Transportation route determination method, transportation route determination device, and computer program

Abstract

Provided are a transportation route determination method, transportation route determination device, and computer program.　In this transportation route determination method, a transportation network is defined in which a plurality of goods collection / delivery points, transportation hubs that a transportation device for transporting the goods is to pass through, and each point in a traffic network are assigned to nodes disposed in a geometric pattern having a prescribed unit pattern that repeats periodically and routes connecting adjacent points or transportation hubs are assigned to edges between nodes, and a transportation route connecting the transportation hubs for the transportation device is determined as a data sequence of identification data for the nodes of the transportation network.

Description

[Technical Field]

[0001] The present invention relates to a transport route determination method, a transport route determination device, and a computer program for determining a route for transporting logistics components on which goods such as pallets, small containers, or cardboard boxes are loaded in order to improve the distribution efficiency of goods. [Background technology]

[0002] In order to improve the efficiency of logistics as a whole, it is necessary to optimize the logistics components on which goods to be transported are loaded and the routes of the transport equipment that carries the logistics components.

[0003] Methods for setting routes for deliveries have been proposed for some time. Patent Document 1 discloses a method for solving a delivery planning problem that assumes that a single transport vehicle transports a load from a delivery base to a delivery destination in a single trip. Patent Document 1 repeatedly creates delivery plans using different algorithms to use multiple delivery vehicles to deliver cargo from a logistics base to multiple delivery destinations and return to the original base in the shortest possible distance, and selects a plan that results in the shortest total travel distance.

[0004] Patent Document 2 discloses a method using an insertion method that repeats the step of inserting delivery destinations into a provisionally determined delivery route and selecting a route with the shortest driving distance. Patent Document 2 generates a provisional solution by inserting customers along the way on a closed route that goes from a logistics center to multiple delivery destinations, under the constraints that the upper limit of driving time and the maximum load capacity of the vehicle are not exceeded, thereby reducing the driving distance or driving time. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2001-188984 [Patent Document 2] Japanese Patent Application Laid-Open No. 2015-038429 Summary of the Invention [Problem to be solved by the invention]

[0006] The methods of setting delivery routes from bases to delivery destinations as disclosed in Patent Documents 1 and 2 are based on the premise that each transport vehicle delivers goods collected at a base center to the final delivery destination, and do not take into account the possibility that a transport vehicle will transfer goods to another vehicle along the way.

[0007] While design based on simulations is being implemented in a variety of fields, including production technology, goods, and architecture, designing delivery routes in logistics is difficult due to the many uncertain conditions. In logistics, the starting and ending points of goods are not fixed, and the starting and ending points are geographically mixed, collection and delivery time requirements are not fixed, and the number of transportation devices is also variable. Furthermore, allowing goods to be transferred to other vehicles along the way results in multiple uncertain conditions overlapping. Separating these uncertain conditions into multiple patterns and calculating them separately would require an enormous amount of calculation. Because goods departures and arrivals and road conditions change constantly, automatically deriving delivery routes through simulation would be unrealistic if the plan had to be reviewed and recalculated every time.

[0008] The present invention has been made in consideration of the above circumstances, and aims to provide a transportation route determination method, a transportation route determination device, and a computer program that make it possible to make the mechanical derivation of transportation routes more practical. [Means for solving the problem]

[0009] A transportation route determination method according to one embodiment of the present disclosure assigns points where multiple items are collected and delivered, transportation hubs through which transportation equipment transporting the items passes, and each point in a transportation network to nodes arranged in a geometric pattern in which a predetermined unit pattern is repeated periodically, defines a transportation network in which routes connecting adjacent points or transportation hubs are assigned to edges between the nodes, and determines a transportation route connecting the transportation hubs of the transportation equipment as a data string of identification data for the nodes in the transportation network.

[0010] A transportation route determination device according to one embodiment of the present disclosure includes a processing unit that assigns points where multiple items are collected and delivered, transportation hubs through which transportation equipment transporting the items passes, and each point in a transportation network to nodes arranged in a geometric pattern in which a predetermined unit pattern is periodically repeated, defines a transportation network in which routes connecting adjacent points or transportation hubs are assigned to edges between the nodes, and determines a transportation route connecting the transportation hubs of the transportation equipment as a data string of identification data for the nodes in the transportation network.

[0011] A computer program according to one embodiment of the present disclosure causes a computer to execute a process of assigning points where multiple items are collected and delivered, transportation hubs through which transportation equipment transporting the items passes, and each point in a transportation network to nodes arranged in a geometric pattern in which a predetermined unit pattern is periodically repeated, defining a transportation network in which routes connecting adjacent points or transportation hubs are assigned to edges between the nodes, and determining the transportation route connecting the transportation hubs of the transportation equipment as a data string of identification data for the nodes in the transportation network.

[0012] In the transportation route determination method, transportation route determination device, and computer program disclosed herein, the multiple points that make up a transportation route and the roads between the points are approximated as a transportation network expressed in a geometric pattern and converted into simple data for use. By associating distance as separate data rather than reflecting the length that reflects the actual distance in the route connecting the points, agent-based modeling processing that probabilistically selects a route to an adjacent node becomes easier. [Effects of the Invention]

[0013] According to the present disclosure, transportation routes in a certain area are converted into a geometric pattern network including a lattice, which enables network analysis or image analysis. This makes it easier to apply existing analysis methods for network analysis or image analysis, and learning methods can also be applied, making it possible to narrow down the search to derive the optimal transportation route. [Brief explanation of the drawings]

[0014] [Figure 1] 1 is a schematic diagram of a logistics system according to the present disclosure. [Figure 2A] FIG. 1 is a schematic diagram illustrating an example of a transportation device. [Figure 2B] FIG. 1 is a schematic diagram illustrating an example of a transportation device. [Figure 3] FIG. 2 is a block diagram showing the configuration of a transportation route determination device. [Figure 4] FIG. 2 is a block diagram showing the configuration of a terminal device used by an operator. [Figure 5] FIG. 1 is an explanatory diagram of conversion from map data to a transportation network. [Figure 6] 10 is a flowchart illustrating an example of a procedure for advance route calculation. [Figure 7] 10 is a flowchart illustrating an example of a procedure for advance route calculation. [Figure 8] FIG. 10 is a diagram illustrating an example of a method for deriving an optimal route. [Figure 9A] FIG. 10 is a diagram illustrating an example of a method for deriving an optimal route. [Figure 9B]FIG. 10 is a diagram illustrating an example of a method for deriving an optimal route. [Figure 10] 10 is a flowchart showing an example of a transportation route determination process performed by the transportation route determination device. [Figure 11] 10 is a flowchart showing an example of a transportation route determination process performed by the transportation route determination device. [Figure 12] Shows the overall pallet path and its relationship to the segmented transportation network. [Figure 13] 11 is a flowchart showing an example of a procedure for advance route calculation according to the second embodiment. [Figure 14] 11 is a flowchart showing an example of a procedure for advance route calculation according to the second embodiment. [Figure 15] FIG. 11 is an explanatory diagram of pixel representation of a transportation network in the third embodiment. [Figure 16] 1 shows a schematic diagram of training a pixelated transportation network. [Figure 17] FIG. 1 illustrates another representation of a transportation network. [Figure 18] FIG. 1 illustrates a transportation network including bypass edges. [Figure 19] FIG. 1 illustrates an example of a pixel representation of a transportation network including bypass edges. DETAILED DESCRIPTION OF THE INVENTION

[0015] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present disclosure will be specifically described with reference to the drawings showing embodiments thereof. In the following embodiments, a logistics system to which the transportation route determination method of the present disclosure is applied will be described.

