JP2017102565A - Anchorage vessel number prediction system, anchorage vessel number prediction method, and anchorage vessel number prediction program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make it possible to predict the number of vessels resting in a port.SOLUTION: With an anchorage vessel number prediction system 1 and an anchorage vessel number prediction method by the same, an overall time map creation unit 11 creates an overall time map (overall time distribution information) representing a distribution of an overall time from a current position of a vessel navigating to a specific port to the specific port and an anchorage time distribution information creation unit 13 creates anchorage time distribution information representing an anchorage time distribution of a vessel in a specific port. It becomes possible to predict the number of vessels resting in a port hereinafter by combining these information relative to AIS data capable of identifying a current position of a vessel.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an anchored vessel number prediction system, an anchored vessel number prediction method, and an anchored vessel number prediction program for predicting the number of vessels anchored in a harbor.

  Ship identification code, type, position, course, speed, navigation status and other safety-related information are automatically transmitted and received by VHF charged waves, and information between ships and between the ship and land navigation aids, etc. As an exchange system, an automatic identification system (AIS) is known.

  According to the AIS data transmitted from the ship, it is possible to grasp the situation relating to the ship being navigated. In recent years, a method for estimating a future movement route of a ship as described in, for example, Patent Document 1 using information in a database in which AIS data is accumulated has been studied.

JP-A-2015-59896

  Since ships are generally intended to carry people or luggage back and forth on the water between the ports, congestion in the ports affects the navigation of the vessels. Therefore, there is a demand for the degree of congestion in the port, that is, the prediction of the number of ships anchored in the port.

  The present invention has been made in view of the above, and an object thereof is to provide a berthing number prediction system, a berthing number prediction method, and a berthing number prediction program capable of predicting the number of vessels anchored in a port. .

  In order to achieve the above object, a berthing number prediction system according to an aspect of the present invention includes an AIS data acquisition unit that acquires AIS data related to navigation of a ship, and a past predetermined period acquired by the AIS data acquisition unit. A required time distribution information generating means for generating required time distribution information indicating a distribution of required time from a staying position of a ship sailing to a specific port to a specific port based on the AIS data; and the AIS data acquiring means. Based on the acquired AIS data for a predetermined period in the past, the berthing time distribution information creating means for creating the berthing time distribution information indicating the berthing time distribution of the ship in the specific port, and the AIS data acquiring means Based on the AIS data, the required time distribution information, and the berthing time distribution information, the number of berths in the specific port is predicted. And anchored Number of vessels prediction means, and having a.

  In addition, the method for predicting the number of anchored vessels according to one aspect of the present invention is based on the AIS data acquisition step of acquiring AIS data related to the navigation of the ship, and the AIS data of the past predetermined period acquired in the AIS data acquisition step. A required time distribution information creation step for creating a required time distribution information indicating a distribution of a required time from a staying position of a ship sailing to a specific port to a specific port, and the past acquired in the AIS data acquisition step Based on the AIS data for a predetermined period, the berthing time distribution information creating step for creating berthing time distribution information indicating the berthing time distribution of the ship in the specific port, the AIS data obtained in the AIS data obtaining step, Based on the required time distribution information and the berthing time distribution information, the ship in the specific port It characterized by having a a berthing number of vessels prediction step of predicting the mooring Vessels.

  A berthing number prediction program according to an embodiment of the present invention is a berthing number prediction program for causing a computer to execute a berthing number prediction of a ship in a port, and an AIS data acquisition step for acquiring AIS data relating to the navigation of the ship; Based on the AIS data acquired in the past predetermined period acquired in the AIS data acquisition step, the required time distribution information indicating the distribution of the required time to the specific port from the staying position of the ship navigating to the specific port is created. The berthing time distribution information which shows the berthing time distribution information which shows the berthing time distribution of the ship in the said specific port based on the AIS data of the past predetermined period acquired in the said required time distribution information creation step and the said AIS data acquisition step Distribution information creation step and AIS acquired in the AIS data acquisition step And a berthing number predicting step of predicting the number of anchored vessels in the specific port based on the data, the required time distribution information, and the anchoring time distribution information, To do.

  According to the above-mentioned anchored vessel number prediction system, anchored vessel number prediction method, and anchored vessel number prediction program, the required time distribution information indicating the distribution of the required time from the staying position of the ship sailing to the specific port to the specific port, and After preparing the berthing time distribution information indicating the berthing time distribution of a ship in a specific port, it is possible to predict the number of ships berthed at a port in the future by combining these information with the AIS data. It becomes possible.

  Here, the number of anchored vessels predicting means is based on the AIS data, the required time distribution information, and the anchoring time distribution information, to the specific port relating to each vessel that is navigating to the specific port. And the departure time from the specific port and the departure time from the specific port related to each ship that has already arrived at the specific port, and the number of ships anchored at the specific port It can be set as the aspect which estimates.

  As mentioned above, when predicting the number of anchored vessels, it is possible to predict the number of anchored vessels with higher accuracy by calculating the arrival time and departure time from a specific port for each ship. .

  The required time distribution information may be an aspect in which a statistic relating to a required time from the stay position of the ship to the specific port is calculated for each mesh obtained by dividing the surface of the water into predetermined areas. it can.

  As mentioned above, the required time distribution information is the information calculated for each mesh on the required time from the stay position of the ship to the specific port for each mesh. The required time can be easily obtained. In addition, by adopting a configuration that calculates a statistic related to the required time for each mesh, it is possible to prevent an excessive increase in the amount of work related to creation of the required time distribution information.

