Delivery plan creation device

Abstract

To provide a delivery plan creation device capable of creating an efficient delivery plan in a system configuration in which a loading platform for loading and unloading cargo is determined depending on the position of a delivery vehicle relative to a cargo delivery port provided at a base.SOLUTION: An AGV 2 that constitutes the delivery control system is equipped with multiple loading platforms 28 and is configured to transfer cargo by stopping the vehicle so that the loading platform to be used for transferring (loading and unloading) the cargo is positioned in front of a cargo delivery port 51 provided at a base 3. The processing device in charge of creating the delivery plan determines the allocation of loading platforms to cargo so as to increase the frequency of simultaneous transfers based on capacity data indicating the number of loading platforms provided at the AGV 2 and origin / destination information (OD information) for each cargo. For example, the processing device reduces the loading platform allocation to a set packing problem or a two-dimensional packing problem to obtain an optimal solution or an approximate solution.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present disclosure relates to a delivery plan creation device that creates a delivery plan for packages using delivery vehicles. [Background technology]

[0002] Patent Document 1 discloses a delivery system that uses a vehicle (hereinafter also referred to as a delivery vehicle) as transportation equipment with multiple loading bays. In the configuration described in Patent Document 1, when the delivery vehicle reaches a stopping area for delivery of packages, the delivery vehicle remains stopped. A carry-out device is provided near the stopping area. The carry-out device has multiple arms whose positions relative to the longitudinal direction of the delivery vehicle can be changed, and by moving the arms, packages loaded on specific loading bays can be removed or packages can be loaded onto empty loading bays. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 6362240 [Non-patent literature]

[0004] [Non-Patent Document 1] Jacques Desrosiers, Yvan Dumas, and Francois Soumis “A Dynamic Programming Solution of the Large-Scale Single-Vehicle Dial-A-Ride Problem with Time Windows” American Journal of Mathematical and Management Sciences Vol. 6, February 1986. Summary of the Invention [Problem to be solved by the invention]

[0005] In a facility configuration where an arm that can access multiple loading bays and whose position relative to the delivery vehicle is variable is installed at a base, there are fewer restrictions on how packages can be loaded onto the multiple loading bays, and the multiple loading bays can be operated flexibly. However, introducing an arm that can freely access multiple loading bays as described above incurs costs.

[0006] Another possible facility configuration is one in which the delivery port at a base station can only access one loading platform located in front of the delivery port. In such an assumed configuration, the loading platform available for loading / unloading (hereinafter referred to as "loading and unloading") is determined by the stopping position of the delivery vehicle. If the loading platform located in front of the delivery port for shipping the package is occupied by another package, the delivery vehicle may need to move forward or backward to change its stopping position to pick up the package and switch the loading platform facing the delivery port. However, Patent Document 1 does not consider at all an efficient method for allocating loading platforms to packages when such an assumed configuration is assumed.

[0007] The present disclosure provides a delivery plan creation device capable of creating an efficient delivery plan in a system configuration in which a loading platform for loading and unloading cargo is determined by the position of a delivery vehicle relative to a delivery port. [Means for solving the problem]

[0008] One of the delivery plan creation devices disclosed herein is a delivery plan creation device that creates a delivery plan for at least one delivery vehicle (2) that has a plurality of loading bays (28) and is configured to be able to transfer cargo by stopping so that the loading bay to be transferred is located at a location corresponding to the delivery port at the base, and includes at least one calculation device (111), and the calculation device is configured to acquire route data indicating the route of the delivery vehicle, capacity data indicating the number of loading bays the delivery vehicle has, layout data indicating the number and locations of delivery ports at the base, and cargo data indicating the shipping location and delivery destination for each cargo, and to create a delivery plan including cargo bay allocation based on the route data, capacity data, layout data, and cargo data.

[0009] According to the delivery plan creation device, when creating a delivery plan including the allocation of cargo bays to cargoes, the layout of delivery ports at base locations and the shipping locations and delivery destinations of each cargo are taken into consideration, thereby enabling the creation of an efficient delivery plan.

[0010] Note that the symbols in parentheses in the claims indicate a correspondence with the specific means described in the embodiments described below as one aspect, and do not limit the technical scope of the present disclosure. [Brief explanation of the drawings]

[0011] [Figure 1] FIG. 1 is a diagram showing the overall configuration including a delivery control system. [Figure 2] FIG. 2 is a diagram for explaining the configuration of an AGV. [Figure 3] FIG. 10 is a diagram for explaining a configuration for transferring cargo. [Figure 4] FIG. 2 is a functional block diagram of the processing device. [Figure 5] FIG. 2 is a diagram illustrating an example of base data. [Figure 6] FIG. 10 is a diagram illustrating an example of AGV data. [Figure 7] FIG. 10 is a diagram illustrating an example of package data. [Figure 8] FIG. 10 is a diagram illustrating an example of a route candidate. [Figure 9] 10 is a flowchart showing the flow of processing executed by the processing device. [Figure 10] FIG. 10 is a diagram for explaining a configuration in which allocation of bins to assigned packages is handled as a two-dimensional packing problem. [Figure 11] FIG. 1 is a diagram showing an example of a mathematical model that formulates a two-dimensional packing problem corresponding to bin allocation. [Figure 12] FIG. 10 is a diagram showing an example of loading platform allocation when priority is given to loading efficiency. [Figure 13] FIG. 13 is a diagram for explaining a transfer action in the allocation example shown in FIG. [Figure 14] 11 is a diagram for explaining a transfer action in the allocation example shown in FIG. 10. FIG. [Figure 15] FIG. 10 is a functional block diagram of a processing device according to a second embodiment. [Figure 16] FIG. 18 is a diagram showing combinations of executable plans under the preconditions shown in FIG. 17. [Figure 17] FIG. 10 is a diagram illustrating a situation in which two AGVs are available when there is a demand to deliver two packages from a first base to a second base. [Figure 18] FIG. 17 is a diagram showing a matrix corresponding to the combination of executable plans shown in FIG. 16. [Figure 19] FIG. 10 is a diagram illustrating an example of a cost vector. [Figure 20] FIG. 1 is a diagram illustrating an example of a mathematical model that formulates a set-packing problem corresponding to bin allocation. [Figure 21] FIG. 10 is a functional block diagram of a processing device according to a third embodiment. [Figure 22] FIG. 1 illustrates an example of a mathematical model for routing and load allocation. [Figure 23] FIG. 10 is a functional block diagram of a processing device according to a fourth embodiment. [Figure 24] FIG. 11 is a functional block diagram of a processing device according to a fifth embodiment. [Figure 25] 10 is a flowchart showing an example of operation of the processing apparatus according to the fifth embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. The present disclosure is not limited to the following embodiments. The configurations disclosed below may be modified in various ways without departing from the spirit of the present disclosure. Various modified examples may be appropriately combined as long as no technical contradictions arise. The present disclosure also includes configurations that are not explicitly stated and are formed by combining multiple modified examples. In the following description, components having the same function may be given the same reference numerals, and specific descriptions thereof may be omitted. Furthermore, components having the same function may be given the same or similar names, and specific descriptions thereof may be omitted. When only a portion of a configuration is mentioned, descriptions given elsewhere may apply to other parts.

[0013] [First embodiment] The delivery control system 1 shown in FIG. 1 is a system that creates a delivery plan for multiple AGVs (Automatic Guided Vehicles) 2 used within a specified area. In the following, a location installed within the area where delivery of goods takes place will be referred to as a base 3. The base 3 may actually be a facility for delivery of goods. The base 3 may also be referred to as a base facility.

[0014] The AGV2 is an unmanned delivery vehicle or unmanned delivery robot used to deliver packages between bases 3. The area in which the AGV2 is used may be within the premises of a facility such as a factory, warehouse, or logistics center. Paths are formed between the bases 3 for the AGV2 to travel through. Each of the multiple AGV2s may be assigned an AGV number, which is a unique identification number. The multiple AGV2s may be distinguished by their AGV numbers. Each of the multiple bases 3 may also be assigned a base number, which is a unique identification number.

