Shipborne shipping information management system

Abstract

The invention relates to the technical field of shipping information management, in particular to a shipborne shipping information management system. The method comprises the following steps: predicting the fault risk time when a ship sails on the sea by introducing the change condition of seasonal influence on the ship, and jointly judging the maintenance time of the ship under the fault risk time by combining the fault risk time with the correlation characteristics of the ship, including the fault type, position, health state and maintenance parts. Then the fault risk time and the maintenance time are introduced into the constructed linear regression model to accurately predict the estimated time when the ship arrives at the next port and station again, so that the ship can be conveniently filled into a shipping information management system in combination with the position of the ship, and therefore, the ports, stations and the like corresponding to the ship can make replacement and planning preparation for shipborne shipping, and the shipping efficiency is improved. And the shipborne shipping efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of shipping information management, and in particular to a shipborne shipping information management system. Background Art

[0002] In the process of shipborne shipping, it is necessary to monitor, analyze and predict the real-time information status of shipborne shipping, and obtain the information data of future shipborne shipping so as to fill it into the subsequent shipping information system. The information data includes the current shipborne shipping position, time, estimated time of arrival at the next target location, etc. This can enable the coordinated allocation of tasks among multiple subsequent ship stations (for example, by determining the estimated time, the ship can plan the cargo and space stored on it in advance when it arrives at the next port or terminal, which is convenient for the subsequent ship to arrive at the station, take over, etc.), and can also output an estimated arrival situation of the cargo to the user.

[0003] However, at present, when predicting the estimated time of ships in shipborne shipping, personnel usually make reference assessments based on weather and distance, which leads to a lack of consideration of the actual conditions of the sailing ships when judging the estimated time. Therefore, it is difficult to predict the accurate estimated time of the ship's arrival at the target location under actual circumstances in the future, resulting in the inability of ports and stations intersecting with them to implement accurate shipping planning plans, affecting the transportation efficiency of shipborne shipping. Summary of the invention

[0004] In view of the above-mentioned shortcomings of the prior art, the present invention provides a shipborne shipping information management system, which can effectively solve the problem in the prior art that the estimated time for shipborne shipping to arrive at the target location is inaccurate, resulting in the inability of ports and stations connected with shipping to accurately implement shipping planning.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0006] The present invention provides a shipborne shipping information management system, including a shipping information management module, which is used to obtain the status information of the shipborne shipping, clarify the node position of the current vehicle, the current time and the estimated time to reach the next stage node, and fill the status information into the shipping information management system, and also includes:

[0007] The voyage time prediction module is used to build a linear regression model based on the historical transportation speed, traffic conditions, weather conditions, road conditions, and cargo status information of the vehicle under the node shipping path to predict the estimated time for the vehicle to arrive at the next stage node, where:

[0008] By introducing the influence of seasonality and combining it with the time series model, the failure risk time of the vehicle during the subsequent voyage is predicted, and the maintenance time is determined based on the historical status of the failure risk time. The determined failure risk time and maintenance time are used as inputs of the linear regression model to regain the estimated time and fill it into the shipping information management system.

[0009] Furthermore, the weather condition is obtained by weighted summation of weather temperature, weather humidity, weather wind speed, rainfall and snowfall;

[0010] The cargo status information is obtained by weighted summation of the current cargo temperature, cargo humidity, cargo type and cargo shape;

[0011] The road condition information is obtained by weighted summation of the channel width, channel depth, channel complexity and wave height of the node shipping path.

[0012] Furthermore, the algorithm expression of the linear regression model constructed by historical transportation speed, traffic conditions, weather conditions, road conditions, and cargo status information is:

[0013] T eta =β 0 +V hiaca β 1 +C traf β 2 +W eath β 3 +T emp β 4 +S lope β 5 +ε

[0014] Among them, T eta is the estimated time to reach the next stage node, V hiaca is the historical transportation speed, C traf For traffic conditions, W eath is the weather condition, T emp is the cargo status information, S lope is the traffic information, β 1 , β 2 , β 3 , β 4 , β 5 are the corresponding weight coefficients, β 0 is the intercept term and ε is the error term.