  When the required time distribution information creating means calculates the required time to the specific port from the stay position of the ship navigating to the specific port from the AIS data of the past predetermined period, the ship is on the water. It is determined whether there is a stoppage time, and if there is a stoppage time on the water, the stoppage time is removed and the required time to the specific port is calculated. It can be set as an aspect.

  As described above, if there is a time when the ship is stopped on the water, the time required to reach a specific port is calculated by removing the time when the ship is stopped. The required time can be calculated without including the time during which the ship is stopped on the water, and the required time distribution information can be created with higher accuracy.

  The berthing time distribution information creating means identifies arrival time and departure time at the specific port in each ship anchored at the specific port from the AIS data of the past predetermined period, and the arrival time and the departure time. The berthing time distribution information of the ship in the specific port can be created by calculating the berthing time in the specific port of each ship based on the above.

  As described above, by specifying the arrival time and departure time at a specific port from the past AIS data related to the ship and calculating the berthing time from these information, for example, the crew of the ship, etc. The anchoring time distribution information can be created with higher accuracy than when the anchoring time distribution information is created based on the input information.

  According to the present invention, there are provided a berthing number prediction system, a berthing number prediction method, and a berthing number prediction program capable of predicting the number of vessels anchored in a port.

It is a functional block diagram of the number of ships anchored system. It is a figure which shows the structure of a docked ship number prediction program. It is a flowchart explaining the number of anchored ship prediction method. It is a flowchart explaining the preparation method of a required time map. It is a figure explaining the creation example of a required time map. It is a flowchart explaining the preparation method of berthing time distribution information. It is a figure explaining the creation example of berthing time distribution information. It is a flowchart explaining the method of estimating the number of anchored vessels from AIS data. It is a figure explaining the example of prediction of the number of anchored vessels.

  DESCRIPTION OF EMBODIMENTS Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to the accompanying drawings. In the description of the drawings, the same elements are denoted by the same reference numerals, and redundant description is omitted.

  FIG. 1 is a functional block diagram of a berthing number prediction system according to an embodiment of the present invention. The anchored vessel number prediction system 1 in FIG. 1 predicts the number of vessels anchored at a port using AIS data transmitted from an automatic identification system (AIS) mounted on a vessel navigating on the water. System. The “port” in the present embodiment refers to a port provided with a plurality of facilities (berths) for a ship to berth for the purpose of loading and unloading cargo. In such a port, many ships repeat berthing, berthing and leaving. There are also many ships that have the same port as their destination. However, since a larger number of ships than the number of berths provided at the port cannot be anchored at the port, when the port is congested, the time is adjusted by changing the route or anchoring before the port. There are also ships. The anchored vessel number prediction system 1 according to the present embodiment is a system that predicts how the number of vessels anchored at a specific port will change in the future using AIS data.

  As shown in FIG. 1, the anchored vessel number prediction system 1 includes an AIS data acquisition unit 10 (AIS data acquisition unit), a required time map generation unit 11 (required time distribution information generation unit), a required time map storage unit 12, Berthing time distribution information creation unit 13 (berthing time distribution information creation means), berthing time distribution information storage unit 14, berthing number prediction instruction receiving unit 20, berthing number prediction unit 21 (berthing number prediction unit), berthing And a ship number prediction result output unit 22.

  The AIS data acquisition unit 10 has a function of acquiring AIS data from an external device. The AIS data stored in the external device includes, for each data, data transmission date and time (data acquisition date and time), information for identifying the ship (identification code: ship ID), ship type (ship type), ship position Information indicating (longitude / latitude), the course of the ship (the destination port), the speed of the ship (navigation speed), the navigation state of the ship, and the like are associated with each other. AIS data is information transmitted at predetermined intervals from a ship equipped with a ship automatic identification device (AIS). AIS data transmitted from a ship is stored in an external device after being made into a database. Therefore, when the anchored vessel number prediction system 1 uses AIS data for the above-mentioned use, AIS data is acquired from an external device.

  In the berthing number prediction system 1, the AIS data is used for two purposes. The first application is the use as basic data for creating the required time map and berthing time distribution information used when predicting the number of anchored vessels, and uses AIS data for a specific period in the past. The second application is used as basic data when the number of anchored vessels is actually predicted. How to use the AIS data will be specifically described later. Acquisition of AIS data by the AIS data acquisition unit 10 is performed periodically for AIS data in a specific period in the past or when an instruction from the operator of the anchored boat number prediction system 1 is received. In addition, acquisition of AIS data as basic data for actually predicting the number of anchored vessels is performed at the time of an instruction from the anchored vessel number predicting unit 21 described later.

  The required time map creation unit 11 has a function of creating a required time map using the AIS data acquired by the AIS data acquisition unit 10. The required time map is a form of required time distribution information, and shows information indicating the distribution of required time to a specific port according to the position where the ship stays in correspondence with the map. The required time map is created for each port, and is used to predict the required time to a specific port for each ship when predicting the number of anchored vessels. A method for creating the required time map will be described later. In addition, it can also be set as the structure which prepares the function etc. which show distribution of the required time according to the position of a ship instead of a required time map as required time distribution information.

  The required time map storage unit 12 has a function of storing the required time map created by the required time map creating unit 11.

  The berthing time distribution information creation unit 13 has a function of creating berthing time distribution information using the AIS data acquired by the AIS data acquisition unit 10. The berthing time distribution information is information indicating the berthing time distribution of a ship in a specific port. The berthing time distribution information is used to predict the time at which a ship arriving at a specific port departs from the port. The berthing time distribution information is created for each port. The method for creating the berthing time distribution information will be described later.

  The berthing time distribution information storage unit 14 has a function of storing the berthing time distribution information created by the berthing time distribution information creation unit 13.