[0015] The following description will be given taking as an example a case where the AGV 2 is used in a factory. The parcels to be delivered may be parts (so-called workpieces) used in the manufacture of an item, or may be finished products. The parcels may also be a collection of multiple workpieces. For example, the parcels may be cases containing multiple workpieces. Of course, the parcels may also be cardboard boxes, etc.

[0016] <Overall structure> The delivery control system 1 is configured to be capable of wireless communication with the AGVs 2. The communication method between the delivery control system 1 and the AGVs 2 may be any method, such as Zigbee (registered trademark), Wi-Fi (registered trademark), Bluetooth (registered trademark), EnOcean (registered trademark), or Cellular V2X (PC5). The delivery control system 1 is also configured to be capable of wireless communication with each base 3. The communication method with the base 3 may be the same as or different from the communication method with the AGVs 2.

[0017] The delivery control system 1 may be connected to some of the bases 3 via a wired communication network installed in the factory. The wired communication network may be Ethernet or the like. Of course, the delivery control system 1 may be connected to all of the bases 3 via a wired communication network. The communication method with the delivery control system 1 may differ depending on the base 3.

[0018] The delivery control system 1 includes a processing device 11, an input device 12, a display 13, a communication device 14, and a database 15. The processing device 11 is connected to the input device 12, the display 13, the communication device 14, and the database 15.

[0019] The processing device 11 is configured to include a computer. The processing device 11 corresponds to a delivery plan creation device. The processing device 11 includes a processor 111, a memory 112, a storage 113, an input / output circuit 114, and the like. The processor 111 is an arithmetic core such as a CPU (Central Processing Unit). The processing device 11 may include multiple processors 111. The processor 111 corresponds to an arithmetic device. The memory 112 is a volatile memory such as a RAM (Random Access Memory). The storage 113 is a rewritable non-volatile memory such as a flash memory. The storage 113 may include a non-rewritable recording medium (ROM: Read Only Memory). The storage 113 stores a program including instructions for causing a computer to function as the processing device 11 of the present disclosure. The input / output circuit 114 is hardware that serves as an interface for the processing device 11 to communicate with the input device 12, the display 13, and the like. IO in FIG. 1 stands for input / output (Input Output). The input / output circuit 114 may include a PHY chip or the like depending on the communication method.

[0020] The processing device 11 is a device that executes calculation processing for creating a delivery plan for the AGV 2 based on data input from the input device 12 and the communication device 14, and data stored in the database 15. Details of the functions of the processing device 11 will be described separately later.

[0021] The input device 12 is a device that allows the processing device 11 to receive operations from an administrator. The administrator here may be a person who manages the factory where the AGV 2 is used, or an employee of the factory. The input device 12 outputs a signal (hereinafter also referred to as an operation signal) in response to an operation by the administrator to the processing device 11. Based on the operation signal input from the input device 12, the processing device 11 can import data stored in the database 15, rewrite data, create a delivery plan, and so on. The input device 12 may include a mouse and a keyboard. The input device 12 may also be a touch panel layered on the screen of the display 13. The input device 12 may also be a hardware switch installed on the display 13 or in the factory.

[0022] The display 13 is a device that displays an image corresponding to the video signal input from the processing device 11. The display 13 may be a liquid crystal display, an organic EL display, or the like. The processing device 11 displays the progress of the calculation process for creating a delivery plan and the calculation results on the display 13. The display 13 and the input device 12 may be integrally formed as a tablet terminal.

[0023] The communication device 14 is a device for communicating with the AGV 2 and the base 3. The communication device 14 may include a wireless communication module compatible with the method used for communication with the AGV 2. The communication device 14 may also be equipped with a communication module for wired communication compatible with a wired communication network installed in the factory. The communication device 14 inputs data received from the AGV 2 and the base 3 into the processing device 11. The communication device 14 also transmits the data input from the processing device 11 to a destination specified by the processing device 11. For example, the communication device 14 transmits data indicating a delivery plan to the AGV 2 based on input from the processing device 11.

[0024] The database 15 is a recording device that stores data necessary for creating a delivery plan. The data necessary for creating a delivery plan includes premise data, which is data indicating prerequisites (in other words, constraints) for allocating loading bays, which will be described later. Details of the premise data, etc. will be described separately later.

[0025] The AGV 2 includes an external sensor 21, a communication device 22, a display unit 23, and a control unit 24. The control unit 24 is connected to be able to communicate with the external sensor 21, the communication device 22, and the display unit 23. In addition to the above components, the AGV 2 also includes components necessary for a delivery vehicle, such as wheels 25, a travel motor 26, a frame 27, a loading platform 28, and a steering device (not shown), as shown in FIG.

[0026] The external sensor 21 is a sensor that detects objects around the AGV 2. The detection results of the external sensor 21 are used to control the travel of the AGV 2, such as stopping the AGV 2 when an obstacle is detected or stopping the AGV 2 at a predetermined stopping position. The external sensor 21 may be a camera, millimeter-wave radar, sonar, LiDAR, or the like. The external sensor 21 may also be a device that reads magnetic markers installed on the factory floor.

[0027] The communication device 22 is a device for wireless communication with the delivery control system 1. The communication device 22 provides data received from the delivery control system 1 (e.g., delivery plan data) to the control unit 24. The communication device 22 also transmits data input from the control unit 24 to the delivery control system 1. The AGV 2 may be configured to be able to perform wireless communication with other AGVs 2 or the base 3. By communicating with other AGVs 2, processing related to the transfer / acquisition of passage rights through an aisle may be performed.

[0028] The display unit 23 is a display attached to the AGV 2. The display unit 23 may be a liquid crystal display or the like. The display unit 23 is a device that displays an image corresponding to a video signal input from the control unit 24. The display unit 23 is an optional element and may be omitted.

[0029] The control unit 24 is a device that controls the travel of the AGV 2. The control unit 24 moves the AGV 2 and operates the rollers (described in detail later) of the loading platform 28 in accordance with delivery plan data received from the delivery control system 1 using the communication device 22. The control unit 24 may also recognize the status of the AGV 2 based on signals from the external sensor 21 or other on-board sensors (not shown). The status of the AGV 2 may include the position of the AGV 2, the load status of the cargo, the status of the power source, or the presence or absence of a malfunction. The control unit 24 may use the communication device 22 to transmit AGV status information, which is information regarding the status of the AGV 2, to the delivery control system 1. The transmission of the AGV status information may be performed periodically or in response to a predetermined event (e.g., departure). The transmitted data may include an AGV number as information indicating the sender.

[0030] The control unit 24 may also display an image on the display unit 23 indicating the load status of the luggage or whether or not there is a malfunction. Displaying an image on the display unit 23 may be a process for notifying a manager or other employees of AGV status information. The image indicating the load status of the luggage may be an image indicating the usage status (empty / occupied) of each loading platform 28. The control unit 24 may also display an image on the display unit 23 notifying or predicting the current or upcoming behavior of the AGV 2, such as that the AGV 2 is scheduled to stop soon, that the luggage is being unloaded, that the AGV 2 is scheduled to depart soon, or that the AGV 2 is cruising. Note that since the display unit 23 is an optional element, the display control by the control unit 24 is also an optional element.

[0031] The travel motor 26 is a motor for rotating the wheels 25. The AGV 2 may be equipped with a travel motor 26 for the front wheels and a travel motor 26 for the rear wheels. The frame 27 is a support member that integrally supports components such as the wheels 25, the travel motor 26, and the loading platform 28.