[0015] Furthermore, the method for determining the failure risk time is:

[0016] Obtain the historical fault data set of the vehicle, which contains the fault records of the vehicle in the past, represented as fault time series data;

[0017] Perform seasonal difference on the original fault time series data to remove seasonal fluctuations in the data and make the data stable. The expression of seasonal difference is:

[0018] y″ t =y t -y t -s

[0019] Among them, y″ t represents the data after seasonal difference, y t-s Represents the failure time data at time point ts;

[0020] Constructing a time series model based on seasonal cycles, we have:

[0021] Φ s (B s )·(1-B s ) D ·(1-B s ) d ·y t =Θ s (B s )·ε t

[0022] Among them, Φ s (B s ) represents the seasonal autoregressive part, (1-B s ) D represents the seasonal difference part, (1-B s ) d represents the non-seasonal difference part, B represents the lag operator, and B s represents the seasonal lag, Θ s (B s ) represents the seasonal moving average part, D represents the order of seasonal difference, d represents the order of non-seasonal difference, ε t represents the noise term;

[0023] Predict vehicle failure risk time:

[0024]

[0025] in, represents the predicted failure risk time in the next h steps, φ 1 ,φ 2 ,···φ p They represent the parameters of the autoregressive part, θ 1 ,···θ q They represent the parameters of the moving average part, ε t +h-1,ε t +h-2,···εt +hq respectively represent the prediction errors.

[0026] Furthermore, after the fault risk time is determined, the fault type of the vehicle is determined. The specific method is as follows:

[0027] Get the vehicle's failure risk time The future weather conditions e t and the load L at that time oad ;

[0028] Obtain the fault type and historical weather conditions W′ corresponding to the historical fault time in the database et And the historical load L′ oad , determine the weather state W et 、Load L oad The same historical weather conditions W′ et And the historical load L′ oad , marking the corresponding fault type as the first fault type and constructing a fault set, thereby determining the fault types at adjacent times in the fault set;

[0029] Determine the time intervals between multiple fault types to build interval sequence groups and determine the fault risk time The risk time interval between the previous historical fault time is used to determine the corresponding time interval in the interval sequence group, and the historical fault type existing in the corresponding time interval is determined as the fault risk time. The fault type.

[0030] Furthermore, the maintenance time is determined as follows:

[0031] Obtain relevant characteristics of the marine environment, including the risk of failure Wave height, wind speed, and weather conditions;

[0032] Get vehicle-related features, including:

[0033] Fault type and fault location;

[0034] and, the health status of the vehicle;

[0035] Identify the repair parts used in the fault type history and determine whether the corresponding repair parts on the current vehicle are sufficient. If not, calculate the repair parts based on the fault risk time. The fault coordinates corresponding to the vehicle are used to obtain other associated vehicles that are closest to the vehicle and pass through the fault coordinates, and whether the required maintenance components are available on the other associated vehicles. If available, the other associated vehicles are recorded as target other associated vehicles, and the response distance between the other associated vehicles and the fault coordinates is determined, as well as the deployment impact status of deploying the maintenance components to the vehicle with the current fault. The deployment impact status is determined by weighted summation of weather, ocean conditions and response distance.

[0036] The wave height, wind speed, weather conditions, fault type, fault location, health status, response distance and deployment impact status are used to build a linear regression model to obtain the maintenance time R pat .

[0037] Furthermore, when the failure risk time and the maintenance time are used as the input of the linear regression model to regain the estimated time, the algorithm expression of the linear regression model is:

[0038]

[0039] Among them, β 6 , β 7 are the corresponding weight coefficients respectively.

[0040] Furthermore, the impact levels of different weather and ocean conditions are obtained respectively, and corresponding numerical values ​​are assigned and weighted summed with the response distance to determine the deployment impact status.

[0041] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements any one of the above-mentioned systems when executing the computer program.

[0042] A computer-readable storage medium stores a computer program, wherein the computer program implements any one of the above-mentioned systems when executed by a processor.