  When creating the required time map and berthing time distribution information, AIS data for a predetermined period in the past (for example, 10 days to several months) is used.

  The anchored vessel number prediction instruction receiving unit 20 has a function of receiving an instruction to start processing related to the prediction of the anchored vessel number (an anchored vessel number prediction instruction). An example of a person who gives an instruction to predict the number of anchored vessels is an operator of the anchoring boat number prediction system 1. It is also conceivable that the anchored vessel number prediction system 1 receives an instruction from an external device (for example, a device that manages a specific port). When the anchor number prediction instruction is received, the anchor number prediction unit 21 starts an instruction relating to the prediction of the anchor number. The instruction to predict the number of anchored vessels includes information for specifying a port to be predicted and information for specifying a prediction period.

  When the anchored vessel number prediction instruction receiving unit 20 receives the anchored vessel number prediction instruction, the anchored vessel number prediction unit 21 performs processing related to the prediction of the anchored vessel number. Specifically, the berthing number prediction unit 21 acquires AIS data related to a port to be predicted by giving an instruction to the AIS data acquisition unit 10. Thereafter, the berthing number predicting unit 21 includes the AIS data, a required time map related to a port to be predicted out of the required time map stored in the required time map storage unit 12, and an anchored time distribution information storage unit 14. The number of anchored vessels is predicted based on the anchoring time distribution information related to the port to be predicted, among the anchoring time distribution information stored in. Details will be described later.

  The anchored vessel number prediction result output unit 22 has a function of outputting the result predicted by the anchored vessel number prediction unit 21. Examples of a method for outputting the anchored vessel number prediction result include a method of outputting to the monitor of the anchored vessel number prediction system 1 and a method of transmitting the anchored vessel number prediction result to the sender of the anchored vessel number prediction instruction.

  The anchored vessel number prediction system 1 includes a central processing unit (CPU), a random access memory (RAM) and a read only memory (ROM) that are main storage devices, a communication module for communication, and an auxiliary storage device such as a hard disk. It is comprised as a computer provided with hardware, such as. And the function of each part is exhibited when a constituent element operate | moves when the docking number prediction program memorize | stored in memory | storage devices, such as RAM or ROM, is performed by CPU.

  A berthing number prediction program P for causing a computer to function as the berthing number prediction system 1 will be described with reference to FIG.

  The berthing number prediction program P includes a main module P01, an AIS data acquisition module P10, a required time map creation module P11, a berthing time distribution information creation module P13, a berth number prediction instruction reception module P20, a berth number prediction module P21, and a berth number A prediction result output module P22 is provided. The main module P01 is a part that comprehensively controls processing related to the prediction of the number of anchored vessels. This is realized by executing the AIS data acquisition module P10, the required time map creation module P11, the berthing time distribution information creation module P13, the berthing number prediction instruction reception module P20, the berthing number prediction module P21, and the berthing number prediction result output module P22. The functions to be performed are the AIS data acquisition unit 10, the required time map creation unit 11, the berthing time distribution information creation unit 13, the berthing number prediction instruction reception unit 20, the berth number prediction unit 21, and the berth number prediction in the present embodiment, respectively. The function of the result output unit 22 is the same.

  The berthing number prediction program P may be provided after being fixedly recorded on a tangible recording medium such as a CD-ROM, DVD-ROM, or semiconductor memory. Alternatively, the berthing number prediction program P may be provided via a communication network as a data signal superimposed on a carrier wave.

  Next, a method for predicting the number of anchored vessels in the anchored vessel number prediction system 1 will be described with reference to FIG. FIG. 3 is a flowchart for explaining a series of flows relating to the prediction of the number of anchored vessels.

  The prediction of the number of anchored vessels in the anchored vessel number prediction system 1 is roughly divided into two steps. That is, a step of preparing data used for predicting the number of anchored vessels and a step of actually estimating the number of anchored vessels. In the flow chart shown in FIG. 3, the required time map creation from the past AIS data (S01: AIS data acquisition step, required time distribution information creation step) and the berthing time distribution information creation from the past AIS data (S02: AIS data acquisition step and berthing time distribution information creation step) are steps for preparing data to be used for predicting the number of berths. The process from the reception of the anchored vessel number prediction instruction (S03) to the output of the prediction result (S06) is the process of actually estimating the number of anchored vessels.

  In the following embodiment, a process of preparing data used for predicting the number of anchored vessels related to the port Z and a step of actually estimating the number of anchored vessels related to the port Z will be described. As a premise, the harbor Z is assumed to have six berths A to F.

  First, the process of preparing data used for predicting the number of anchored vessels will be described. The anchorage number prediction system 1 creates a required time map in the required time map generating unit 11 based on the past AIS data acquired by the AIS data acquiring unit 10 and stores the required time map in the required time map. Stored in the unit 12 (S01). The berthing time distribution information creating unit 13 creates berthing time distribution information based on the past AIS data acquired by the AIS data obtaining unit 10, and the berthing time distribution information is stored in the berthing time distribution information storage unit. 14 (S02).

  The required time map and the berthing time distribution information are both created from AIS data for a predetermined period in the past. Further, the required time map and the berthing time distribution information are updated regularly (for example, about one week to several months). That is, the required time map and the berthing time distribution information are periodically created and updated in the berthing number prediction system 1 in order to improve the prediction accuracy of the number of berths. In addition, the structure produced every time based on the instruction | indication from the operator etc. of the number-of-berthing vessels prediction system 1 may be sufficient.

  A method for creating a required time map using AIS data will be described with reference to FIGS. FIG. 4 is a flowchart illustrating a method for creating a required time map. FIG. 5 is a diagram schematically showing a specific procedure for creating a required time map.