[0032] The loading platform 28 is a platform on which cargo is placed. In the AGV2, the loading platform 28 may also be called a frontage. The AGV2 is equipped with multiple loading platforms 28. The multiple loading platforms 28 equipped on the AGV2 may be identified by numbers. Here, as an example, the number for identifying the loading platform 28 is also referred to as the loading platform number. The loading platform number may be referred to as the loading platform identifier. The loading platform number may be expressed as numbers, letters, or a combination of numbers and letters. The loading platform number may be set so that there is no duplication (i.e., it is unique) among the loading platforms 28 equipped on one AGV2.

[0033] For convenience of explanation, the AGV 2 is set to have a front and a back, and the leading platform 28 is also referred to as the first platform 28. However, in other embodiments, the AGV 2 may be configured to be able to move in either direction equally without distinction between front and back. Platform numbers may be assigned in accordance with a predetermined rule.

[0034] 1, the base 3 includes a communication device 31 and a base management device 32. The communication device 31 is connected to the base management device 32 so as to be able to communicate with each other. Furthermore, the base 3 includes a plurality of delivery ports 51 as shown in FIG.

[0035] The communication device 31 is a device for wireless communication with the delivery control system 1. The communication device 31 receives collection and delivery plan data from the delivery control system 1 and provides it to the control unit 24. The collection and delivery plan data may include information on the scheduled arrival time of the AGV 2. The communication device 31 also transmits data input from the base management device 32 to the delivery control system 1. The base 3 may be configured to be able to perform wireless communication with the AGV 2 or other bases 3. The base 3 may communicate with the AGV 2 that is scheduled to arrive at its base within a certain time, thereby sharing the order in which cargo will be loaded and unloaded.

[0036] The base management device 32 is a computer installed at the base 3. The base management device 32 may send a request message related to the dispatch or reception of a package to the delivery control system 1 using the communication device 31. The request message related to the dispatch of a package may include information about the destination base 3 (for example, the base number). A request related to the reception of a package may be a request for parts replenishment, etc. In the present disclosure, a request to the delivery control system 1 related to the dispatch or reception of a package is also referred to as an order. The base management device 32 may control the operation of a roller motor arranged at the delivery port 51 based on delivery plan data received from the delivery control system 1. The delivery port 51 will be described later.

[0037] <Configuration related to loading and unloading of luggage> As shown in FIGS. 2 and 3, the AGV 2 is configured as a delivery vehicle having multiple wheels 25. The multiple carriers 28 are arranged in a row along the longitudinal direction of the AGV 2. FIG. 2 illustrates, as an example, a case in which the AGV 2 has three carriers 28. 28a shown in FIGS. 2 and 3 represents the leading (i.e., first) carrier 28, 28b represents the second (i.e., last) carrier 28, and 28c represents the third (i.e., last) carrier 28. The number of carriers 28 provided on the AGV 2 may be any value, such as 2, 4, or 6. The number of carriers 28 may differ for each AGV 2. In this embodiment, for simplicity of explanation, the number of carriers 28 provided on the AGV 2 is assumed to be the same (e.g., 3). The multiple carriers 28 are arranged at a predetermined distance from each other. The multiple carriers 28 may be separated by walls or frames.

[0038] The installation intervals of the loading platforms 28 on the AGV 2 may be the same as the placement intervals of the delivery openings 51 at the base 3. Furthermore, the width of one loading platform 28 may be the same as the width of the delivery opening 51. The height of the loading platform 28 relative to the road surface may be set to the same height as the delivery opening 51. Here, "same" does not necessarily mean completely the same, but may also mean approximately the same.

[0039] 3, the AGV 2 and the base 3 are configured to automatically load and unload cargo when the AGV 2 stops at a predetermined position in front of the delivery gate 51 at the base 3. For example, the loading platform 28 may include a plurality of rollers 45 and a roller motor that rotates the rollers. The rotation axes of the rollers 45 are set parallel to the longitudinal direction of the AGV 2, and as the rollers 45 rotate in the same direction, the cargo placed on the loading platform 28 moves in the width direction of the AGV 2.

[0040] When the rotation direction of the roller motor is reversed, the rotation direction of the rollers 45 is reversed. Accordingly, the movement direction of the cargo is also reversed. In other words, the AGV 2 is configured to be able to send (drop off) cargo to the delivery port 51 of the base 3 and receive (pick up) cargo from the base 3 by switching the rotation direction of the roller motor. In this disclosure, pickup and drop off of cargo, i.e., loading and unloading of cargo, are also referred to as transfer or transfer action. A roller motor may be installed for each loading platform 28. The control unit 24 of the AGV 2 may be configured to be able to control the rotation direction of the rollers 45 independently for each loading platform 28.

[0041] The delivery port 51 of the base station 3 may also be designed to correspond to the configuration of the loading platform 28 of the AGV 2 described above. The delivery port 51 may be configured to be able to send and receive packages using rollers and roller motors, similar to the loading platform 28. While FIG. 3 illustrates a pattern in which the base station 3 has only two delivery ports 51, the number of delivery ports 51 provided at the base station 3 is not limited to two. The number of delivery ports 51 provided at the base station 3 may be three or more, or may be only one. The number of delivery ports 51 installed at each base station 3 may differ. A small-scale base station 3 may have three to four delivery ports 51. Furthermore, a large-scale base station 3 may have 20 or more delivery ports 51. A large-scale base station 3 may be a base station 3 where packages produced at small-scale base stations 3 are collected.

[0042] The base station 3 and the AGV 2 are configured to enable simultaneous transfer. Simultaneous transfer refers to loading and unloading of cargo onto multiple loading bins 28 in parallel (at the same time). Possible simultaneous transfer patterns include a pattern in which multiple cargoes are loaded onto different loading bins 28 at the same time, a pattern in which cargoes from multiple loading bins 28 are dropped off simultaneously, and a pattern in which pickup and drop-off are mixed. FIG. 3 shows a pattern in which an action to unload cargo from the second loading bin 28b and an action to load cargo onto the third loading bin 28c are executed simultaneously (i.e., a mixed pattern). Simultaneous transfer is the key to efficient delivery, and the processing device of this embodiment allocates cargo to loading bins 28 so that simultaneous transfer can be performed as much as possible. That is, the processing device 11 creates an efficient delivery plan that includes simultaneous transfer from multiple loading bins 28.

[0043] In a production facility, the line through which parts flow and the line through which finished products flow are often also fixed. In light of this, in this embodiment, as an example, the roles (sending / receiving) of the multiple delivery ports 51 provided at the base 3 are assumed to be fixed in relation to the production line. For convenience, the delivery port 51 used to receive delivery is also referred to as the delivery port 51a, and the delivery port 51 used to send delivery is also referred to as the delivery port 51b. The number of delivery ports 51a and the number of delivery ports 51b, as well as their arrangement order, may differ for each base.

[0044] <Processing device functions> 4, the processing device 11 includes a condition registration unit F1a, a reading unit F2a, and an allocation unit F3a. All or part of these components may be realized by a computer that constitutes the processing device 11 executing a program. The processing device 11 generates delivery plan data and transmits it to the AGV 2.

[0045] The delivery plan data here may include route data, data on stopping positions (hereinafter referred to as stopping positions) for loading and unloading luggage, information on the target platform which is the loading platform to be moved at the stopping position, and the action type of the target platform. The action type of the target platform is receiving (pickup) the luggage or delivering (drop-off) the luggage. The stopping position is uniquely determined according to the number of the loading platform used for loading and unloading the luggage and the location of the delivery gate. The delivery plan data may include data such as where to go, at which base, which loading platform to use, and what transfer action to perform at which receiving gate.

[0046] The condition registration unit F1A is configured to register data (also called premise data) required to create a delivery plan in the database 15, and the reading unit F2a and allocation unit F3a essentially correspond to the configuration for creating a delivery plan.