[0016] FIG. 1 is a schematic diagram of a logistics system 100 according to the present disclosure. The logistics system 100 includes a transportation device 1 that transports goods, such as agricultural products, a base center (base) 2 that is a transfer center where the transportation device 1 stops, a collection and delivery center 3 that is a distribution center that collects and delivers goods, and a control center 400. The logistics system 100 is operated based on route information determined by a transportation route determination device 4. The transportation route determination device 4 is capable of communication with the transportation device 1, devices in the base center 2, and devices in the collection and delivery center 3. The transportation route determination device 4 is also capable of communication with terminal devices 5 used by operators at each location, such as the producer or manufacturer, the base center 2, and the collection and delivery center 3.

[0017] The transportation equipment 1 is a vehicle such as a transport truck, a train, a transport plane, a ship, etc. The transportation equipment 1 may also be a transport robot that travels by automatic driving.

[0018] The base center 2 is a transit center provided on a so-called trunk line in logistics, and is installed at a location that serves as a base for the transportation equipment 1, such as a port, airport, freight station, or an interchange (IC) in a road network. The base center 2 is provided, for example, at predetermined distances. The base center 2 is provided with a group of devices that accepts the storage of the transportation equipment 1 at the base center 2 and carries in and out pallets P from the transportation equipment 1, and a base controller 20 that controls the group of devices. The base controller 20 is capable of communicating with the transportation route determination device 4, and controls the group of devices based on instructions from the transportation route determination device 4.

[0019] In a logistics system centered on base center 2, transport objects are loaded onto logistics components and transported. Specifically, logistics components are pallets P and small containers C. The following explanation focuses on pallets P, but small containers C can also be treated in the same way. Logistics components may also be flexible containers, so-called FIBCs, steel containers, or cardboard boxes. In addition, bags, plates, and boxes used to place or store items in logistics are also included in logistics components.

[0020] A pallet center 22 may be installed alongside the base center 2. Pallets P are collected and delivered at the pallet center 22. A pallet controller 23 that controls devices that carry in and out pallets P is installed in the pallet center 22. The pallet controller 23 can be connected to a transportation route determination device 4 for communication, and controls a group of devices based on instructions from the transportation route determination device 4.

[0021] The collection and delivery center 3 is a base that collects goods to be transported from the base of a producer or manufacturer and transports them to the base center 2, or conversely, receives goods from the base center 2, stores them, and transports the goods to the end user. The collection and delivery center 3 corresponds to a wholesaler or a distribution center. It is preferable that multiple collection and delivery centers 3 are provided for the base center 2. The collection and delivery center 3 may be managed by the consignee, who is a retailer. When the collection and delivery center 3 collects goods to be transported to the base center 2, the goods may be loaded onto a pallet P at the collection and delivery center 3. A collection and delivery location device 30 that receives instructions from the transportation route determination device 4 is installed in the collection and delivery center 3. The collection and delivery location device 30 receives input of goods to be shipped and pallets P from the operator at the collection and delivery center 3, and receives input of arrived goods. The collection and delivery location device 30 stores the correspondence between the pallet identification information of the pallet P on which the goods to be shipped are loaded and the identification information of the goods, and the correspondence between the identification information of the goods that have arrived and the pallet identification information of the pallet P on which the goods are loaded, and transmits these to the transportation route determination device 4.

[0022] In a region, there are one or more base centers 2 and multiple collection and delivery centers 3. Regions are not limited to administrative divisions and can be defined to be divided into any units, and are stored using latitude and longitude information, identification data for base centers 2 and collection and delivery centers 3, and identification data for the region to which they belong. Regions may overlap. Collection and delivery centers 3 may belong to different regions. Base centers 2 and collection and delivery centers 3 are both collection points and delivery points for transportation within the region (corresponding to "collection and delivery points"). Multiple collection and delivery centers 3 within a region are both collection points and delivery points. Any one collection and delivery center 3 may function only as a collection point, and similarly any other collection and delivery center 3 may function only as a delivery point.

[0023] The transport route determination device 4 sequentially acquires the position of the transport equipment 1, and sequentially collects information on pallets P stored in the transport equipment 1 and information on pallets P awaiting shipping at the collection and delivery center 3. The transport route determination device 4 sequentially determines the movement route of the pallets P and the transport route of the transport equipment 1 from the position sequentially acquired from the transport equipment 1, position information including the location of each pallet P, destination information of the collection and delivery center 3 to which each pallet P should be delivered, and information on the time when it should be delivered.

[0024] The transportation route determination device 4 instructs the transportation equipment 1 on the base center 2 at which it should stop, based on the determined transportation route of each transportation equipment 1 and the movement route of the pallet P. The transportation route determination device 4 instructs the base center 2 on the pallet P to be removed from the arrived transportation equipment 1 and the pallet P to be carried into the transportation equipment 1. Based on the determined transportation route and movement route, the transportation route determination device 4 instructs the collection and delivery center 3 and the pallet center 22 on the pallet P to be shipped, and instructs the transportation equipment 1 on the pallet P to be carried out.

[0025] In this disclosure, a process will be described in which the transportation route determination device 4 sequentially optimizes the movement route of the pallet P and the transportation route of the transportation equipment 1 based on information on the goods to be transported, position information of the transportation equipment 1, and the like.

[0026] Pallets P will now be described. Pallets P are logistics components measuring 90cm square, similar to the pallets widely used in logistics sites, and are preferably made of resin. Pallets P may also be made of a variety of other materials, such as wood, stainless steel, or cardboard, and are preferably provided with lift holes suitable for transportation by forklift. Pallets P should preferably be made of materials that comply with import and export regulations. A tag storing pallet identification information is attached to the pallet P. The tag is preferably a wireless tag using RFID or the like.