  First, as shown in FIG. 4, the required time map creation unit 11 extracts route data relating to a ship heading to a specific port (here, port Z) from the past AIS data acquired by the AIS data acquisition unit 10. (S11).

  In general, since AIS transmission from a ship is repeatedly performed, the past AIS data acquired by the AIS data acquisition unit 10 includes a plurality of pieces of information transmitted from the same ship. Therefore, in the acquired AIS data, after the data is collected for each ship based on the information (ship ID) for specifying the ship, the AIS data from the same ship is rearranged in time series. Thereby, the log | history which concerns on a movement of a ship can be confirmed. Furthermore, the ship repeats navigation to a berth provided in a certain port and anchoring at that berth (loading and unloading luggage). Therefore, when AIS data indicating that the berth is anchored can be identified, the ship is navigating to the port (a specific berth in the port) by going back in time from the AIS data. AIS data can be extracted.

  In order to determine whether or not a specific berth in the port is anchored from the AIS data, the following method can be used. That is, a specific berth in a harbor (a region including a region where a ship is actually anchored and a region where an error derived from the position of positioning equipment or positioning accuracy can be estimated to be a berth) Is stored in advance. Then, based on the AIS data of the same ship, when it can be confirmed that the ship stays at a specific berth in the port for a predetermined period or longer, the ship There is a way to determine that you are anchored in Bath.

  In addition, even if information indicating the position of a berth in a harbor is not prepared in advance, information indicating the position of a coastline (a region that can be estimated to be a coastline at a level where errors derived from the position of positioning equipment or positioning accuracy can be tolerated If it is held in advance, it can be determined whether or not the ship is anchored near the coastline based on the AIS data of the same ship. Therefore, based on the AIS data of the same ship, when it can be confirmed from the AIS data that the ship is staying along the coastline for a predetermined period of time, the place where the ship is anchored is the port. It is possible to use a method of determining that the ship is anchored at the berth.

  Moreover, when the information which shows the position of the berth in a harbor is not prepared beforehand and the other information is not hold | maintained, the method of judging individually with reference to a map etc. can be used. For example, if the AIS data includes information indicating that the state in which the ship does not move has not continued for a predetermined period or longer, whether or not the position where the ship stays is a position corresponding to the coastline. It is possible to determine whether or not the ship is anchored at the berth in the harbor by the method of confirming with comparison with the map.

  Based on the above criteria, a series of history (AIS data) related to movement to a berth in a certain port is used as one route data, and a new identification code (route ID) is added to the AIS data included in the route data. Is allocated. Thereby, it is possible to obtain route data in which a route of the same ship moving toward a specific berth in the harbor is collected in time series from a large amount of AIS data.

An image of the route data is shown in FIG. One route data includes a plurality of pieces of position information related to one ship heading for a specific berth provided in the destination port Z. Since the position of the ship is indicated by longitude / latitude, it can be considered that the position of the ship is coordinate information substantially indicated by two coordinate axes. FIG. 5A plots and visualizes the positions indicated by the position information P ₀ , P _{1 to} P _N in each AIS data. In FIG. 5A, the position information _PN is information indicating the berth A in the port Z. Further, the position information P ₀ , P _{1 to} P _N is associated with the date and time when the position information is acquired. In FIG. 5 (A), t ₀ , t _1, ..., T _N indicate the acquisition date and time of each AIS data. As a result of connecting each AIS data with a straight line in order of date and time, the route of the ship is known as shown in FIG. Here, for example, the difference between the date and time t ₀ and the date and time t ₁ becomes s ₁ , and can be said to be the time required for navigation between two points. When there are N + 1 pieces of AIS data included in the route data, N pieces of information specifying the time required for moving between two points are created. Data including all of these becomes one route data.

  The required time map creation unit 11 extracts the route data relating to the ship heading for the berth A provided in the specific port (port Z) from the route data as described above (S11). As a reference for extracting route data, it is possible to use whether or not the destination (final destination) of the route data is a port for which a required time map is to be created. Route data is extracted based on whether or not the berth A in the port Z is a destination (final destination).

  The extracted route image is shown in FIG. In FIG. 5B, a plurality of route data relating to a ship heading to the berth A of the port Z is shown. Thus, when the route data which goes to the berth in the specific harbor Z is extracted, the data which show the route to the berth A which concerns on a some ship are obtained. Thereafter, the same processing is performed on the berths B to F, so that it is possible to extract the route data relating to the ship heading to the berth in units of berths. Thereby, all the route data which show the route to the port Z concerning a some ship can be acquired.

  Next, the required time map creation unit 11 calculates a required time actual value from the position to the port for each coordinate information from the extracted AIS data (S12). The route data includes a plurality of AIS data. The required time performance value is calculated for each AIS data. Moreover, it calculates as a required time actual value to a port irrespective of the berth where a ship arrives.

  Since the position of the ship is indicated by longitude / latitude, it can be considered that the position of the ship is substantially coordinate information indicated by two coordinate axes. Considering simply, since the arrival date and time at a specific berth in the destination port can be known based on the route data, the AIS data included in one route data is the arrival date and time of the port (berth) in the route data. And the difference between the acquisition date and time of each AIS data is the required time until the arrival of the port (berth) in each AIS data. However, in reality, there is a case where a ship moving to a specific port Z performs a so-called “offshore waiting” in which the ship stops at a predetermined time (water) different from the port. “Oki wait” is often performed to adjust the arrival date and time at the port. Therefore, if the required time to the port is calculated in a state where “offshore waiting” is included, a value larger than the required time required for actual navigation is calculated. Accordingly, when calculating the required time actual value, it is determined whether the ship is “moving or stopped”, and the time during which the ship is stopped (offshore waiting time) is removed.