[0047] The premise data may include base data, AGV data, parcel data, route candidate data, and cost setting data. The base data is data related to the base 3. The base data includes delivery port data for each base 3. The delivery port data is data indicating the layout of delivery ports at the base 3. The delivery port data may include data related to the base number, the number of delivery ports, the receiving port, the shipping port, and the stop position, as shown in FIG. 5, for example. In FIG. 5, the base 3 with the base number "001" has six delivery ports, with delivery ports 4 to 6 being for receiving packages and delivery ports 1 to 3 being for shipping packages. The stop position may be the stop position of the AGV 2 when the loading platform No. 1 is positioned in front of delivery port No. 1. The stop position data may include the stop position for each delivery port. The items included in the delivery port data are not limited to the combination shown in FIG. 5. The delivery port data may include position information for each delivery port. The delivery port data corresponds to the layout data.

[0048] The AGV data is data related to the AGVs 2. The AGV data includes capacity data, which is a data set indicating the number of platters equipped on each AGV 2. The capacity data may be data indicating the number of platters equipped on each AGV 2 in list or matrix format, as shown in Figure 6, for example. The number of platters corresponds to the number of packages that can be carried when fully loaded. The AGV data may also include management data such as the current position of each AGV 2 and its operating status (normal / abnormal).

[0049] As shown in FIG. 7, the parcel data may be a dataset indicating the origin and destination of each parcel. Here, the origin refers to the base 3 from which the parcel is shipped. The origin may be alternatively referred to as the requester, pickup base, shipping location, etc. The destination refers to the base 3 to which the parcel is delivered. The destination may be alternatively referred to as the destination or drop-off base. In this disclosure, information indicating the combination of the origin and destination of a parcel is also referred to as OD (Origin-Destination) information. N1, N2, N3, etc. in the figure represent bases 3. In addition to the OD information, the parcel data may also include the number of the delivery port. If the delivery port and the delivery port for the parcel are fixed, the origin and destination may be expressed at a finer granularity of the delivery port rather than at the base unit. Such a configuration may enable more efficient allocation of pallets taking into account the location of the delivery port.

[0050] Such parcel data may be generated based on assumptions. For example, the parcel data may be generated based on past performance. The parcel data may also be expressed in the form of an array, a matrix, or the like. In one embodiment, the parcel data may be generated based on a request (so-called order) for parcel delivery received from the base 3.

[0051] The route candidate data is data indicating routes that the AGV 2 can take (hereinafter referred to as route candidate). The multiple route candidates differ in at least one of the combination of bases 3 that the AGV 2 passes through and the order in which they pass through. The route candidate may be determined according to the layout of the area in which the AGV 2 is used (here, a factory). The route candidate may be set based on topology data in which paths that the AGV 2 can travel are defined as links and bases and intersections are defined as nodes.

[0052] FIG. 8 shows an example of a route candidate. The first route, indicated by a dashed line in FIG. 8, is a route that passes through N1, N2, N3, N4, and N5. The second route, indicated by a two-dot dashed line in FIG. 8, is a route candidate that passes through N1, N2, N3, N4, N6, N7, and N8. The solid arrows in the figure indicate paths that AGV2 can travel. In FIG. 8, the circles on the solid arrows indicate bases 3 on the route. The stars in the figure indicate home positions. The home position is the location (i.e., the base) from which AGV2 departs and returns. The home position is sometimes called a depot. The home position may be a location where charging equipment is installed. Home positions may be located in multiple locations. Note that for convenience, only two types of routes are shown in FIG. 8, but three or more route candidates may be set. The set of route candidates may be different for each AGV2.

[0053] The cost setting data is data of cost parameters used in the allocation calculation. The costs used in the allocation calculation include travel cost and transfer cost. The travel cost is the cost required to travel between bases 3. The travel cost may be set for each candidate route. The travel cost may be set for each link. The travel cost may be set to a larger value the longer the travel distance. In cases where congestion occurs, the travel cost may be set to a larger value depending on the situation. The travel cost setting value may be dynamically changed based on reports from AGV2 or the output of sensors installed in the aisles. The travel cost setting data corresponds to travel cost data. Furthermore, data indicating the transfer cost setting value corresponds to transfer cost data.

[0054] The transfer cost is a parameter that represents the cost incurred when picking up / dropping off a load (i.e., a transfer action) once. As shown in Fig. 3, when multiple transfers are performed simultaneously, the processing device 11 may be programmed to apply the transfer cost for one transfer. In other words, one simultaneous transfer may be counted as the cost of one transfer action.

[0055] Additionally, the cost data may include a cost setting value for small movements for position adjustment within a base (hereinafter, "intra-base movement"). Intra-base movement means, for example, stopping the first platform so that it is positioned in front of a first delivery gate at a certain base to drop off a package, and then moving forward / backward so that the first platform is positioned in front of another delivery gate at the same base to pick up the package. Given the configuration of the AGV2 disclosed herein, it should be noted that such intra-base movement may occur depending on the platform allocation. In one embodiment, the processing device 11 performs platform allocation so as to reduce the frequency of intra-base movement, as will be described separately below.

[0056] Some of the premise data (in other words, prerequisites) described above may be registered by the administrator via the input device 12. The condition registration unit F1A may generate some or all of the premise data described above based on an operation signal input from the input device 12, and store it in the database 15.

[0057] Furthermore, the condition registration unit F1A may acquire part of the premise data through communication with the base 3 or communication with the AGV 2 and register it in the database 15. For example, the condition registration unit F1A may generate or update the parcel data based on an order received from the base 3. The condition registration unit F1A may update the capacity data based on receiving a notification from an AGV 2 that one of the loading platforms is soiled or broken. The condition registration unit F1A may update the capacity data so as not to use an unusable loading platform.

[0058] The reading unit F2a is configured to read various data registered in the database 15. The reading unit F2a executes reading of premise data based on input of a predetermined start trigger. Reading of premise data corresponds to the initial step of creating a delivery plan. The start trigger may be interpreted as a trigger that starts creating a delivery plan. The start trigger may be designed as appropriate. For example, the start trigger may be a predetermined operation signal input from the input device 12. Creation of a delivery plan may be executed according to a predetermined schedule. The start trigger may be the passage of a certain amount of time since the previous creation or the arrival of a specific time. The processing device 11 may execute creation of a delivery plan based on switching the power from off to on. The start trigger may be power on. The start trigger may be the accumulation of a predetermined number of orders from the base 3 related to package shipping / parts supply. The data read by the reading unit F2a is temporarily stored in the memory 112 and referenced by the allocation unit F3a, etc.

[0059] The allocation unit F3a is configured to determine the route and assigned load for each AGV2, and allocate a loading platform for each assigned load. The assigned load for a certain AGV2 is the load that the AGV2 is responsible for delivering (specifically, picking up and dropping off). The loading platform allocation may be the allocation of a loading platform to be used for delivering the load for the assigned load.

[0060] The allocation unit F3a includes a route setting unit F31a and a provisional allocation unit F32a as sub-functions for performing loading platform allocation, etc. The route setting unit F31a is configured to select a route for each AGV 2 from route candidates in accordance with predetermined route setting rules. For example, the route setting unit F31a may be configured to select the route with the shortest route length among the route candidates. Route setting is performed for each AGV 2. To ensure route variation, the combination of route candidates for each AGV 2 may be different. Furthermore, the route setting unit F31a may be programmed so that a route set for a previous AGV 2 is not assigned to a subsequent AGV 2. An upper limit may be set on the number of AGVs 2 that can be assigned to one route candidate. The route setting unit F31a generates data indicating the determined route for each AGV 2.