[0043] Compared with the known prior art, the technical solution provided by the present invention has the following beneficial effects:

[0044] By determining the current location of the ship and the location to be reached, a linear regression model is constructed based on historical transportation speed, traffic conditions, weather conditions, road conditions, and cargo status information to achieve a preliminary prediction of the estimated time for the ship to arrive at the designated location;

[0045] By introducing the changes in ships affected by seasonality, the risk time of failure that may occur when ships are sailing on the ocean is predicted, and the associated characteristics of the risk time of failure and the ship, including the type of failure, location, health status and repair parts, are combined to jointly determine the repair time of the ship under the risk time of failure. Then, the risk time of failure and the repair time are introduced into the constructed linear regression model to make an accurate prediction of the estimated time for the ship to arrive at the next port or station, which is convenient for filling in the shipping information management system in combination with the location of the ship, so that the ports and stations corresponding to the ship can make preparations for the succession and planning of ship-borne shipping, thereby improving the transportation efficiency of ship-borne shipping. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0047] Figure 1 It is the overall module block diagram of the present invention. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] The present invention will be further described below in conjunction with the embodiments.

[0050] Example 1 (see Figure 1 ): Shipborne shipping information management system, including at least:

[0051] The shipping information management module is used to monitor the status information of shipborne shipping. The status information includes the current node position of the shipborne shipping (vehicle), the current time and the estimated time to reach the next position (stage node). It is worth noting that in the shipborne shipping information management system, the shipping path is generated according to the origin and destination of this shipborne shipping, and multiple stage nodes are preset in the shipping path (usually the transit positions in the current shipping cargo transportation process), and each stage node needs to fill in the current shipborne shipping status information.

[0052] In the above, the current position and time of the ship-borne shipping are usually obtained based on the positioning equipment installed on the vehicle, that is, the real-time position and time of the vehicle can be clearly determined, so that when the stage node of the corresponding position is reached, the corresponding status information is generated and input into the shipping information management system, so that the back-end management personnel can monitor the specific situation of the ship-borne shipping in real time.

[0053] However, it should be noted that when the vehicle arrives at the stage node and in the process of sailing, it is also necessary to generate an estimated time to reach the next location, that is, the time to reach the next stage node, so that the next stage node can make corresponding preparations for cargo replacement and other related steps according to the output estimated time, improve the coordination between ship-borne shipping and improve shipping efficiency. Therefore, in order to determine the estimated time to reach the next stage node, it also includes a sailing time prediction module, which includes a correlation data analysis unit, a sailing time prediction unit and a sailing time optimization unit;

[0054] The associated data analysis unit is used to obtain the related data respectively:

[0055] Get the historical transport speed V hiaca The historical transportation speed is determined according to the node shipping path between the current stage node or position of the vehicle and the next stage node and the database to determine the corresponding multiple historical transportation speeds, and the multiple matching historical transportation speeds are screened according to the weather conditions in the node shipping path and the average value is obtained;

[0056] Get the traffic conditions of the future node shipping path of the current vehicle (the navigation path between the current position of the vehicle and the next stage node) C traf , including traffic flow (vehicles / hour) and the number of traffic and shipping accidents. Therefore, different weights are assigned according to traffic flow and the number of traffic accidents (traffic situation C traf In the above equation, the sum of the weights of traffic flow and traffic accident number is 1. The sum of the weights corresponding to other characteristic parameters below is 1. Similarly, the traffic situation of the current node shipping path is obtained by weighted summation.

[0057] Get the weather conditions W of the node shipping path where the current vehicle will be in the future eath , including weather temperature, weather humidity, weather wind speed, rainfall, snowfall (for bad weather, the vehicle's driving speed is usually reduced, thereby increasing the transportation time). Similarly to the above, the weather conditions of the current node shipping route are obtained by assigning different weights to specific parameters in the weather data;

[0058] Get the cargo status information T of the cargo in the current vehicle emp(The status information of the goods may affect the transportation conditions during transportation. For example, some types of goods may need to be transported under specific temperature and humidity conditions. Too high or too low a temperature may cause damage to the goods, thus affecting the smooth progress of the transportation process), including obtaining the current temperature and humidity of the goods (for perishable goods or goods that require constant temperature and humidity, large fluctuations in temperature and humidity during transportation may cause damage to the goods, thereby extending the transportation time and may require detours or delayed processing, etc.), obtaining the vibration frequency borne in the previous node shipping path (if the goods are subjected to large vibrations, it may affect the safety of the goods or require additional time for inspection or packaging adjustment), the type of goods (including fragile, pressure-resistant, chemicals, etc., among which fragile or dangerous goods may require more careful handling or slower transportation speeds to ensure the safety of the goods) and the shape of the goods, thereby assigning corresponding weights according to the temperature, vibration frequency, type and shape of the goods (generally speaking, the more dangerous and more easily damaged the goods are, the higher the weight is, which can be set according to the actual situation) and weighted summing to obtain the cargo status information;