The calculation of the required time performance value considering whether the ship is “moving or stopped” can be determined by the following method. As one method, there are two points for information s _{1 to} s _N indicating the movement time (required time) between two points of the previous data in the AIS data arranged in time series. A flag indicating movement or stoppage is given based on the distance between the two. In addition, when a flag indicating that the vehicle is moving or stopped is given, information indicating the navigation speed included in each AIS data, information indicating the navigation state of the ship, and the like may be used. Then, in response to the position information P _0, P ₁ ~P _N, when calculating the time required to Bath A harbor Z is a destination (corresponding to the position information P _N) are ports from each location Of the time required to connect the points up to the berth A of Z, only the information to which the moving flag is attached is totaled and calculated. For example, the actual time required value from the point specified by the position information P ₀ to the berth A of the port Z is calculated as the sum of s ₁ to s _N when there is no time to which the stop flag is given. Can do. However, when there is information indicating the travel time between two points to which the stop flag is assigned, the information indicating the travel time to which the stop flag is assigned is removed from the total of s ₁ to s _N described above. Thereby, the required time performance value from the point specified by the position information P ₀ to the berth A of the port Z can be appropriately calculated. The actual method for calculating the actual time required value is not limited to the above, but if it is determined whether or not the ship is stopped, the time zone part is removed if it is included. Then, by calculating the required time actual value up to the berth A of the port Z, a more accurate required time actual value can be obtained.

  Thereafter, for the required time actual value for each AIS data calculated in the above process, a statistic relating to the required time is calculated for each mesh obtained by dividing the coordinates for each predetermined section on the water (S13). The mesh refers to a region having latitude and longitude range information as shown in FIG. Although the required time actual value was obtained for each AIS data in the previous stage, this required time actual value is information indicating the required time from the specific position specified by the position information, and is not plane information. . Therefore, the distribution of the required time actual value is obtained in a region (mesh) unit obtained by dividing the surface of the water into predetermined areas. In addition, as a statistic related to the required time calculated for each mesh, a numerical value that can indicate the distribution of the required time actual value such as an average value and a standard deviation can be used. Moreover, it is good also as a structure which calculates the statistic corresponding to a required time actual value by calculating the distribution of the required time actual value contained in one mesh using a histogram etc. FIG. As described above, the statistics can be appropriately selected using a known method. In the present embodiment, a case will be described in which the average required time actual value is calculated for each mesh as a statistic related to the required time.

  Specifically, meshes for displaying the distribution of required time are set in advance, and a mesh number is assigned to each mesh. Thereafter, for each AIS data for which the required time performance value has been calculated, the mesh to which the AIS data belongs is specified based on the position information, and a mesh number corresponding to the mesh is assigned. Then, based on the AIS data to which the same mesh number is assigned, the average of the calculated required time values is calculated. By performing this for each mesh, it is possible to obtain a distribution of required time results for each mesh.

  FIG. 5C shows an average distribution of actual time required values for each mesh. In the example shown in FIG. 5C, each mesh is a square having the same area, and the required time (the required number of days) is different for each mesh. In addition, it can replace with an average value and the distribution of a required time actual value can also be created using the representative value of the required time actual value for every mesh.

  The size of the mesh can be appropriately changed and set so that one side is several hundred m to several tens km. The size of the mesh may be changed depending on the place in the required time map. For example, a mesh can be set small around a main route where the frequency of vessel navigation is large, and the distribution of required time actual values can be more accurately grasped according to the location of the vessel. Moreover, it can also be set as the structure which increases the number of the data used for the average value calculation of the required time performance value in the mesh in the area where the navigation frequency of the ship is small. Further, the shape of the mesh is not limited to a square, and can be appropriately changed in consideration of topography and the like.

  The distribution of the required time actual value calculated by the above processing becomes the required time map of the ship heading to the port Z. The generated required time map is stored in the required time map storage unit 12. (S13).

  Thus, the required time map is created for each port and stored in the required time map storage unit 12.

  Next, a method for creating berthing time distribution information using AIS data will be described with reference to FIGS. FIG. 6 is a flowchart illustrating a method for creating the berthing time distribution information. FIG. 7 is a diagram schematically showing a specific procedure for creating the berthing time distribution information. Here, an example in which the berthing time actual value and the berthing time distribution are created in units of berths and then the berthing time distribution information as a port is created will be described, but the present invention is not limited to this.

  First, in the berthing time distribution information creation unit 13, AIS data relating to a ship including route data regarding a ship heading to a specific port (here, port Z) from past AIS data acquired by the AIS data acquisition unit 10. Is extracted (S21). The route data is a series of data (AIS data) related to movement to a berth included in a certain port as described above, and is collected as a series of data groups. The berthing time distribution information is information indicating the distribution of how much ships arriving at a berth in a specific harbor anchored at that berth. For example, the berth of a ship having route data with the berth in the harbor as the destination Need to get staying time at. Accordingly, the AIS data of the ship including the route data with the berths A to F included in the port Z as the destination is acquired here.

  Next, the arrival time and departure time at the berth included in the specific port are specified from the extracted AIS data of the ship, and a difference berthing time actual value is calculated (S22).

A method for calculating the actual berthing time will be described with reference to FIG. In FIG. 7A, the AIS data of the ship including the route data to the berth A in the port Z is plotted in time series corresponding to the time, and in FIG. 7A, five AIS data D are plotted. It shows the ₁ ~D _5.