[0061] The temporary allocation unit F32a is configured to temporarily allocate assigned loads to each AGV2. The operation of the temporary allocation unit F32a will be described in detail later. The allocation unit F3a determines whether the temporarily allocated loads can be loaded and performs processing to determine the allocation of loading pallets to be used to transport each load. Among the loads provisionally allocated by the temporary allocation unit F32a, the allocation unit F3a excludes loads that are determined to be unhandlable (in other words, unloadable) from the loads assigned to the target AGV. The target AGV is the AGV2 that is being focused on as the target of processing such as loading pallet allocation. The loads assigned to each AGV2 are finally determined based on the results of the loading pallet allocation processing.

[0062] As described above, the processing device 11 (allocation unit F3a) of this embodiment creates an operation plan for each AGV 2 in the following order: route setting, provisional allocation of assigned cargo, determination of feasibility of delivery of the provisionally allocated cargo, and determination of assigned cargo and loading platform allocation. Figure 9 is a flowchart showing an outline of the operation of the processing device 11 in this embodiment. This flowchart is executed for each AGV 2. Below, the AGV 2 to be processed is also referred to as the target AGV. The AGV 2 to be processed may be selected in a predetermined order. The following steps are executed by the processing device 11. The processing device 11 as the entity executing the processing may be replaced with the processor 111 or the allocation unit F3a as appropriate.

[0063] S101 is a step in which the processing device 11 selects a target AGV. After selecting the target AGV, the processing device 11 executes S102. S102 is a step in which the processing device 11, functioning as the route setting unit F31a, sets a route for the target AGV. In S102, the processing device 11 selects one route from among route candidates for the target AGV based on a predetermined route selection rule (e.g., distance priority).

[0064] In step S103, the processing device 11 as the provisional allocation unit F32a extracts candidates for the packages to be handled by the target AGV by comparing the OD information for each package indicated in the package data with the route of the target AGV. The processing device 11 extracts packages that can be picked up and dropped off on the route of the target AGV as candidates for the packages to be handled.

[0065] For example, the provisional assignment of assigned packages will be described here assuming the packages shown in FIG. 7 and the route of the target AGV is set to the first route shown in FIG. 8. As can be seen by comparing the package data shown in FIG. 7 with the bases passed through on the first route shown in FIG. 8, for packages with package numbers 1 to 3 and 5 to 6, the origin and destination appear on the first route in that order. Therefore, the processing device 11 provisionally assigns the packages with package numbers 1 to 3 and 5 to 6 as assigned packages for the target AGV. The provisionally assigned packages correspond to assigned candidate packages. On the other hand, for the package with package number 4, the origin (base N3) is on the first route, but the destination (base N7) is not on the first route. Therefore, the processing device 11 does not provisionally assign the package with package number 4 as the assigned package for the target AGV. On the route set in this way, packages whose origin and destination appear in that order can be candidates for assigned packages. After extracting the assigned package candidates based on the route data and package data, the processing device 11 executes S104.

[0066] In step S104, the processing device 11 determines the allocation of platters to be used for delivery of each AGV based on the OD information of the candidate platters. As shown in FIG. 10, the processing device of this embodiment solves a two-dimensional packing problem in which the number of platters (i.e., capacity) of the target AGV is considered to be the height (= depth) of the base, the route length is considered to be the width of the base, and the area from the departure point to the destination is considered to be the occupied area of ​​the item. This determines whether each platter can be handled and which platter to use for each platter. The term "item" here refers to the platter to be packed. This also applies to items that are assumed to be transferred simultaneously. Whether a single platter that is not being transferred simultaneously can be loaded without exceeding the number of platters (capacity) can be determined by counting the number of platters, taking into account the timing of loading and unloading, and determining whether the number of platters (capacity) is exceeded. The lifetime of an item is subject to the constraints of the origin and destination. For example, a platter with platter number 1 (L1) is placed so that it occupies the area from N1 as the departure point to N3 as the destination.

[0067] In this packing problem, we consider a route r, a package i delivered by route r, and a set M r is given. M rLet Q be the set of pairs of parcels with overlapping delivery periods in r , and let y be an integer variable that represents the loading bay to which luggage i is assigned. i , the maximum number of carriers that AGV2 has is p max Then, the packing problem is formulated as shown in Figure 11. Note that dii' indicates the distance between the entrances when loading luggage i and luggage i' simultaneously.

[0068] Equation (1a) in Figure 11 is a setting that specifies that a solution that satisfies the constraints needs to be found, and means that there is no need to maximize or minimize the value of the objective function. Equation (2a) means that there is no overlap in the periods when two packages are assigned to a certain opening. Equation (3a) means that given simultaneous transfers are assigned while maintaining the distance dii' between packages. Equation (4a) is a constraint that, when multiple packages are assigned to one AGV2 on the same line (base), the opening number to which the next package i' is assigned is set to be larger than the opening number of the AGV2 to which the first package i is assigned, allowing the AGV2 to transfer packages sequentially without backing up.

[0069] When allocating a loading platform, the processing device 11 may determine, for each package, whether there is a loading platform available for delivery of the package. If the processing device 11 cannot find a loading platform available for delivery of the package to be determined from the viewpoint of loading platform capacity, the processing device 11 may decide to remove the package from the packages handled by the target AGV and have another AGV 2 deliver the package.

[0070] Furthermore, when allocating loading bays, the processing device 11 may determine loading bay allocation so as to satisfy the simultaneous transfer constraint for cargo that has a constraint on simultaneous transfer. A cargo that has a constraint on simultaneous transfer is a cargo that should be loaded and unloaded at the same time as other cargo. The presence or absence of a constraint on simultaneous transfer may be registered in advance for each cargo. For example, the administrator may use the input device 12 to register whether or not simultaneous transfer is required for each cargo. The cargo data may include data regarding whether or not simultaneous transfer is required. A cargo that requires simultaneous transfer may be linked to data on cargo that should be loaded and unloaded at the same time as that cargo.

[0071] The processing device 11 may be programmed to allocate bins preferentially to bins that have simultaneous transfer constraints among candidate bins, and then allocate bins for bins that do not have simultaneous transfer constraints. For example, the processing device 11 may be configured to solve a packing problem for bins that have simultaneous transfer constraints, and then solve a packing problem for bins that do not have simultaneous transfer constraints, while inheriting the allocation result.

[0072] Incidentally, when solving the bin allocation for the luggage (items) shown in Figure 10 as a packing problem, it may appear at first glance that the bin allocation shown in Figure 12 has a higher bin utilization efficiency than the bin allocation shown in Figure 10. However, although the bin allocation shown in Figure 12 increases the bin utilization efficiency itself, it decreases the implementation efficiency of transfers. This is because when attempting to use one bin continuously, slight movements (i.e., movements within the bin) are required to adjust the position within the hub.

[0073] Specifically, the movement of AGV2 at base N2 is shown in Figure 13, taking the allocation shown in Figure 12 as an example. When the allocation shown in Figure 12 is adopted, AGV2 unloads package L2 from the first platform at base N2 and then places package L3 on the first platform. However, at base N3, the delivery entrances used for receiving package L2 and sending package L3 are different. AGV2 must first stop where the first platform is in front of receiving entrance 51a, and then move forward until the first platform is located directly in front of sending entrance 51b. Continuously using one platform in this way requires movement to adjust its position within a single base, which can reduce efficiency. In other words, the number of stops and starts increases.

[0074] In view of the above, the processing device 11 may be configured so that one loading platform is not used continuously. The processing device 11 may be configured to solve a problem with a constraint that a loading platform immediately after drop-off is not used to pick up other packages. For example, items handled in a packing problem may be treated as items that virtually occupy a loading platform up to the next base from the actual destination. For example, package L2 may be treated as an item that occupies a loading platform up to base N3, which is a transit base one base beyond base N2, the actual delivery destination. This makes it possible to avoid allocations such as those shown in FIG. 12. As a result, the frequency of simultaneous transfers may increase, and delivery efficiency may also improve.