[0027] The tag stores pre-assigned pallet identification information of the pallet P so that it can be read by a wireless reader. Note that the pallet identification information of the pallet P on the tag cannot be rewritten, but information about one or more items to be loaded on the pallet P can be written to the tag using a writer. The item information includes the type and item of the item, weight, collection date, item delivery number, base identification information of the most recently passed base center 2, the destination base center 2, consignee information, sender information, etc. Instead of a tag, a predetermined medium on which a one-dimensional code or two-dimensional code corresponding to the pre-assigned pallet identification information of the pallet P may be printed may be used. The one-dimensional code or two-dimensional code may be printed on the tag. The item information is stored in association with the pallet identification information on the transport route determination device 4 side.

[0028] In this embodiment, the transportation equipment 1 is a transport truck. Figures 2A and 2B are schematic diagrams showing an example of the transportation equipment 1. Figure 2A shows an example of a loading platform with a side-opening door, and Figure 2B shows an example of a loading platform with a rear door. In both of the examples shown in Figures 2A and 2B, the transportation equipment 1 has a pallet frame 11 provided inside the loading platform to transport items in units of pallets P. Specifically, the pallet frame 11 is a platform that divides the loading platform into two levels, upper and lower, in half in height from the floor to the ceiling. The loading platform has internal dimensions in the width direction of the vehicle that allow two pallets P to be placed side by side. The storage position of the pallet P can be identified by the upper and lower levels and the left and right levels divided by the pallet frame 11.

[0029] The pallet frame 11 may be configured so that it can be pulled out entirely from the rear door of the loading platform, as shown in Figure 2B. In this case, if the pallet frame 11 is considered to be a giant pallet, the pallet frame 11 can also be managed as a logistics component in a nested state, with multiple pallets P loaded on it and multiple small containers C loaded on each pallet P. Note that the pallet frame 11 does not have to be a two-tiered platform as shown in Figures 2A and 2B, but may be in the form of a plate laid on the floor of the loading platform.

[0030] When the transportation equipment 1 is a vehicle such as a train, a transport aircraft, or a ship, a large container having a structure similar to that of the loading platform portion of FIG. 2A or FIG. 2B is used, and the large container may be configured to include a pallet frame 11.

[0031] The transportation equipment 1 is equipped with an on-board device 10 and a reader that reads pallet identification information from a tag on the pallet P. The on-board device 10 has a GPS receiver and sequentially acquires position information of the transportation equipment 1 and transmits it to the transportation route determination device 4. When a pallet P is carried into the loading platform of the transportation equipment 1, the on-board device 10 reads the pallet identification information from the tag on the pallet P with the reader. The pallet identification information of the stored pallet P and storage unit identification information that identifies the storage position in the pallet frame 11 are transmitted to the transportation route determination device 4 in association with pre-stored identification information of the transportation equipment 1. The on-board device 10 reads the pallet identification information from the tag of the pallet P that is being removed from the loading platform of the transportation equipment 1 with the reader, and transmits the pallet identification information of the removed pallet P to the transportation route determination device 4 in association with pre-stored identification information of the transportation equipment 1. The on-board device 10 may notify the transportation route determination device 4 of the removal of the pallet P by associating it with the storage unit identification information of the storage unit in which the pallet P was stored.

[0032] This allows the transport route determination device 4 to constantly grasp the position information of the pallet P. The transport route determination device 4 associates the position information of the transport equipment 1 transmitted from each transport equipment 1 with the equipment identification information of the transport equipment 1 and stores it as transport equipment information 413 (see FIG. 3). The transport route determination device 4 associates the equipment identification information and storage unit identification information of the transport equipment 1 carrying the pallet P, or the identification information of the base center 2 or collection and delivery center 3 where the pallet P is placed, with the pallet identification information of the pallet P and stores it as pallet information 412 (see FIG. 3).

[0033] 3 is a block diagram showing the configuration of the transportation route determination device 4. The transportation route determination device 4 is a server computer and includes a processing unit 40, a storage unit 41, and a communication unit 42. The transportation route determination device 4 may be configured to use one server computer (hardware), or may be configured to distribute processing among multiple server computers, or may be one of multiple server computers (instances) virtually generated in a large computer. The transportation route determination device 4 may perform calculations using a quantum computer, except for the process of updating information in the logistics DB 410 in the storage unit 41.

[0034] The processing unit 40 is a processor that uses a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit). The processing unit 40 executes processing using built-in memories such as a ROM (Read Only Memory) and a RAM (Random Access Memory). The processing unit 40 can sequentially acquire time information using a built-in timer.

[0035] The storage unit 41 includes a non-volatile storage medium such as a hard disk or an SSD (Solid State Drive). The storage unit 41 stores a control program 40P. The processing unit 40 executes a process of deriving an optimal route by calculation, which will be described later, based on the control program 40P stored in the storage unit 41.

[0036] The storage unit 41 stores a Web server program, and the processing unit 40 performs a Web server function, which may receive requests for sharing the loading platform of the transport equipment 1 from the terminal device 5.

[0037] A logistics DB (Data Base) 410 is constructed in the storage unit 41 or an external storage device. The processing unit 40 is capable of reading and writing data from and to the logistics DB 410 using a database operation module. The logistics DB 410 stores, for example, user information 411, pallet information 412, and transportation equipment information 413, as will be described later. The logistics DB 410 also stores transportation route information 414 that has been derived in advance by a calculation, which will be described later.

[0038] The communication unit 42 realizes communication over the public communication network N1 or the carrier network N2. The processing unit 40 can transmit and receive information to and from the terminal device 5 via the public communication network N1 or the carrier network N2 using the communication unit 42. The processing unit 40 can also transmit and receive information to and from an onboard device mounted on the transportation equipment 1 via the carrier network N2 using the communication unit 42. The processing unit 40 can communicate with the base controller 20 of the base center 2 and the collection and delivery location device 30 of the collection and delivery center 3 via the public communication network N1, the carrier network N2, or a dedicated line using the communication unit 42.

[0039] 4 is a block diagram showing the configuration of a terminal device 5 used by an operator. The terminal device 5 is a desktop, laptop, or tablet personal computer. The terminal device 5 may also be a smartphone. The terminal device 5 includes a processing unit 50, a storage unit 51, a communication unit 52, a display unit 53, and an operation unit 54.

[0040] The processing unit 50 is a processor using a CPU or a GPU. The processing unit 50 executes processing based on a program stored in a storage unit 51 using built-in memories such as a ROM and a RAM.

[0041] The storage unit 51 includes a non-volatile storage medium such as a hard disk or an SSD. The storage unit 51 stores a terminal program 50P. The processing unit 50 receives an operation from an operator based on the terminal program 50P stored in the storage unit 51 and executes a process of displaying data.

[0042] The communication unit 52 realizes communication in the public communication network N1 or the carrier network N2. The processing unit 50 can transmit and receive information to and from the transportation route determination device 4 via the public communication network N1 or the carrier network N2 using the communication unit 52.