Since position information is associated with each of the AIS data D _{1 to} D ₅ , data including the position information of the berth A is extracted from the AIS data of the ship. In FIG. 7A, it is assumed that information indicating that the ship is anchored at berth A is included in the two AIS data D ₃ and D ₄ . At this time, an arrival time at the acquisition date and time Bath A of AIS data D _3, becomes the starting time of the acquisition date and time Bath A of AIS data D _4, the difference (here, 10 hours) of Bath A It becomes stay time. If there are three or more consecutive AIS data indicating staying at berth A, the first and last data of the AIS data indicating staying at berth A, Between is the berthing time. In this way, by arranging the AIS data related to the ship having the route data toward the berth A of the port Z in time series, the actual berthing time value at the berth A of the port Z can be calculated. Moreover, the acquisition period of the past AIS data is long, and one ship may visit Berth A many times during that period. In this case, processing such as determining whether or not the data is continuous based on whether or not there is route data before and after is specified, and the actual berthing time value at the berth A in the port Z is specified. .

  When determining whether or not a ship is anchored at a berth in port Z, the position information of the ship included in the AIS data includes the area where the ship is actually anchored and the position of the positioning device or The determination is made based on whether or not the area that can be estimated to be staying in the same berth is continuously staying in the area including the area that can be estimated to be staying at the same berth at a level that allows an error derived from the positioning accuracy.

  The actual berthing time value at the berth A in the port Z is calculated for each ship and for each berth at the berth A in the port Z. Next, distribution information based on the result is created (S23). FIG. 7B is an example of creation of distribution information of actual berthing time values. FIG. 7B shows the distribution of the berthing time at the berth A in the port Z as a histogram, showing that 9 hours is the mode value. Thus, the distribution information of the actual berthing time value can be created as a histogram, for example. Further, a probability density function related to the actual berthing time value can also be calculated as distribution information of the actual berthing time value.

  The distribution of the berthing time actual value calculated by the above processing becomes the berthing time distribution information of the ship in the berth A in the harbor Z. By performing the series of processes described above for the other berths B to F in the port Z, it is possible to obtain the distribution of the berthing time actual value in units of berths. And the berthing time distribution information of the ship in the port Z can be obtained by calculating the sum total of the distribution of the berthing time actual value which concerns on berths A-F. The created berthing time distribution information is stored in the berthing time distribution information storage unit 14. (S24).

  In this way, the berthing time distribution information is created for each port and stored in the berthing time distribution information storage unit 14.

  Returning to FIG. 3, the process of actually predicting the number of anchored vessels will be described. When the number of anchored vessels is predicted, the anchored vessel number prediction instruction receiving unit 20 first receives an anchored vessel number prediction instruction (S03). The berthing number prediction instruction includes information for specifying a port (here, port Z) to be predicted and information for specifying a prediction period. When the anchored vessel number prediction instruction receiving unit 20 receives the anchored vessel number prediction instruction, the anchored vessel number predicting unit 21 performs processing related to the prediction of the anchored vessel number.

  The anchored vessel number prediction unit 21 first acquires AIS data used to predict the number of anchored vessels by instructing the AIS data acquisition unit 10 (S04: AIS data acquisition step). Here, the AIS data used by the anchored vessel number prediction unit 21 to predict the number of anchored vessels is the latest position information of each ship on the water and information indicating the destination. However, the acquisition date and time of AIS data from each ship is different from each other. Therefore, the AIS data acquisition unit 10 acquires AIS data for a past predetermined period (for example, several days to several weeks).

  Next, the number of anchored vessels predicting unit 21 predicts the number of anchored vessels in a specific port (here, port Z) based on the AIS data acquired by the AIS data acquiring unit 10 (S05: anchoring vessel number predicting step).

  A method for predicting the number of anchored vessels using AIS data will be described with reference to FIGS. FIG. 8 is a flowchart illustrating the prediction of the number of anchored vessels. Moreover, FIG. 9 is a figure which shows typically the specific procedure which concerns on the prediction of the number of anchored vessels.

  First, the anchored vessel number prediction unit 21 extracts AIS data relating to a ship related to a specific port (port Z) from the AIS data acquired by the AIS data acquisition unit 10 (S31). The AIS data acquired by the AIS data acquisition unit 10 includes information related to all the ships that have acquired the AIS data. Among them, ships that are sailing to the port Z and ships that are anchored at the port Z This is a ship that may be anchored in the port Z within the period of predicting the number of anchored ships. Therefore, by extracting AIS data related to the ship that has set the port Z as the course (destination port) of the ship in the AIS data, AIS data related to the ship that is sailing toward the port Z is extracted. be able to. In addition, after collecting data for each ship for each acquired AIS data, the latest position information of each ship can be acquired by rearranging the AIS data from the same ship in chronological order. At this time, it can be considered that the ship whose latest position information is the port Z is currently staying in the port Z. Based on this, AIS data of the ship related to the port Z can be extracted.

  Next, the arrival time prediction value and the departure time prediction value of the port Z are calculated for each ship from the AIS data of the ship related to the extracted port Z (S32).

  For ships that are navigating, the predicted arrival time at the port Z can be calculated from the position information and the required time map included in the latest AIS data for the ship. Specifically, the required time to the port Z is calculated from the required time map by determining which mesh in the required time map related to the port Z the position information of the ship indicated by the AIS data belongs. Can do. Therefore, the time when the time specified by the required time map to the port Z from the acquisition date and time of the latest AIS data relating to the ship has elapsed is the predicted arrival time.

  On the other hand, the berthing time distribution information is used to calculate the predicted departure time from the port Z for the ship. As described above, the berthing time distribution information is created as a histogram or a probability density function. That is, it can be said that the berthing time distribution information is information indicating the probability distribution of the berthing time (indicating the width). Therefore, when predicting the departure time, the representative value is used as a numerical value related to the berthing time distribution information. In addition, as a representative value of berthing time distribution, the average value of berthing time distribution can be used, for example, However, It is not limited to this.