[0075] According to the loading platform allocation mode shown in FIG. 10, the loading platform for dropping off the package L2 and the loading platform for receiving the package L3 are separately allocated, and these are compatible with the layout of the delivery gates at the base 3. As a result, as shown in FIG. 14, drop-off from the second loading platform and pickup to the first loading platform can be performed simultaneously. The loading platform used for pickup may be determined in consideration of the arrangement order of the receiving gate 51a and the sending gate 51b at the base 3 where the package is to be dropped. If the receiving gate 51a is located on the entrance side (upstream side) of the base 3, it is preferable to place the package on a loading platform closer to the rear. If the receiving gate 51a is located on the exit side (downstream side) of the base 3, it is preferable to place the package on a loading platform closer to the front. According to this policy, when the AGV 2 stops to drop off the package, the number of loading platforms facing the group of delivery gates at the base 3 can be increased, which is expected to make it easier to simultaneously drop off and pick up packages.

[0076] Of course, the above-described platform allocation policy is merely an example and is not limiting. Any constraints may be set in the mathematical model for platform allocation according to the allocation policy. The mathematical model used for platform allocation is not limited to the model shown in FIG. 11 and may be modified as appropriate.

[0077] After determining the assigned loads and their pallet allocation for one AGV2, the processing device 11 may execute S101 to S104 for the AGV2 to which the loads have not yet been assigned. In this case, the loads that have been assigned an AGV2 responsible for delivery among the list of loads indicated in the load data may be deleted from the load list. Using the loads to which no assigned vehicle has been assigned as the parent population, the processing device 11 may perform provisional assignment of the loads to the target AGV, determination of whether the load can be delivered, and pallet allocation. The assigned vehicle for a certain load is the AGV2 responsible for delivering that load. By repeating this process, the assigned vehicle for each load, the pallet to be used for delivery, etc. are finally determined. In the repeated process, the target AGV in the first round of processing corresponds to the first vehicle, and the target AGV in the second round and subsequent rounds of processing corresponds to the second vehicle.

[0078] When setting the route for the second AGV 2 (S101), the processing device 11 may be configured to preferentially select a route that can handle a parcel that the first AGV 2 is unable to handle. For example, if the first AGV 2 selects the first route, the parcel with parcel number 4 shown in FIG. 7 will remain unassigned to a delivery vehicle. In this case, the route for the second AGV 2 may be set to a route that can deliver the parcel with parcel number 4, i.e., the second route. In this way, the processing device 11 may be configured to select a route that can deliver a parcel that has not been assigned to a delivery vehicle in S101 for the second and subsequent AGVs. This reduces the occurrence of parcels that are not delivered to any AGV 2.

[0079] When the processing device 11 completes route selection for each AGV2, selection of assigned cargo, and allocation of loading bins, it transmits delivery plan data including these to the AGV2. The delivery plan data transmitted to the AGV2 may be delivery plan data for the AGV2. The AGV2 starts operating in accordance with the delivery plan data received from the processing device 11. As mentioned above, the delivery plan data may include instruction data regarding a stopping position for a transfer action. The stopping position may be identified from the loading bin number to be used for transferring the cargo and the position data of the delivery gate to be used for transferring the cargo. The transmission signal for the delivery plan data corresponds to the instruction signal.

[0080] The processing device 11 does not necessarily need to create a delivery plan that uses all available AGVs 2. For example, if all parcels can be delivered by operating five of nine AGVs 2, the processing device 11 may create and instruct a delivery plan for those five AGVs 2. The remaining four AGVs may wait at their home positions.

[0081] <Effects> Generally, base 3 as a production facility or the like may have restrictions such as the number of delivery ports and the purpose of each delivery port (for shipping / receiving). In light of such circumstances, depending on how the loading platform is used, a delivery vehicle may need to move back and forth multiple times at a single facility to load and unload packages. However, by devising a loading platform allocation for each package, it may be possible to reduce the frequency with which the delivery vehicle needs to change position for loading and unloading, thereby enabling more efficient deliveries.

[0082] This embodiment was created based on the above idea, and the processing device 11 determines the route and the assigned luggage taking into consideration the order of pickup and drop-off for each luggage. Furthermore, the processing device 11 determines the loading platform allocation so that there is no overlap in time or space. In other words, the processing device 11 determines the loading platform to be used to deliver the assigned luggage so that one loading platform is responsible for only one luggage at a time. Furthermore, when allocating loading platforms, the processing device 11 determines the number of stops to be reduced and simultaneous transfers are actively carried out. This can improve delivery efficiency. Specifically, the relative positions of the loading platforms and the delivery entrances can reduce the frequency of unnecessary movements and stops and starts. Furthermore, by utilizing simultaneous transfers, it is expected that the time required for loading and unloading can be reduced.

[0083] [Second embodiment] This second embodiment is a modification based on the previous embodiment, and the description of the previous embodiment can be used.

[0084] The processing device 11 of this embodiment includes a condition registration unit F1b, a reading unit F2b, and an allocation unit F3b, as shown in Fig. 15. The condition registration unit F1b is configured to acquire premise data and store it in the database 15, as in the first embodiment. The condition registration unit F1b registers base data, AGV data, cargo data, and cost setting data in the database 15. The reading unit F2b is configured to read out various data stored in the database 15 upon receiving a predetermined start trigger. The data read by the reading unit F2b is temporarily stored in the memory 112 and is referenced by the allocation unit F3b, etc.

[0085] The allocation unit F3b is configured to determine the platoon allocation by reducing the route, assigned cargo, and platoon allocation to a set packing problem (or set covering problem) and solving it for each AGV 2. The allocation unit F3b lists plans that satisfy the capacity constraint (i.e., are feasible) based on the cargo data and capacity data, and determines a plan that satisfies a predetermined objective function based on the listed plans.

[0086] Here, the plan is a tentative proposal for bin allocation, as shown in Figure 16. The plan may be referred to as a schedule proposal, allocation pattern, or delivery plan. Each row shown in Figure 16 corresponds to one plan. A plan number corresponding to the row number may be set for each plan. The plan may be a combination of bins used to deliver packages. Since the departure and arrival points are set for each package, once the packages to be loaded onto the bins are determined, the route is also determined according to a predetermined rule (for example, so that the route length is minimized). In other words, from another perspective, the plan represents a route. Therefore, the plan may be referred to as a route. However, even if the route is the same, different bins may be used to deliver the packages. For this reason, the tentative proposal for bin allocation is referred to as a plan here. Plans may be comprehensively listed within a range that satisfies capacity constraints.

[0087] For simplicity, Figure 16 shows a plan for delivering two packages L1 and L2 sent from base N1 to base N2, assuming that two AGVs 2 are available, as shown in Figure 17. Each plan may be based on the allocation of packages to the loading pallets that minimizes the cost of delivering a package. The dashed-dotted line in Figure 17 indicates route R1 that can be adopted by vehicle A, the first AGV 2, and the dashed-dotted line indicates route R2 for vehicle B, the second AGV 2. Vehicles A and B have different home positions (i.e., depots), but are both designed to pass through bases N1 and N2. Route R2 is longer than route R1. For example, route R1 is equivalent to a cost of 50, while route R2 is equivalent to a cost of 62. The cost here may be an evaluation value of the physical length of the route (i.e., route length) or an evaluation value of travel time.

[0088] In Figures 16 and 17, "A1" represents the front loading platform of vehicle A, and A2 represents the second loading platform of vehicle A. B1 represents the front loading platform of vehicle B, and B2 represents the second loading platform of vehicle B. The first and fifth plans (#1, #5) shown in Figure 16 indicate that no luggage is assigned to either vehicle A or vehicle B, and luggage L1 and L2 are not delivered. The second plan (#2) indicates a pattern in which luggage L1 is delivered using loading platform A1. Note that if plan (#2) is adopted, luggage L2 that is not delivered by vehicle A will be left behind or delivered by another vehicle. The third plan (#3) indicates a pattern in which luggage L2 is delivered using loading platform A1. The fourth plan (#4) indicates a pattern in which luggage L1 and L2 are picked up simultaneously using loading platforms A1 and A2 and dropped simultaneously at location N2. An eighth plan (#8) shows a pattern in which vehicle B simultaneously picks up luggage L1 and luggage L2 using loading platform B1 and loading platform B2, and simultaneously drops them at base N2.