[0059] Get the road condition information S of the future node shipping path of the current vehicle lope , including the channel width, channel depth, channel complexity (obtained by summing the number of bends and the total length of multiple bends) and wave height (the height of the wave, i.e. the vertical distance between the crest and the trough, the maximum wave height is determined by the wave data currently collected in the node shipping path), thereby assigning corresponding weights and weighted summing to obtain the road condition information;

[0060] Then, the voyage time prediction unit can calculate the historical transportation speed V according to the above-defined hiaca Traffic conditions C traf 、Weather conditions eath , Cargo status information T emp And traffic information lope Construct a linear regression model to predict the estimated time T to reach the next stage node eta :

[0061] T eta =β 0 +V hiaca β 1 +C traf β 2 +W eath β 3 +T emp β 4 +S lope β 5 +ε

[0062] Among them, β 1 , β 2, β 3 , β 4 , β 5 are the corresponding weight coefficients, β 0 is the intercept term, ε is the error term, and the more accurate estimated time T for the current vehicle to reach the next stage node is obtained. eta , so that the estimated time T eta The node location and current time are entered into the shipping information management system to facilitate management personnel to monitor and control the ship's status in a timely manner.

[0063] Furthermore, considering that the vehicles sailing on the ocean are susceptible to seasonal influences, such as being more prone to failures in cold and hot seasons, the navigation time optimization unit is used to combine seasonal factors to predict the failure risk time and the repair time of the failure in the subsequent navigation process, so as to use the failure risk time and the repair time as inputs and combine the above linear regression model to re-predict the estimated time T for the vehicle to reach the next stage node eta , the specific steps are as follows:

[0064] Obtain the historical fault data set of the vehicle, which contains the fault records of the vehicle in the past, represented as fault time series data. Therefore, y t Represents the failure time data (such as hours, days, etc.) at time point t (such as month, quarter, etc.);

[0065] For ocean shipping, the seasonal cycle is usually 12 months (i.e. annual seasonality), or 4 quarters, etc. Therefore, the appropriate seasonal cycle s (12 months or 4 quarters) is determined, which is usually the seasonal fluctuation cycle of the failure time;

[0066] Perform seasonal difference on the original fault time series data to remove seasonal fluctuations in the data and make the data stable. The expression of seasonal difference is:

[0067] y″ t =y t -y t-s

[0068] Among them, y″ t represents the data after seasonal difference, y t-s Represents the failure time data at time point ts;

[0069] Constructing a SARIMA model (time series model) based on the seasonal cycle, we have:

[0070] Φ s (B s )·(1-B s ) D·(1-B s ) d yt=Θ s (B s )·εt

[0071] Among them, Φ s (B s ) represents the seasonal autoregressive part, which is used to capture the seasonal autoregressive relationship, (1-B s ) D represents the seasonal difference part, which is used to eliminate seasonal fluctuations, (1-B s ) d represents the non-seasonal difference part, which is used to eliminate the long-term trend in the data, B represents the lag operator, and B s represents the seasonal lag (e.g., failure data from 12 months ago), Θ s (B s ) represents the seasonal moving average part, capturing the impact of seasonal errors, D represents the order of seasonal differences, d represents the order of non-seasonal differences, and ε t represents the noise term;

[0072] Predict vehicle failure risk time:

[0073]

[0074] in, represents the predicted failure risk time in the next h steps, φ 1 ,φ 2 ,···φ p They represent the parameters of the autoregressive part, θ 1 ,···θ q They represent the parameters of the moving average part, ε t +h-1,ε t +h-2,···ε t +hq respectively represent the prediction error;

[0075] Based on the predicted failure risk time Determine the fault risk time Is it between the current time and the estimated time T? eta If it is in the range of , it means that the vehicle is more likely to fail during navigation, which will affect the estimated time T to reach the destination. eta Therefore, in order to estimate the time T eta To further determine, here the vehicle is expected to be at the time T eta Determine the possible fault type to estimate the subsequent maintenance time, so as to get the estimated time T again. eta , the specific steps are as follows:

[0076] Get the vehicle's failure risk time The future weather conditions et And its load L at that time oad ;

[0077] Obtain the fault type and historical weather conditions W′ corresponding to the historical fault time in the database et And the historical load L′ oad , determine the weather state W et 、Load L oad The same historical weather conditions W′ et And the historical load L′ oad , mark the corresponding fault type as the first fault type and construct a fault set, thereby determining the fault type at adjacent times in the fault set, determining the time intervals between multiple fault types to construct an interval sequence group, and determining the fault risk time The risk time interval between the previous historical fault time is used to determine the corresponding time interval in the interval sequence group, and the historical fault type existing in the corresponding time interval is determined as the fault risk time. The fault type is determined by the weather state W et 、Load L oad And the gradual screening of time intervals to analyze the failure risk time The accurate prediction of the failure type of the download tool can be based on the determined failure risk time. The fault type is calculated, and the maintenance time R corresponding to the fault type is calculated. pat , then:

[0078] Obtain relevant characteristics of the marine environment, including the risk of failure Wave height, wind speed, weather conditions (including rainfall, haze, air pressure, etc.);

[0079] Get vehicle-related features, including:

[0080] Fault type and fault location. Different fault types have different corresponding severity levels, and different fault locations have different impacts on time.

[0081] The vehicle health status is obtained by assigning corresponding weight coefficients according to the total number of historical failures, the total time of historical failures, and the total number of historical failure repairs, and then taking the weighted sum;

[0082] Identify the repair parts used in the fault type history and determine whether the corresponding repair parts on the current vehicle are sufficient. If the repair parts are insufficient, calculate the repair parts based on the fault risk time. The fault coordinates corresponding to the vehicle (the coordinate information when the vehicle is in a faulty state) are obtained to obtain other associated vehicles that are closest to it and pass through the fault coordinates, and whether the required maintenance parts are available on other associated vehicles. If so, the other associated vehicles are recorded as target other associated vehicles (it should be noted that for vehicles sailing on the ocean, the information will be filled into the shipping information management system in advance. Therefore, the status information of the target other associated vehicles is also predicted based on the above linear regression model), and the response distance between it and the fault coordinates, as well as the deployment impact state of deploying the maintenance parts to the currently faulty vehicle are determined. The deployment impact state is based on the obstacles that may be encountered during the deployment process, including weather and ocean conditions. The degree of influence (influence value) of different weather and ocean conditions is obtained respectively, and the corresponding numerical value is assigned and determined by weighted summation with the response distance (which is also assigned the corresponding influence value and weight).

[0083] The repair time R is obtained by linear regression model based on wave height, wind speed, weather conditions, fault type, fault location, health status, response distance and deployment impact status. pat It should be noted that the linear regression model constructed here and the linear regression model constructed by using historical transportation speed, traffic conditions, weather conditions, road conditions, and cargo status information mentioned above have a common principle structure, so they will not be repeated here.

[0084] It is worth noting that in the above, by analyzing whether the corresponding maintenance parts on the current vehicle are sufficient, if the maintenance parts are insufficient, the actual operation expectation of the current vehicle in the subsequent actual fault situation is predicted, so as to analyze the definition of maintenance time according to the predicted actual operation expectation. For the above prediction of the maintenance time of the vehicle in the case of a fault, it is usually predicted based on historical conditions or the most convenient method at present (that is, the vehicle in the above fault risk time In this case, the system predicts that the faulty vehicle will be deployed to other related vehicles nearby).

[0085] Furthermore, the predicted failure risk time of the vehicle during driving in the future and maintenance time R pat The linear regression model used initially is introduced to further determine the estimated time T to reach the next stage node eta , the algorithm expression is:

[0086]

[0087] Among them, β 6 , β 7 are the corresponding weight coefficients respectively, so the estimated time T for the subsequent vehicle to reach the next stage node under navigation can be accurately predicted eta, so that more accurate information can be entered into the shipping information management system. In this way, the ports and stations corresponding to the next stage nodes can make relevant preparations such as ship cargo replacement in advance to ensure that the transportation efficiency of ship shipping is not affected.