[0043] The display unit 53 is a display such as a liquid crystal display, an organic EL (Electro Luminescence) display, etc. The display unit 53 displays a screen based on the data stored in the storage unit 51.

[0044] The operation unit 54 is a user interface such as a keyboard and a pointing device that can input and output data to and from the processing unit 50. The operation unit 54 may be a voice input unit. The operation unit 54 may be a touch panel of the display unit 53.

[0045] In the logistics system 100 configured as described above, the transportation route determination device 4 determines the transportation route of the transportation device 1 and the movement route of the pallet P based on the positions of each transportation device 1 and the positions of the pallet P, and delivery conditions such as the scheduled shipping date and required arrival time of the goods. The transportation route of the transportation device 1 includes data connecting the waypoints that the transportation device 1 stops at in order for collection or loading / unloading from the base center 2 or collection / distribution center 3 at the departure point (shipping point) to the base center 2 or collection / distribution center 3 at the destination. The movement route of the pallet P is the history of the storage units of one or more transportation devices 1 in which the pallet P is stored on its way from the departure point to the destination. The pallet P may be transported to the destination by one transportation device 1, but it is also possible to transfer from one transportation device 1 to another transportation device 1 at waypoints such as the base center 2 or collection / distribution center 3 in the delivery destination area.

[0046] When determining the transportation route of the transportation equipment 1 and the movement route of the pallet P, the transportation route determination device 4 not only starts calculations after receiving actual delivery conditions (the position of the transportation equipment 1, the position of the pallet P, the dispatch schedule, and the arrival time), but also derives an optimal route within the target area in advance. When the actual delivery conditions are given, the transportation route determination device 4 extracts matching or similar routes from the optimal routes derived for various conditions derived in advance, and performs a simulation based on the extracted routes to derive the actual transportation route and movement route. The optimal route may be completed by selecting partial routes for each area.

[0047] This section explains how to derive the optimal route in advance. The route is derived in advance by first creating a transportation network from the geographic information of the target area. Secondly, the route calculation process is carried out by simulating the agent-based modeling using the converted transportation network.

[0048] Figure 5 is an explanatory diagram of the conversion from map data to a transportation network. Image A in Figure 5 shows an example of the original map data, image B shows a transportation network created from the map data, and image C shows a segmented transportation network. As shown in A and B, the transportation network is created by assigning intersections, branching points, or intermediate points (base center 2 or distribution center 3, and also pallet P supply points) to vertices (nodes) arranged in a grid pattern from a road network that is by no means a grid. The geographical distance between each point does not apply to the length of the transportation network, but can simply be stored as separate data as the distance between nodes. The transportation network B is segmented into sections large enough to allow for calculations (Figure 5(C)). The size large enough to allow for calculations can be appropriately set based on the processing capacity of the transportation route determination device 4, based on calculation time trials as described below. As shown in images B and C, edges are removed if there is no corresponding road on the map. By converting to such a transportation network, the road traffic network can be standardized.

[0049] A transportation route can be expressed as a sequence of node identification data (rows and columns). For example, the transportation route of transportation device 1 starting from the central point in Figure 5C can be expressed as point (i, j) → point (i, j+1) → point (i-1, j+1) → ...

[0050] 5 may be automatically created by the processing unit 40 of the transportation route determination device 4 based on map data acquired from the communication unit 42. Alternatively, the transportation network may be created by displaying a transportation network temporarily created by the processing unit 40 of the transportation route determination device 4 or another computing device on the display unit 53 of the terminal device 5, and accepting modifications such as deletion of edges or addition of edges from the operator of the terminal device 5.

[0051] The transportation route determination device 4 divides the grid-like transportation network shown in Fig. 5 into sections as needed to allow for calculation, and executes the second simulation by providing delivery conditions for each transportation network. The division unit may be set according to the processing capacity of the transportation route determination device 4. The divided ranges may partially overlap, for example, with the base center 2 at the center. For example, it has been found that when the transportation network size increases from 3x3 to 5x5, the calculation time described below increases by 20 times.

[0052] Next, we will explain how to calculate a route in advance using an agent-based modeling simulation that uses a transportation network. Figures 6 and 7 are flowcharts showing an example of the processing procedure for calculating a route in advance. The transportation route determination device 4 (or another calculation device) repeats the following processing for each transportation network by changing the delivery conditions, and derives an optimal route for each delivery condition.

[0053] The processing unit 40 sets delivery conditions for the target transportation network, including the initial position and number of transportation equipment 1, the initial position and number of pallets P, and the destination of each pallet P within the network (S101). For example, the delivery conditions may be that the transportation equipment 1 is located at a node corresponding to a base center 2 in the transportation network, 100 pallets P are present at that specific node, and 40, 30, and 20 of these pallets P are transported to three other nodes. The number of transportation equipment 1 may be one or more, or the delivery conditions may state that the transportation equipment 1 is located at different nodes. The type of pallets P may be one or more.

[0054] The processing unit 40 sets the number of executions to an initial value (S102). The processing unit 40 executes the following process multiple times, and derives the route that requires the fewest number of executions (the number of steps from the state where the transportation equipment 1 is present at a node to the state where it moves to the next node) until the pallet P that was present at a specific node that is the starting point is distributed to the delivery points within the transportation network, or the route that requires the shortest transportation travel distance of the transportation equipment 1.

[0055] The processing unit 40 sets the number of steps to an initial value (S103). As shown below, the processing unit 40 executes processing based on agent-based modeling in which the transportation equipment 1 is an agent, until the pallet P that was present at a specific node that is the departure point is completely distributed to each destination within the network.

[0056] In this step, the processing unit 40 acquires the number of pallets P (loaded pallets P) associated with the agent (transportation equipment 1) (S104). The processing unit 40 acquires the number of pallets P present at the node where the agent is located (S105). The processing unit 40 determines and stores the number of pallets P to be stacked at the node where the agent is located (S106).

[0057] In S106, if the node where the agent is located is the departure point, the processing unit 40 determines to load pallets P present at the departure point so as to fill the empty storage compartment of the transportation equipment 1 that is the agent. If the node where the agent is located is a waypoint or destination, the processing unit 40 compares the number of pallets P stored in the storage compartment with the number of pallets P required for the node, and determines the number of pallets P so as to reduce the remaining number of pallets P required for the node.

[0058] The processing unit 40 probabilistically selects an edge to the next adjacent node to proceed to (S107). Of course, edges that do not exist in the transportation network are not selected here.

[0059] In the selection process of S107, the processing unit 40 basically selects probabilistically using random numbers (Solution A). As will be described later, the selection may be made based on a probability based on the weight for the node to be advanced to in the next step, or may be made under the condition that the agent returns to the starting point (or supply point) when the number of pallets (loading number) associated with the agent becomes zero (Solution B). The processing unit 40 may learn a weighting factor for selecting a node by reinforcement learning that provides a reward such as loading efficiency each time a step is advanced (Solution C). In this case, the processing unit 40 is more likely to arrive at an optimal solution more quickly. The processing unit 40 may also make a large change to a parameter (e.g., probability) by one step, such as by mutation (Solution D). This reduces the possibility of falling into a local solution.

