cargo normally or adjust the berthing area of ​​the vessel according to the determination result; If the management type of the ship at the port is the abnormal berthing and loading and unloading management type, the selected ship management method is to determine the adjusted berthing area of ​​the ship according to the cargo data of the ship and the congestion amount of each ship berthing area; Determine whether cargo can be loaded directly, and determine whether the ship can load cargo normally or adjust the ship's berthing area based on the determination results, including: Data identifying the type of cargo the vessel is unloading and the type of cargo the vessel is to load; If the cargo type data of the ship's unloading is the same as the cargo type data of the ship to be loaded, it is determined that the cargo can be loaded directly, and the ship is loaded with cargo normally; If the cargo type data of the ship unloading is different from that of the ship to be loaded, it is determined that the cargo cannot be loaded directly, and the ship's berthing area needs to be changed after the ship unloads.

2. The port data analysis, prediction and management method based on the cloud platform according to claim 1 is characterized in that: The process of determining the single variable includes, Determine the simulation results of each influencing variable in the analysis and prediction model; Calculating the correlation coefficients between the influencing variables of the simulation results; Determine whether the influencing variables are the same single variable based on the size relationship between the difference between the correlation coefficients and the preset correlation standard; If the difference between the correlation coefficients is less than or equal to the preset correlation standard, it is determined that the influencing variables are the same single variable.

3. The port data analysis, prediction and management method based on the cloud platform according to claim 2 is characterized in that: In step S2, determining the influence coefficient of each single variable includes: Calculate the simulation results of each single variable in the analysis and prediction model; Calculate the simulation result difference between the variable difference of each single variable and the corresponding simulation result; The influence coefficient of each single variable is determined based on the difference between the variable and the simulation result.

4. The port data analysis, prediction and management method based on the cloud platform according to claim 3 is characterized in that: In step S3, the ship berthing area includes a fast berthing area, a regular berthing area and a special berthing area.

5. The port data analysis, prediction and management method based on cloud platform according to claim 1 is characterized in that: Determine the adjusted docking area for the vessel including, Determine the amount of congestion in each vessel docking area; The congestion allowable adjustment amount is obtained by subtracting the congestion amount corresponding to the non-special parking area from the free amount in the non-special parking area; Determining, according to the congestion allowable adjustment amount, whether to adjust the vessel in a non-docking area or to a special docking area; If the congestion allowable adjustment amount is a positive number, it is determined that the ship needs to be adjusted to a special berthing area; if the congestion allowable adjustment amount is a non-positive number, it is determined that the ship needs to be adjusted in a non-berthing area.

6. The port data analysis, prediction and management method based on cloud platform according to claim 5 is characterized in that: The steps of determining whether to update the analysis and prediction model according to the expected arrival date of the ship and the actual arrival date of the ship include: Determine the error rate of simulation results of analytical prediction models; Determine eligibility criteria based on historical data combined with simulation results from analytical forecasting models; Comparing the simulation result error rate with the qualified standard to determine whether the prediction result of the analysis and prediction model is qualified; Among them, if the error rate of the simulation result is less than or equal to the qualification standard, the prediction result of the analysis and prediction model is determined to be qualified, and there is no need to update the analysis and prediction model.

Images