[0089] The allocation unit F3b may have a function to automatically generate the above plans based on the cargo data and capacity data. The above multiple plans may be automatically generated by inputting the OD information for each cargo and the number of platoons for each AGV2 into a predetermined program so as to prevent overlapping of platoons.

[0090] When the generation of the plans is completed, the allocation unit F3b creates a matrix A corresponding to the listed plans and determines the allocation of loading bays for each cargo using the matrix A. As shown in Fig. 18, the matrix A may be a matrix in which 1 is set to elements in the columns corresponding to the plans where there is a loading / unloading action and 0 is set to elements where there is no loading / unloading action.

[0091] The allocation unit F3b also creates a cost vector c having cost values ​​corresponding to each plan as elements. As shown in FIG. 19, the cost vector c may be a row vector with a number of columns corresponding to the number of plans. The value of the jth column of the cost vector c represents the cost of the jth plan. For example, the first to fourth columns indicate the costs of plans #1 to #4. The first column corresponds to the first plan (#1) that does not deliver cargo, so a cost value of (0) is inserted. However, since plans #2 to #4 all involve vehicle A traveling along route R1, a common cost value (50) is inserted. Furthermore, the fifth to eighth columns of the cost vector c indicate the costs of plans #5 to #8. The sixth to eighth columns correspond to plans (#6 to #8) that involve vehicle B traveling along route R2, so a common cost value (62) is inserted. The cost value (0) for no delivery is applied to the fifth column. The cost may be the sum of the travel time along the route and the transfer time. The transfer time may be a value obtained by multiplying the number of transfer actions performed by the time required for one transfer action.

[0092] When the generation of the matrix A and the cost vector c is completed, the allocation unit F3b uses these data to solve the optimization problem shown in Figure 20 to determine the bin allocation. Figure 20 is a formulation of the set packing problem. ij" means the element in the jth column and ith row of matrix A, and takes the value 0 or 1. Matrix A is an m x n matrix consisting of 0-1 variables. m is the total number of loading and unloading actions, for example, twice the total number of packages. n corresponds to the number of plans. As mentioned above, matrix A is generated to include all plans that each AGV2 can realize.

[0093] a ij When = 1, it means that package i is transported by plan j, and a ij When = 0, it means that luggage i is not transported on plan j. N = {1, 2, ..., n} is the index of matrix A. N h ∈N represents a subset of routes (in other words, plans) starting from the home position h. j " represents the cost (basically the minimum value) when delivering using plan j. "β h " is the number of AGV2s waiting at home position h.

[0094] Formula (1b), which constitutes the optimization problem shown in Figure 20, is the objective function. Formula (1b) is a function whose objective is to maximize the number of parcels to be delivered while minimizing the total sum of costs resulting from movement and reloading. The first term in formula (1b) represents the total number of parcels that can be transported, and the second term represents the total sum of costs. "C" in Figure 20 digit " is a coefficient to change the number of digits in the first and second terms. "C digit " may be set to a sufficiently large value so that the first term takes priority. The second term in equation (1b) may be interpreted as a factor for determining the relative merits of plans from the viewpoint of cost when there are multiple plans that can deliver the same number of packages. Note that this objective function is just an example, and other objective functions may be applied. The objective function may be to minimize the route length, maximize the number of packages, minimize the delay time, etc.

[0095] Equation (2b) is a constraint that specifies that each parcel is transported only once. Equation (3b) states that the number of routes to be assigned to an AGV2 waiting at home position h is the number of AGVs waiting at home position h (β h) or less. Of course, the set-packing problem corresponding to bin allocation may be formulated using another mathematical model. The allocation unit F3b may be configured to obtain a solution for bin allocation using a mathematical model different from the mathematical model shown in FIG. 20.

[0096] The allocation unit F3b may solve the above optimization problem using, for example, an optimization solver. The allocation unit F3b may be implemented with an optimization solver. The optimization solver is software for obtaining an optimal solution to the optimization problem. The optimization solver may be any solver, such as Gurobi Optimizer, OptCover, IBM CPLEX, or SCOP.

[0097] The allocation unit F3b determines the bin allocation for each parcel according to the results of solving the formulated optimization problem. With the above configuration, the route, assigned parcels, and bin allocation for each AGV2 can be optimized. Therefore, it is expected that delivery efficiency will be further improved. The objective function of the optimization problem is not limited to minimizing cost. The objective function may be minimizing the number of AGV2s in operation, minimizing delivery time, or maximizing the number of parcels delivered. The allocation unit F3b may also determine the bin allocation using another solution method, such as an approximate solution method. For example, the allocation unit F3b may solve the problem using a column generation method, a local search method, or the like.

[0098] [Third embodiment] This third embodiment is a modification of the preceding embodiments as a basic form, and the descriptions of the first and second embodiments can be used. The processing device 11 of this embodiment includes a condition registration unit F1c, a reading unit F2c, and an allocation unit F3c, as shown in Fig. 21. The condition registration unit F1c, like the first embodiment, is configured to acquire premise data such as base data and AGV data and store the data in a database 15.

[0099] The reading unit F2c is configured to receive a predetermined start trigger and read out various data stored in the database 15. The data read by the reading unit F2c is temporarily stored in the memory 112 and is referenced by the route allocation unit F31c and the like.

[0100] The allocation unit F3c is configured to determine the route and assigned cargo for each AGV2, and further allocate bins. The allocation unit F3c, in general, determines the route and assigned cargo for each AGV2 by solving a first optimization problem regarding the allocation of the route and assigned cargo for each AGV2. Then, based on the results of the first optimization problem, the allocation unit F3c determines the bin allocation by solving a second optimization problem regarding the bin allocation.

[0101] The allocation unit F3c includes a route allocation unit F31c as a sub-function for performing the above processing. The route allocation unit F31c is configured to calculate a first optimization problem related to route allocation (hereinafter also referred to as a route and luggage allocation problem) based on luggage data, capacity data, and route candidate data, and determine the route and luggage assigned to each AGV2.

[0102] The route and package allocation problem may be an optimization problem with the objective of maximizing the number of packages within a certain route length or delivery time, maximizing the number of packages along the shortest route length, or other objectives. The route and package allocation problem may be a problem that takes into account pass constraints, delivery constraints, capacity constraints, and delivery time constraints. A pass constraint may be that the same base is passed only once. A delivery constraint may be that the package must pass through a shipping base before reaching the delivery destination.

[0103] The capacity constraint means that the AGV cannot carry more parcels than the number of bins when repeatedly picking up and dropping parcels. For example, if the target AGV has three bins, it cannot carry more than three parcels. The delivery time constraint may be a constraint that specifies that each parcel must be delivered while satisfying the requirement that it be delivered within a certain period of time. The route and parcel assignment problem may include a constraint on the maximum route length instead of or in addition to the delivery time constraint. This type of problem is known as the Pick and Delivery Problem (PDP), and the route and parcels assigned to each AGV2 may be determined, for example, by using the method disclosed in Non-Patent Document 1.

[0104] As an example, the route assignment unit F31c may determine the route and assigned loads for each AGV2 by solving the optimization problem shown in FIG. 22. The objective of this optimization problem is to minimize the total distance while satisfying the delivery time constraints for each load and the upper limit of the number of loads that the AGV2 can carry. Equations (1c) to (5c) shown in FIG. 22 are constraints on the number of trips, i.e., constraints that AGV2 must visit each base only once. Equation (6c) is a time constraint. Equation (7c) is a delivery constraint, i.e., constraints that require that the AGV2 must drop off the load after picking it up. Equation (9) is a capacity constraint, i.e., constraints that require that the AGV2 must satisfy the upper limit of the number of loads that it can deliver at one time. Equations (8) and (10) are constraints for removing partial routes. Of course, the route assignment unit F31 may use another mathematical model to determine the route and assigned loads for each AGV2.