[0060] The processing unit 40 calculates the number of pallets P present at the node to which the agent has moved via the selected edge when the loaded pallets P are unloaded at that node (S108). In S108, the processing unit 40 calculates the sum of the number of pallets P that existed at the node before the agent moved and the number associated with the agent.

[0061] The processing unit 40 adds up and stores the number of steps (S109), and stores the identification data of the node to which the agent has moved and the number of pallets P at each node (S110). The processing unit 40 may store the state of each node after the agent has moved, treating it as a multidimensional matrix (vector).

[0062] The processing unit 40 determines whether or not delivery that meets the delivery conditions has been completed within the transportation network based on the number of pallets P at each node (S111).

[0063] If it is determined that the delivery is not completed (S111: NO), the processing unit 40 adds the number of steps and the distance associated with the edge to the cumulative travel distance (S112), and returns the process to S104. In S112, the processing unit 40 may add only either the number of steps or the cumulative travel distance for that execution.

[0064] If it is determined that the delivery is completed (S111: YES), the processing unit 40 stores the history of the identification data of the nodes that the agent passed through (transportation route) and the number of steps (S113).

[0065] The processing unit 40 determines whether the processes of S104-S113 have been executed a predetermined number of times or more (S114). If it is determined that the processes have been executed less than the predetermined number of times (S114: NO), the processing unit 40 increments the number of executions (S115) and returns the process to S103.

[0066] In S114, the processing unit 40 may determine whether or not the execution has been performed a predetermined number of times or more, but whether or not it is determined that the optimal solution (minimum number of steps or shortest cumulative distance) has been obtained through the execution up to that point, or whether or not the optimal solution can be derived by further execution.

[0067] If it is determined in S114 that the process has been executed a predetermined number of times or more (S114: YES), the processing unit 40 extracts a transportation route with the minimum number of steps or the shortest cumulative travel distance within the predetermined number of times (S116). The processing unit 40 stores the extracted transportation route as an optimal solution for the delivery conditions (S117), and ends the process.

[0068] The optimum route for each transportation network derived by the processing procedures shown in the flowcharts of FIGS. 6 and 7 is stored in the storage unit 41 of the transportation route determination device 4 as transportation route information 414 for each delivery condition.

[0069] 8, 9A, and 9B are diagrams showing examples of a method for deriving an optimal route. FIG. 8 is a schematic diagram of a transportation network. Each node is assigned identification information in the form of a queue number. An agent (transportation equipment 1) initially located at i, j selects an adjacent node probabilistically. Here, the probability may be completely random, may be based on a pre-set weight, or may be described as a function of a state described by the number of pallets P remaining at the departure point, the number of pallets P missing at the destination, the number of pallets loaded by the agent, etc.

[0070] This shows an example of applying a reinforcement learning technique. Figures 9A and 9B are schematic diagrams of a transportation network. The transportation networks in Figures 9A and 9B are assigned numerical values ​​representing evaluations (rewards) for each state. Figure 9A shows the initial state, and Figure 9B shows the time-series node selection history of the agent. In Figures 9A and 9B, each node is assigned identification information in the form of a queue number, and the starting point is node (i+1, j), there are two pallets P, and the delivery conditions are that they should be delivered to node (i, j-1) and node (i+1, j+1), respectively.

[0071] In Figure 9A, initially, the departure point is given a negative reward equal to the number of pallets P, the agent is given a positive reward equal to the number of pallets P loaded, and the agent is given a positive reward equal to the number of pallets P delivered to the destination. By performing reinforcement learning with the profit as the sum of the rewards, learning to select each edge is realized.

[0072] Next, a description will be given of the processing procedure for determining the movement route of the pallet P and the transportation route of the transportation equipment 1 based on an actual order for goods and a transportation request. The transportation route determination device 4 sequentially updates the logistics DB 410 based on information obtained from the on-board device 10 due to the movement of the transportation equipment 1 and the dispatch of goods, and on information on the location of goods obtained from the base center 2 and the collection and delivery center 3. The transportation route determination device 4 determines the movement route of the transportation equipment 1 and the pallet P of the transportation equipment 1, and controls transportation, loading and unloading at the transportation equipment 1, base center 2, and collection and delivery center 3 based on this update of the logistics DB 410.

[0073] 10 and 11 are flowcharts showing an example of a transportation route determination process performed by the transportation route determination device 4. In FIG.

[0074] The processing unit 40 of the transportation route determination device 4 acquires the arrangement of the pallet P from the pallet information 412 and the transportation equipment information 413 (S201).

[0075] The processing unit 40 acquires conditions indicating where each pallet P should be transported after a predetermined time or at a specific time based on the destination and desired arrival time of each pallet P (S202). In S202, the processing unit 40 tallies the number of pallets P at each base center 2 or collection and delivery center 3 that should be there, for example, three hours after the processing of S201.

[0076] The processing unit 40 determines the transportation conditions in the transportation network based on the current position of the pallet P and the position of the transportation equipment 1 obtained in the processing of S201, and the desired conditions of the pallet P and the transportation equipment 1 after a predetermined time obtained in the processing of S202 (S203).

[0077] In the process of S203, the processing unit 40 determines the transportation conditions for each time period, i.e., where the pallet P should be in the transportation network and how far it should reach. FIG. 12 shows the relationship between the overall route of the pallet P and the divided transportation network. The base centers 2 are connected in a net-like manner. As shown in FIG. 12, each pallet P passes through multiple divided transportation networks spanning from the base center 2 at its departure point to the collection and distribution center 3 at its final destination. In the example of FIG. 12, one pallet P departs from the base center 2 in Nagano Prefecture and is destined for the collection and distribution center 3 in Shiga Prefecture. By determining where the pallet P should be transported from based on its position at each time point and overlaying the data for multiple pallets P, the transportation conditions can be determined for each divided transportation network, such as how many pallets P should be collected at the base center 2 and transported to another base center 2 or collection and distribution center 3.

[0078] The processing unit 40 selects one transportation network (S204), and extracts and stores the optimal route with delivery conditions similar to the transportation conditions for that transportation network from those stored in the storage unit 41 (S205). In S205, the processing unit 40 derives the Euclidean distance from the delivery conditions as a vector having dimensions of the node identification information of the departure point, node identification information of the destination, the type of pallet, and the number of pallets, which describe the transportation conditions, and extracts those with shorter distances as those with higher similarity.