  For ships currently moving toward port Z, after arriving at port Z, it is predicted that they will be anchored for a period specified by the representative value of the anchoring time distribution. Therefore, by adding a representative value of the berthing time distribution to the predicted arrival time of the ship at the port Z, the predicted departure time from the port Z is obtained.

  On the other hand, if the average value of the berthing time distribution is added to the ship that has already arrived at the port Z, the berthing time is estimated to be longer than the berthing time distribution indicated by the berthing time distribution information. Therefore, for a ship arriving at the port Z, not only the latest AIS data but also the previous AIS data is referenced to identify the time of arrival at the port Z and estimate the elapsed time from that time. Then, the estimated berthing time at the future port Z is estimated in consideration of the elapsed time, and the predicted departure time is calculated.

  Specifically, for example, there is a method of performing processing by assuming that every ship arrives at the port Z and that the ship is anchored at the port Z by a representative value (average value) of the anchoring time distribution uniformly. As another method, when there is a bias in the berthing time distribution in the berthing time distribution information, the information indicating the berthing time distribution included in the berthing time distribution information is after arrival at the port Z. It is conceivable to calculate the average berthing time only from information having a longer berthing time than the elapsed time, and to estimate the berthing time at the port Z based on this. In any case, the predicted departure time is calculated in consideration of the time of arrival at the port Z.

  Next, the above calculation results are totaled to create time-series data (S33). The calculation of the predicted arrival time and the predicted departure time of the port Z is performed for each ship. Therefore, based on the result, the number of ships anchored in the port Z is counted in time series.

In the upper part of FIG. 9, the estimated arrival time and the estimated departure time calculated for each ship are shown along the time. Also, the lower part of FIG. 9 shows the result of totalization. Specifically, in FIG. 9, the port arrival time predicted value and the port departure time predicted value are calculated for four ships and are shown in time series. Note that FIG. 9 does not show a ship that has already arrived at the port Z, but for a ship that has already arrived at the port Z, there is no start point on the arrival time t _si side and only t _ei on the end point side. Is calculated and shown.

In FIG. 9, a time zone in which there is a line connecting the predicted port arrival time and the predicted port departure time indicates that the ship is anchored at the port Z. Therefore, the number of ships anchored at the port Z at the time t may be counted by counting the number of ships having a line connecting the predicted port arrival time and the predicted port departure time at the time t. In the case of time t shown in FIG. 9, since there is only a line connecting t _s3 and t _e3 , it can be predicted that there is one ship anchored in the port Z. In this way, the result of counting the change in the number of ships anchored corresponding to the time is the histogram at the bottom of FIG. This result corresponds to the predicted number of anchored vessels. This histogram is created in correspondence with the period indicated by the information specifying the period of prediction included in the instruction to predict the number of anchored vessels.

  Therefore, the anchored vessel number prediction result output unit 22 outputs data along the time series calculated as described above as the anchored vessel number prediction result (S34). Thus, the prediction of the number of anchored vessels by the anchored vessel number prediction system 1 is completed.

  As described above, in the anchored vessel number prediction system 1 and the anchored vessel number prediction system 1 and the anchored vessel number prediction system 1 according to the present invention, in the required time map creation unit 11, the stay position of a vessel sailing to a specific port A required time map (required time distribution information) indicating the distribution of required time from a ship to a specific port is created, and in the berthing time distribution information creation unit 13, a berth time distribution indicating a berth time distribution of a ship in a specific port Create information. Then, by combining these pieces of information with AIS data that can identify the current position of the ship, it is possible to predict the number of ships to be anchored in the port in the future.

  Traditionally, it has been necessary to predict the number of anchored vessels in order to manage ships entering and leaving the port. At this time, conventionally, it has been studied to use the estimated arrival time (ETA) of the ship included in the AIS data. However, since the estimated arrival time is information that is manually input based on the judgment of the crew of the ship, it is conceivable that an input error or forgetting to update to the latest information may occur. Therefore, the accuracy of the information that is the basis for predicting the number of anchored ships at a port was not sufficient.

  In contrast, in the anchored vessel number prediction system 1, the anchored vessel number prediction method, and the anchored vessel number prediction program according to the present embodiment, the required time map (required time distribution information) and the anchored time distribution information based on past AIS data. To predict the number of anchored ships in the future port from the current ship position. Since the required time map (required time distribution information) and the anchored time distribution information used for predicting the number of anchored vessels are information based on actual results, the number of anchored vessels can be predicted with higher accuracy information.

  Specifically, in the berthing number prediction system 1, when the number of berths is predicted, arrival time at a specific port and departure time from a specific port are calculated for each ship. Using the travel time map (travel time distribution information) and berthing time distribution information increases the accuracy of calculating the arrival time and departure time from a specific port as described above. Can be predicted.

  In addition, the required time distribution information is information obtained by calculating, for each mesh, the statistics related to the required time from the ship's stay position to the specific port, so that the required time to the specific port according to the ship's stay position. Time can be easily obtained. In addition, it is possible to prevent the amount of work related to creation of required time distribution information from being extremely increased by adopting a configuration that calculates the statistical value related to the required time for each mesh as in the required time map described above. Can do.

  In addition, when creating the required time distribution information, if there is a time when the ship is stopped on the water, the time required for the specific port can be calculated by removing the stopped time. For example, the required time can be calculated without including the time stopped on the water for time adjustment and the required time distribution information can be created with higher accuracy.