[0088] In the above, the non-time parameters are converted into time parameters through weight coefficients to obtain the estimated time T eta ,include:

[0089] The historical transport speed V hiaca Traffic conditions C traf 、Weather conditions eath , Traffic Information lope And cargo status information T emp Normalization or standardization makes their value ranges relatively consistent, ensuring that unit differences of different features do not lead to imbalanced weights;

[0090] The weight coefficient β in the above 1 , β 2 , β 3 , β 4 , β 5 Represent each feature (historical transport speed V hiaca Traffic conditions C traf etc.) on the linear effect of time increment;

[0091] Historical transport speed V hiaca : The unit of the voyage distance is km, which indicates the total mileage of the ship in this voyage. tV indicates the time term of the impact of the historical transport speed on the estimated time. V hiaca The unit is km / h, β1 represents the weight coefficient learned by the model, which is used to quantify the impact of unit speed change on the estimated time, and the unit is hour / (km / h);

[0092] Traffic Conditions traf :tC=β 2 ·C traf , β 2 It represents the weight coefficient learned by the model, which is used to quantify the impact of unit traffic congestion change on the estimated time. The unit is hour / traffic congestion unit. tC represents the time term of the impact of traffic conditions on the estimated time.

[0093] Weather conditions eath :tW=β 3 ·W eath , β 3 Represents the weight coefficient learned by the model, which is used to quantify the impact of weather conditions (rainfall, snowfall, etc.) on the estimated time. The unit is hour / weather unit. tW represents the time term of the impact of weather conditions on the estimated time.

[0094] Traffic Information lope :tS=β 5 ·s lope , β 5 Represents the weight coefficient learned by the model, which is used to quantify the linear impact of unit road condition changes (such as channel curvature, etc.) on the estimated time. The unit is hour / feature unit. tS represents the time term of the impact of road condition information on the estimated time.

[0095] Cargo status information emp :tT=β 4 ·T emp , β 4 It represents the weight coefficient learned by the model, which is used to quantify the linear impact of the change of the unit cargo status (such as temperature, vibration, etc.) on the estimated time. The unit is hour / cargo characteristic unit. tT represents the time term of the impact of cargo status information on the estimated time.

[0096] Failure risk time and maintenance time R pat It is a direct time feature and no conversion is required.

[0097] Furthermore, in the present solution, based on the status information in each stage node determined by the system (including the location of the vehicle, the current time, and the time to arrive at the next stage node), the information can be pushed to the user end, and can be pushed via email, text messages, etc. The method of pushing information is not limited here, so that users can be informed of the real-time information of the current shipping cargo, so that subsequent users can plan for subsequent matters based on the real-time information.

[0098] Finally, the present invention also provides:

[0099] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements any one of the above-mentioned systems when executing the computer program.

[0100] A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any of the above-mentioned systems, and the above-mentioned systems may be referred to and will not be described in detail here.

[0101] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the embodiments of the present invention.

Claims

1. A shipborne shipping information management system, including a shipping information management module, which is used to obtain the status information of the shipborne shipping, clarify the node position of the current vehicle, the current time and the estimated time to reach the next stage node, and fill the status information into the shipping information management system, characterized in that: Also includes: The voyage time prediction module is used to build a linear regression model based on the historical transportation speed, traffic conditions, weather conditions, road conditions, and cargo status information of the vehicle under the node shipping path to predict the estimated time for the vehicle to arrive at the next stage node, where: By introducing the influence of seasonality and combining it with the time series model, the failure risk time of the vehicle during the subsequent voyage is predicted, and the maintenance time is determined based on the historical status of the failure risk time. The determined failure risk time and maintenance time are used as inputs of the linear regression model to regain the estimated time and fill it into the shipping information management system.

2. The shipborne shipping information management system according to claim 1, characterized in that: The weather condition is obtained by weighted summation of weather temperature, weather humidity, weather wind speed, rainfall and snowfall; The cargo status information is obtained by weighted summation of the current cargo temperature, cargo humidity, cargo type and cargo shape; The road condition information is obtained by weighted summation of the channel width, channel depth, channel complexity and wave height of the node shipping path.