[0079] In S205, the processing unit 40 performs route calculations (FIGS. 6 and 7) based on agent-based modeling for the optimal route with similar delivery conditions, based on the difference from the transportation conditions, to derive the optimal route for each condition. When selecting nodes probabilistically, the simulation may be performed by selecting edges similar to the optimal route with similar delivery conditions. A reinforcement learning method may also be incorporated to achieve the shortest cumulative travel distance.

[0080] The processing unit 40 determines whether or not all transportation networks have been selected (S206), and if it determines that all networks have not been selected (S206: NO), returns the process to S204.

[0081] If it is determined that all transportation networks have been selected (S206: YES), the processing unit 40 acquires and stores the number of transportation equipment 1 required for each transportation network, the optimal transportation route for each required number of transportation equipment, and its delivery conditions (type, number, and number of pallets P to be transported) (S207).

[0082] The processing unit 40 assigns and stores pallet identification information to the pallet storage unit on the loading platform of the transportation equipment 1 at each location based on information about the pallets P already loaded and the information acquired in the processing of S207 (S208).

[0083] The processing unit 40 selects a transportation route in the divided transportation network as a partial route for each transportation equipment 1, and determines an entire transportation route (S209). In S209, the processing unit 40 may determine a route in the transportation network for the transportation equipment 1 as an entire transportation route.

[0084] The processing unit 40 determines (updates) the movement route of each pallet P, that is, the pallet storage unit of the transportation equipment 1 to which the pallet P should be transferred before reaching the destination (S210).

[0085] Based on the movement route of each pallet P, the processing unit 40 lists, for each transport equipment 1, the pallet identification information of the pallet P to be transported at the base center 2, which is the transit point, and the storage unit identification information of the storage unit in which the pallet P is stored (S211).

[0086] Based on the movement route of each pallet P, the processing unit 40 lists, for each transport equipment 1, the pallet identification information of the pallet P to be transported at the base center 2, which is the transit point, and the storage unit identification information of the pallet storage unit where the pallet P should be stored (S212).

[0087] The processing unit 40 outputs the determined transportation route of the transportation equipment 1, the movement route of the pallet P, and a list of pallets P to be carried in and out at each base center 2 (S213). Based on the output contents, the processing unit 40 transmits instructions to the on-board device 10, the base controller 20, the collection and delivery area device 30, and the pallet controller 23 (S214), and ends the processing.

[0088] In this way, the transportation route of the transportation equipment 1 and the movement route of the pallet P can be determined by referring to the optimum route calculated in the divided transportation network, and the movement of the entire pallet P can be controlled based on this.

[0089] In the processing steps shown in the flowcharts of Figures 10 and 11, it is preferable to perform step S205 of the process, rather than simply selecting a previously derived optimal route, and instead add calculations related to the differential conditions based on an optimal route with similar conditions to the previously derived route. For example, for a delivery condition specifying the departure point, destination, and type and number of pallets P in a 5x5 transportation network, in addition to the information on the number of transportation devices and transportation routes that have already been calculated, calculations are performed and integrated to calculate the number of transportation devices and transportation routes corresponding only to the additional collection and delivery points, destinations, and number of pallets. By storing the route calculation results in advance, the only calculation time required is the time to access the data, thereby shortening the calculation time.

[0090] (Embodiment 2) In the second embodiment, the process of stochastically moving the placement of the agent and pallet P one step at a time from the initial state is repeated a predetermined number of times, and the search is terminated when it becomes unlikely that a better route can be derived than the transportation route with the shortest number of steps so far.

[0091] Figures 13 and 14 are flowcharts showing an example of a processing procedure for advance route calculation in embodiment 2. Among the processing procedures shown in the flowcharts of Figures 13 and 14, steps common to the steps shown in the flowcharts of Figures 9 and 10 are given the same reference numerals, and detailed description thereof will be omitted.

[0092] In the second embodiment, the processing unit 40 defines a distance-based transportation network within the region in which adjacent base centers 2 or collection and distribution centers centered on the selected base center 2 are nodes, and the routes between the nodes are edges, with distance information associated with the edges (S121).

[0093] When the processing unit 40 calculates the number of pallets P present at the node to which the agent has moved (S108), it not only adds the number of steps but also adds and stores the distance associated with the moved edge as the cumulative movement distance (S129).

[0094] When the processing unit 40 determines in the process of S111 that delivery within the transportation network is completed (S111: YES), it stores the history of identification data of the nodes that the agent passed through (transportation route), the accumulated travel distance, and the number of steps (S132). As a result, the number of steps, travel distance, and transportation route required to complete each delivery are stored.

[0095] If it is determined in the processing of S111 that delivery within the transportation network is not complete (S111: NO), the processing unit 40 adds up the number of steps and the cumulative travel distance (S112) and calculates the progress rate of delivery throughout the entire transportation network (S133). In S133, the processing unit 40 calculates the ratio of the number of pallets P that have reached their destination within the area to the total number of pallets P as the progress rate. In S133, the processing unit 40 may also calculate the ratio of the number of undelivered pallets P to the total number as the progress rate.

[0096] In the process of S133, the processing unit 40 may derive a criterion for determining the timing for determining whether or not the route being calculated in S137 described later is likely to result in the shortest route. Therefore, the criterion is not limited to the progress rate, and may be the number of calculations or time.

[0097] The processing unit 40 determines whether the progress rate calculated in the process of S133 satisfies the condition (S134). In S134, for example, the processing unit 40 determines whether the progress rate (%) is a multiple of 10. Alternatively, if it is the number of calculations instead of the progress rate, it may be determined whether it is a multiple of 5, etc., or if it is time, it may be determined whether the elapsed time is a multiple of 5, etc.

[0098] If it is determined in the process of S134 that the condition is not satisfied (S134: NO), the processing unit 40 returns the process to S104 and proceeds to the next step.

[0099] If it is determined in S134 that the condition is satisfied (S134: YES), the processing unit 40 stores the moving distance so far in association with the progress rate calculated in the process of S133 (S135). The processing unit 40 compares the stored moving distance with the moving distance of the shortest moving distance transport route among the moving distances for which the transport route has been calculated multiple times, with a width, with the distance at the progress rate calculated in the process of S333 (S136).

[0100] In S136, the processing unit 40 calculates, for example, as shown in the following formula (1). Let the moving distance so far be k, and the moving route at the progress rate (prog) of the shortest distance transport route among the transport routes obtained so far be kmin prog. Then, formula (1) is k < kmin + kmin × (1 - progress rate)…(1) kprog < kmin prog + kmin prog × (1 - progress rate(prog))…(1) That is. The moving distance k10 when the progress rate is 10% is compared with 1.9 times the moving distance kmin 10 at the progress rate of 10% of the shortest distance transport route. The moving distance k20 when the progress rate is 20% is compared with 1.8 times the moving distance kmin 20 at the progress rate of 20% of the shortest distance transport route. When the progress rate is 90%, the moving distance k90 is compared with 1.1 times the moving distance kmin 90 at the progress rate of 90% of the shortest distance transport route. If the moving distance k obtained so far is greater than or equal to the comparison value, it can be determined that there is no prospect of deriving a shorter distance transport route.