  For the berthing time distribution information, the arrival time and departure time at a specific port are specified from the past AIS data related to the ship, and the berthing time is calculated from these information. The berthing time distribution information can be created with higher accuracy compared to the case where the berthing time distribution information is created based on information input by the crew or the like.

  As described above, the anchored vessel number prediction system, the anchored vessel number prediction method, and the anchored vessel number prediction program according to the embodiment of the present invention have been described, but the present invention is not necessarily limited to the above-described embodiment, and departs from the gist thereof. Various changes can be made without departing from the scope.

  For example, the information that can be used in the anchored vessel number prediction system 1 is not limited to the information described in the above embodiment. Therefore, the required time map and the berthing time distribution information can be created in consideration of the navigation speed and ship type information included in the AIS data. For example, the time required for the destination may differ depending on the ship type. Therefore, it is good also as a structure which prepares more than one required time map created for every harbor in the said embodiment for every ship type. Similarly, a plurality of berthing time distribution information may be prepared for each ship type.

  Moreover, in the said embodiment, although the case where the anchorage number prediction system 1 was implement | achieved by one apparatus was demonstrated, the anchorage number prediction system 1 may be comprised by the some apparatus.

  Moreover, in the said embodiment, the case where the calculation which distinguished the berth in a harbor was performed in the case of the prediction of the number of anchored ships was demonstrated. That is, in the above embodiment, the route data is extracted by determining whether or not the ship is anchored at a specific berth in the harbor from the AIS data. Therefore, the case where the route data is extracted individually for each berth to be anchored has been described. Therefore, in the creation of the berthing time distribution information, an example has been described in which the berthing time distribution is created for each berth and then compiled to create the berthing time distribution information in the port. However, the above processing may be performed in units of ports instead of units of berths. In other words, route data is extracted based on whether a ship is anchored in a specific port, not whether the ship is anchored in a specific berth unit, and the required time map (required) Time distribution information) and berthing time distribution information may be created.

DESCRIPTION OF SYMBOLS 1 Anchorage number prediction system 10 AIS data acquisition part 11 Required time map preparation part 12 Required time map storage part 13 Anchorage time distribution information preparation part 14 Anchorage time distribution information storage part 20 Anchorage number prediction instruction reception part 21 Anchorage number prediction part 22 Anchorage Ship number prediction result output section

Claims (7)

AIS data acquisition means for acquiring AIS data related to navigation of the ship;
Based on the AIS data of the past predetermined period acquired by the AIS data acquisition means, the required time distribution information indicating the distribution of the required time to the specific port from the staying position of the ship navigating to the specific port is created. Time required distribution information creation means,
The berthing time distribution information creating means for creating the berthing time distribution information indicating the berthing time distribution of the ship in the specific port based on the past AIS data acquired by the AIS data obtaining means,
Based on the AIS data acquired by the AIS data acquisition means, the required time distribution information, and the berthing time distribution information, the berthing number prediction means for predicting the number of berthing vessels in the specific port;
The number of anchored vessels prediction system.   The berthing number predicting means is based on the AIS data, the required time distribution information, and the berthing time distribution information, and the arrival time at the specific port relating to each ship that is navigating to the specific port. And a departure time from the specific port, and a departure time from the specific port related to each ship that has arrived at the specific port, and predicting the number of ships anchored at the specific port Item 1. The number of anchored vessels prediction system according to item 1.   3. The required time distribution information is obtained by calculating a statistic relating to a required time from the stay position of the ship to the specific port for each mesh obtained by dividing the surface of the water into predetermined areas. The number of anchored vessels prediction system described in.   When the required time distribution information creating means calculates the required time to the specific port from the stay position of the ship navigating to the specific port from the AIS data of the past predetermined period, the ship is on the water. It is determined whether there is a stoppage time, and if there is a stoppage time on the water, the stoppage time is removed and the required time to the specific port is calculated. The number-of-berthing vessel number prediction system according to any one of claims 1 to 3.   The berthing time distribution information creating means identifies arrival time and departure time at the specific port in each ship anchored at the specific port from the AIS data of the past predetermined period, and the arrival time and the departure time. The berthing time distribution information of the ship in the specific port is created by calculating the berth time in the specific port of each ship based on the berthing according to any one of claims 1 to 4. Ship number prediction system. An AIS data acquisition step for acquiring AIS data relating to the navigation of the ship;
Based on the AIS data of the past predetermined period acquired in the AIS data acquisition step, the required time distribution information indicating the distribution of the required time to the specific port from the staying position of the ship navigating to the specific port is created. Time distribution information creation step,
Based on the AIS data of the past predetermined period acquired in the AIS data acquisition step, the berthing time distribution information creation step of creating berthing time distribution information indicating the berthing time distribution of the ship in the specific port;
Based on the AIS data acquired in the AIS data acquisition step, the required time distribution information, and the berthing time distribution information, the berthing number prediction step of predicting the number of berthing vessels in the specific port;
A method for predicting the number of anchored vessels. A program for predicting the number of anchored vessels in a port, which causes a computer to execute a prediction of the number of anchored vessels,
An AIS data acquisition step for acquiring AIS data relating to the navigation of the ship;
Based on the AIS data of the past predetermined period acquired in the AIS data acquisition step, the required time distribution information indicating the distribution of the required time to the specific port from the staying position of the ship navigating to the specific port is created. Time distribution information creation step,
Based on the AIS data of the past predetermined period acquired in the AIS data acquisition step, the berthing time distribution information creation step of creating berthing time distribution information indicating the berthing time distribution of the ship in the specific port;
Based on the AIS data acquired in the AIS data acquisition step, the required time distribution information, and the berthing time distribution information, the berthing number prediction step of predicting the number of berthing vessels in the specific port;
A program for predicting the number of anchored vessels.