3. The shipborne shipping information management system according to claim 1, characterized in that: The algorithm expression of the linear regression model constructed by the historical transportation speed, traffic conditions, weather conditions, road condition information, and cargo status information is: T eta =β0+V hiaca β1+C traf β2+W eath β3+T emp β4+S lope b5+e Among them, T eta is the estimated time to reach the next stage node, V hiaca is the historical transportation speed, C traf For traffic conditions, W eath is the weather condition, T emp is the cargo status information, S lope is the road condition information, β1, β2, β3, β4, β5 are the corresponding weight coefficients, β0 is the intercept term, and ε is the error term.

4. The shipborne shipping information management system according to claim 3, characterized in that: The method for determining the failure risk time is: Obtain the historical fault data set of the vehicle, which contains the fault records of the vehicle in the past, represented as fault time series data; Perform seasonal difference on the original fault time series data to remove seasonal fluctuations in the data and make the data stable. The expression of seasonal difference is: and" t =and t -and t-s Among them, y″ t represents the data after seasonal difference, y t-s Represents the failure time data at time point ts; Constructing a time series model based on seasonal cycles, we have: F s (B s )·(1-B s ) D ·(1-B s ) d ·y t =Θ s (B s )·e t Among them, Φ s (B s ) represents the seasonal autoregressive part, (1-B s ) D represents the seasonal difference part, (1-B s ) d represents the non-seasonal difference part, B represents the lag operator, and B s represents the seasonal lag, Θ s (B s ) represents the seasonal moving average part, D represents the order of seasonal difference, d represents the order of non-seasonal difference, ε t represents the noise term; Predict vehicle failure risk time: in, represents the predicted failure risk time in the next h steps, φ 1, φ2,···φ p denote the parameters of the autoregressive part, θ1, ···θ q They represent the parameters of the moving average part, ε t +h-1,ε t +h-2,···ε t +hq respectively represent the prediction errors.

5. The shipborne shipping information management system according to claim 4, characterized in that: After the fault risk time is determined, the fault type of the vehicle is determined. The specific method is as follows: Get the vehicle's failure risk time The future weather conditions et And the load L at that time oad ; Obtain the fault type and historical weather conditions W′ corresponding to the historical fault time in the database et And the historical load L′ oad , determine the weather state W et 、Load L oad The same historical weather conditions W′ et And the historical load L′ oad , marking the corresponding fault type as the first fault type and constructing a fault set, thereby determining the fault types at adjacent times in the fault set; Determine the time intervals between multiple fault types to build interval sequence groups and determine the fault risk time The risk time interval between the previous historical fault time is used to determine the corresponding time interval in the interval sequence group, and the historical fault type existing in the corresponding time interval is determined as the fault risk time. The fault type.

6. The shipborne shipping information management system according to claim 5, characterized in that: The method for determining the maintenance time is as follows: Obtain relevant characteristics of the marine environment, including the risk of failure Wave height, wind speed, and weather conditions; Get vehicle-related features, including: Fault type and fault location; and, the health status of the vehicle; Identify the repair parts used in the fault type history and determine whether the corresponding repair parts on the current vehicle are sufficient. If not, calculate the repair parts based on the fault risk time. The fault coordinates corresponding to the vehicle are used to obtain other associated vehicles that are closest to the vehicle and pass through the fault coordinates, and whether the required maintenance components are available on the other associated vehicles. If available, the other associated vehicles are recorded as target other associated vehicles, and the response distance between the other associated vehicles and the fault coordinates is determined, as well as the deployment impact status of deploying the maintenance components to the vehicle with the current fault. The deployment impact status is determined by weighted summation of weather, ocean conditions and response distance. The wave height, wind speed, weather conditions, fault type, fault location, health status, response distance and deployment impact status are used to build a linear regression model to obtain the maintenance time R pat .

7. The shipborne shipping information management system according to claim 6, characterized in that: In the process of using the failure risk time and the maintenance time as inputs of the linear regression model to regain the estimated time, the algorithm expression of the linear regression model is: Among them, β6 and β7 are the corresponding weight coefficients respectively.

8. The shipborne shipping information management system according to claim 6, characterized in that: The impact of different weather and ocean conditions is obtained respectively, and the corresponding numerical values ​​are assigned and weighted summed with the response distance to determine the deployment impact status.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the system according to any one of claims 1 to 8 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the system according to any one of claims 1 to 8 is implemented.

Images