[0101] The processing unit 40 determines whether or not there is a possibility that the shortest transportation route can be derived by continuing the delivery simulation from now on (S137) based on the comparison process result of S136. If it is determined that there is no possibility (S137: NO), the processing unit 40 ends the current calculation midway and proceeds to the process of S114.

[0102] If it is determined in S137 that there is a possibility (S137: YES), the processing unit 40 continues the calculation and returns the process to S104.

[0103] This allows the calculation to be interrupted if there is no prospect of deriving the shortest path even if the calculation is continued further, thereby shortening the calculation time.

[0104] (Embodiment 3) In the third embodiment, the transportation network is further converted from the form shown in Fig. 5 into a pixel representation for use in calculations. The hardware configuration of the logistics system 100 in the second embodiment is the same as that in the first embodiment, so the same reference numerals are used for the common components and detailed description thereof will be omitted.

[0105] Figure 15 is an explanatory diagram of a pixel representation of a transportation network in embodiment 3. The image in the upper left of Figure 15 shows a two-dimensional image corresponding to the pixel representation of the transportation network, and the image in the middle left of Figure 15 shows a two-dimensional image showing the delivery conditions for that pixel representation. The image in the upper left of Figure 15 and the image in the middle left of Figure 15 are checkerboard-shaped images of the same size.

[0106] As shown in the upper left image of FIG. 15 , in the second embodiment, the transportation network is represented by a two-dimensional image with a checkered pattern. Every other square in the vertical and horizontal directions corresponds to a node in the transportation network. Squares adjacent to a square corresponding to a node in the vertical and horizontal directions correspond to edges between the nodes, i.e., roads between points. Squares corresponding to nodes and squares corresponding to edges are represented in different colors. If there are no roads connecting the nodes, they are represented as transparent or colorless (white). It is preferable to represent the nodes with different attributes, such as patterns or marks, rather than just colors. The pixel representation of the transportation network can be divided into a departure image shown in the middle left of FIG. 15 , in which the squares of the departure point under the placement conditions are indicated by specific hatching; a destination image shown in the middle right of FIG. 15 , in which the delivery point under the placement conditions is indicated; and a road image corresponding to geographic information shown in the image at the bottom right of FIG. 15 . In the departure image, pixel ranges corresponding to nodes of points where pallets P are initially collected, such as collection points such as base center 2, and supply points, are represented by specific hatching.

[0107] As shown in the upper left image of Figure 15, a pixel representation combining an origin image, a destination image, and a road image allows for a simple description of a transportation network including nodes that are candidate transit points and edges that are roads between the nodes.

[0108] Furthermore, in the two-dimensional image in the upper left of Figure 15, the geographical distance of the edge may be represented by the shade of color or the level of brightness in the pixel range, and in the two-dimensional image in the middle left of Figure 15, the number of palettes P may be represented by the shade or level of brightness in the pixel range corresponding to the point where the palette P is located.

[0109] Standardizing the data in this way makes it easier to process transportation networks in different regions using the same computer program. Geographic information relating to locations on a map is approximated by the vertex positions of the transportation network. Roads existing in the geographic information are represented as edges of the transportation network. Diagonal roads, three-way intersections, branching points, corners, etc. may also be approximated by the length and width of the transportation network. This standardization makes it easier to handle the data in the manner described below.

[0110] The usage of the pixel representations shown in Fig. 15 will be described. These pixel representations may be displayed on the display unit 53 of the terminal device 5 and used to display the status of a route search or an optimal route. They may also be used as input data and output data for deep learning to derive an optimal route. In order to allow an operator to visually confirm the results of the shortest route derived in the first embodiment, the pixel representations may be displayed on the display unit 53 of the terminal device 5.

[0111] This section explains how to derive the optimal route when deep learning is employed. As shown in Figure 15, the transportation network and its delivery conditions can be treated as image data. Therefore, learning using a convolutional neural network with a proven track record in image processing makes it possible to output data that indicates the weighting of edges to be selected (Figure 16).

[0112] Fig. 16 shows a schematic diagram of learning a transportation network expressed in pixels. For example, as shown in Fig. 16, when an image of a transportation network expressed in pixels (the image in the upper left of Fig. 15) is input to a model using a neural network, it becomes possible to output the number of transportation equipment 1 and the transportation route as a vector (a sequence of identification data for the transportation equipment number and the vertices of the transportation route).

[0113] As shown in Figure 15, the transportation route determination device 4 may separate the image into three parts: a departure image, a destination image, and a road image, input each of them into a model using a neural network, and learn to output a vector indicating the number of transportation equipment 1 and the transportation route.

[0114] In this way, by using a transportation network that expresses delivery conditions and road conditions in pixel form, it is possible to apply techniques that have been developed in neural networks, such as image recognition and pattern recognition.

[0115] Furthermore, by using pixel representation as shown in Fig. 15, an operator operating the terminal device 5 can intuitively understand the transportation network by visually checking the transportation network displayed on the display unit 53. Furthermore, by visually checking the transportation network as shown in Fig. 15 displayed on the display unit 53, the operator can intuitively understand the delivery conditions, such as at which node the pallet P is located and to which node it should be transported. Therefore, it becomes easy to at least partially automate the determination of the validity of the calculation results and the correction of the number of transportation equipment 1 and transportation route, which have conventionally relied on the intuition of an expert.

[0116] In the first to third embodiments described above, the transportation network has a grid, lattice, or checkerboard pattern. However, the definition of a transportation network is not limited to these vertical and horizontal grid patterns. FIG. 17 illustrates another representation of a transportation network. As shown in FIG. 17, a transportation network may be defined by nodes arranged in a periodically repeating equilateral triangle shape and edges connecting the nodes, with collection and delivery points, transportation hubs, and various points in the transportation network assigned to the nodes. Furthermore, as shown in FIG. 17, the transportation network may be defined by a two-dimensional image including geometric patterned areas, rather than a checkerboard pattern. In FIG. 17, the areas corresponding to nodes arranged at a predetermined interval in any two directions in the pattern are represented by dark gray (hatching), and the areas corresponding to the edges between the nodes are represented by a dotted pattern different from the areas corresponding to the nodes. Defining a transportation network using such a simple shape facilitates route calculation. The geometric pattern is not limited to equilateral triangles; it may also be a honeycomb network consisting of regular hexagons.

[0117] The transportation network may be expressed in multiple three-dimensional forms based on the actual state of the transportation network. Here, the actual state of the transportation network refers to the existence of routes such as expressways and motorways for roads, and railroads, airplanes, and ships if different types of transportation 1 are used. These different routes may be defined as bypass edges. Figure 18 is a diagram showing a transportation network including bypass edges. In Figure 18, bypass edges are shown with thick dashed lines regardless of their type. Different layers may be used for different types of transportation 1, such as railroads, airplanes, and ships. The transportation network may include bypass edges expressed on a layer containing only the expressway network. By including these bypass edges in the selection process of S107, it becomes possible to perform route calculations for complex systems even in simplified transportation networks.

[0118] A transportation network including bypass edges may be defined by superimposing two-dimensional checkerboard images. FIG. 19 is a diagram illustrating an example of a pixel representation of a transportation network including bypass edges. As shown in FIG. 19, the transportation network is defined by superimposing two two-dimensional checkerboard images. In FIG. 19, the image superimposed at the top is the same as the pixel representation of the transportation network in FIG. 15. Grids corresponding to nodes and grids corresponding to roads are shown using pixel representations with different attributes, such as colors, patterns, or marks. In FIG. 19, the image shown at the bottom is an example of an image of a bypass edge. In the pixel representation of a bypass edge, grids corresponding to nodes to which the bypass edge is connected are represented using specific attributes. In the image of the bypass edge, bypass edges that are expressways are represented using grids with the same attributes as roads, and bypass edges that are water transportation are represented using grids with different attributes from roads. By representing different types of transportation networks using pixels in this way, the operator can visually recognize them in a simplified manner on the display unit 53, and techniques developed in neural networks, such as image recognition and pattern recognition, can be applied to route calculations, etc.

[0119] The embodiments disclosed above are illustrative in all respects and are not restrictive. The scope of the present invention is defined by the claims, and includes all modifications within the meaning and scope of the claims. [Explanation of symbols]

[0120] 1 Transport equipment P Palette 4. Transportation route determination device 40 Processing section 41 Storage section 410 Logistics DB 414 Transportation Route Information 40P control program

Claims

1. The computer A transportation network is created based on map data, in which collection and delivery points for a plurality of items, transportation bases through which transportation equipment for transporting the items can pass, and each point in the transportation network are assigned to nodes arranged in a geometric pattern in which a predetermined unit pattern is periodically repeated, and routes connecting adjacent points or transportation bases are assigned to edges between the nodes; A transportation route connecting the transportation bases of the transportation equipment directly or via each of the transportation bases is determined as a data string of identification data of the nodes in the transportation network. Transport routing method.

2. The transportation network is defined by nodes and edges arranged in a grid pattern. The transportation route determination method according to claim 1 .

3. Each point in the transportation network includes a branch point in the transportation network. The transportation route determination method according to claim 1 or 2.

4. The computer Displaying the transportation network on a display; Accepting a correction to the displayed transportation network by deleting or adding the node or edge through an operation unit; Determine the transportation route on the corrected transportation network. The transportation route determination method according to any one of claims 1 to 3.

5. Each of the edges is associated with geographical distance data. The transportation route determination method according to any one of claims 1 to 4.

6. The transportation network is defined as a two-dimensional image of a geometric pattern; The nodes of the transportation network are represented by regions having predetermined attributes including predetermined colors, patterns, or marks arranged at predetermined intervals in any two directions in the geometric pattern, and the edges between the nodes are located between the regions of the predetermined attributes and are represented by regions of attributes different from the predetermined attributes. The transportation route determination method according to any one of claims 1 to 5.

7. the transportation network is defined as a two-dimensional image of a grid pattern; The nodes of the transportation network are represented by squares of predetermined attributes containing predetermined colors, patterns, or marks, arranged alternately vertically and horizontally in the two-dimensional image, and the edges between the nodes are represented by squares of predetermined attributes that are adjacent to the squares of the predetermined attributes vertically or horizontally and have attributes different from the predetermined attributes. The transportation route determination method according to any one of claims 1 to 5.

8. The two-dimensional image of the transportation network is obtained by superimposing a departure image in which the squares corresponding to the nodes of the collection points of the goods are expressed with specific attributes, a destination image in which the squares corresponding to the nodes of the delivery points of the goods are expressed with the predetermined attributes, and a road image in which the squares corresponding to the edges are expressed with the different attributes. The transportation route determination method according to claim 7.

9. The transportation network includes bypass edges that are different from the edges that connect specific nodes. The transportation route determination method according to any one of claims 1 to 8.

10. the transportation network is defined by superimposing a plurality of two-dimensional images of geometric patterns; The nodes of the transportation network are represented by regions requiring predetermined attributes including predetermined colors, patterns, or marks arranged at predetermined intervals in any two directions of the geometric pattern in one two-dimensional image, and edges between the nodes are located between regions of the predetermined attributes and are represented by regions of attributes different from the predetermined attributes; The bypass edge is expressed as a region of a specific attribute between regions corresponding to specific nodes in another two-dimensional image. The transportation route determination method according to claim 9.

11. The computer a simulation process of repeating a selection process of stochastically selecting a node of a transportation base or a delivery point through which the transportation equipment will next pass from a transportation base or a collection point corresponding to a node of the created transportation network until the transportation of the goods to the delivery point is completed; A step of calculating a predetermined evaluation value for the transportation route obtained in the simulation step; and A step of determining a transportation route based on the predetermined evaluation amount from among a plurality of transportation routes obtained by repeating the simulation step and the calculation step of the predetermined evaluation amount within a predetermined range. The transportation route determination method according to any one of claims 1 to 10, comprising:

12. The computer a learning model that is trained to output vector data indicating a transportation route of the item in the transportation network when image data of the transportation network is input; When an image showing a collection point and a delivery point of the item is input, the transportation route is determined based on the vector data output from the learning model. The transportation route determination method according to claim 7 or 8.

13. A transportation network is created based on map data, in which collection and delivery points for a plurality of items, transportation bases through which transportation equipment for transporting the items can pass, and each point in the transportation network are assigned to nodes arranged in a geometric pattern in which a predetermined unit pattern is periodically repeated, and routes connecting adjacent points or transportation bases are assigned to edges between the nodes; A transportation route connecting the transportation bases of the transportation equipment directly or via each of the transportation bases is determined as a data string of identification data of the nodes in the transportation network. A transportation route determination device including a processing unit.

14. On the computer, A transportation network is created based on map data, in which collection and delivery points for a plurality of items, transportation bases through which transportation equipment for transporting the items can pass, and each point in the transportation network are assigned to nodes arranged in a geometric pattern in which a predetermined unit pattern is periodically repeated, and routes connecting adjacent points or transportation bases are assigned to edges between the nodes; A transportation route connecting the transportation bases of the transportation equipment directly or via each of the transportation bases is determined as a data string of identification data of the nodes in the transportation network. A computer program that executes a process.

Images