[0105] The route allocation unit F31c may update the list of unassigned loads for each AGV 2 and repeatedly execute the problem calculation each time the loads assigned to each AGV 2 are determined. In one aspect, such a route allocation unit F31c may be considered as a configuration that optimizes the allocation of routes and loads.

[0106] The allocation unit F3c calculates an optimization problem for platoon allocation (hereinafter referred to as platoon allocation problem) based on the route and assigned baggage data for each AGV2 determined by the above process. The calculation of the platoon allocation problem may be performed independently for each AGV2. The calculation of the platoon allocation problem for each AGV2 may also be performed in parallel.

[0107] The allocation unit F3c may calculate the bin allocation problem by using the solution method described in the second embodiment. That is, the allocation unit F3c may be configured to solve the bin allocation problem using a matrix (hereinafter, candidate matrix) that mechanically lists bin allocation patterns for assigned cargo. In the allocation unit F3c, each column of the candidate matrix represents a bin allocation pattern and can be called an allocation plan. The allocation unit F3c may calculate the total cost including the transfer cost for each plan (each column of the candidate matrix), and output the allocation pattern (in other words, the plan) that minimizes the cost as the optimal solution. The optimal solution may be calculated using the mathematical model shown in FIG. 20 or a partially modified mathematical model.

[0108] The cost used to evaluate a plan may be the sum of the travel cost between bases and the transfer cost. However, as in this embodiment, when the route is already fixed at the time of considering the allocation of loading bays, the travel cost between bases for each plan is constant. Therefore, the cost used to evaluate a plan may be the sum of the transfer cost, excluding the travel cost. The more simultaneous transfers a plan allows, the lower the calculated cost. Simultaneous transfers are possible when two or more transfer actions can be performed without moving within the base. Whether or not transfer within the base is necessary may be determined based on the layout of the delivery port at base 3 and the loading bay number planned to be used at that base 3. The cost may be added based on the number of times transfers within the base occur.

[0109] The allocation unit F3c may generate data indicating the results of a determination of whether simultaneous transfers are possible or whether intra-site transfers are required at bases where multiple transfer actions occur for each plan, and select a plan based on the data indicating the number of intra-site transfers. The allocation unit F3 may determine that simultaneous transfers are possible if two adjacent loading pallets are available when the two packages arrive at a base where they are scheduled to be picked up. The allocation unit F3 may determine that simultaneous transfers are possible if the two packages scheduled to be dropped off are loaded on adjacent loading pallets when the two packages arrive at a base where they are scheduled to be dropped off. The allocation unit F3 may determine that simultaneous transfers are possible if a loading pallet next to the loading pallet where the two packages are scheduled to be dropped off is available when the two packages arrive at a base where one drop-off and one pickup are scheduled to be performed. For simplicity of explanation, a pattern in which simultaneous transfers are performed using adjacent loading pallets has been described here, but the delivery ports and loading pallets where simultaneous transfers can be performed do not have to be adjacent. Simultaneous transfers may also be performed using a combination of loading pallets that are not adjacent to each other.

[0110] According to the above configuration, the route and assigned load for each AGV2 are optimized, and then the allocation of load bins to each AGV2 is also optimized. This is expected to further improve delivery efficiency. In addition, the problem handled by the processing device 11 is roughly divided into two parts, and optimization is performed in two stages. According to the third embodiment, the calculation time can be reduced compared to the method of the second embodiment, which optimizes everything from route allocation to load bin allocation all at once. Note that the objective function for the optimization problem related to load bin allocation is not limited to minimizing cost. The objective function may also be minimizing the number of movements within a base station or maximizing the number of simultaneous transfers.

[0111] The allocation unit F3c may derive an optimal solution by brute-forcely evaluating the cost for each allocation pattern as described above, but the configuration of the allocation unit F3c is not limited to this. As described in the first embodiment, the allocation unit F3c may determine bin allocation by solving a two-dimensional packing problem in which the period during which a package occupies a bin is considered as one item. The two-dimensional packing problem related to bin allocation for AGV2 is reduced to a mathematical model that uses, as input data, OD information for packages that can be delivered assuming that AGV2 travels along a route indicated in certain route data and information regarding the simultaneous transfer of those packages, and includes constraints on the number of bins indicated in the capacity data and constraints on the overlapping occupancy times of multiple packages. The allocation unit F3c may determine bin allocation using a mathematical model that includes constraints corresponding to the occupancy times of packages. In this way, the allocation unit F3c may be configured to optimize everything from route allocation to assigning assigned packages, and then perform bin allocation in the same manner as in the first embodiment. This configuration also improves delivery efficiency. The allocation unit F3c may also be configured to output an approximate solution (quasi-optimal solution) for bin allocation using other heuristic methods.

[0112] [Fourth embodiment] This fourth embodiment is a modification of the preceding embodiments as a basic form, and the descriptions of the first, second, and third embodiments may be used. The route and assigned cargo for each AGV 2 may be designed by an administrator. The database 15 may contain constraint data for the route and assigned cargo for each AGV 2 created by the administrator. In this case, the processing device 11 may execute a process for optimizing the bin allocation based on the route and assigned cargo data for each AGV 2 registered in the database 15.

[0113] In this fourth embodiment, data on the routes and assigned cargo for each AGV2 is already registered in the database 15, and the processing device 11 is configured to solve the optimization problem of loading bay allocation based on the data on the routes and assigned cargo for each AGV2 registered in the database 15.

[0114] The processing device 11 of this embodiment includes a condition registration unit F1d, a reading unit F2d, and an allocation unit F3d, as shown in Fig. 23. The condition registration unit F1d generates route data and assigned baggage data for each AGV 2 based on an operation signal received via the input device 12, and stores the data in the database 15. Note that the database 15 may also contain pre-registered AGV data including capacity data for each AGV 2, base data including the location of delivery ports at the base 3, and cost setting data. Upon receiving a predetermined start trigger, the reading unit F2d reads out the route data and assigned baggage data for each AGV 2 stored in the database 15 and expands them in the memory 112.

[0115] The allocation unit F3d is configured to solve the bin allocation problem using a candidate matrix that mechanically lists bin allocation patterns for assigned cargo, similar to the allocation unit F3c of the third embodiment. That is, the allocation unit F3d is configured to use a predetermined mathematical model to output, as a solution, a plan that minimizes the cost from among multiple plans.

[0116] The above configuration also makes it possible to efficiently allocate loading bays. In addition, since the routes and cargo for each AGV2 are already determined, the calculation time itself can be shortened.

[0117] The allocation unit F3 may determine the allocation of bins by solving a two-dimensional packing problem in which the period during which a package occupies a bin is regarded as one item, as described in the first and third embodiments. This configuration can also improve delivery efficiency.

[0118] [Fifth embodiment] The processing device 11 may include multiple calculation units F4 that perform loading bay allocation using different methods. For example, as shown in FIG. 24, the processing device 11 may include a first calculation unit F4a, a second calculation unit F4b, a third calculation unit F4c, and a management unit F5 in addition to a condition registration unit F1 and a reading unit F2. These functional units may be realized by the processor 111 executing a predetermined program. The condition registration unit F1 registers prerequisites in the database 15. The reading unit F2 reads data registered in the database 15 and stores it in the memory 112.

[0119] The first calculation unit F4a is configured to set routes and allocate carriers for each AGV2 using the method of the first embodiment. The second calculation unit F4b is configured to set routes and allocate carriers for each AGV2 using the method of the second embodiment. The third calculation unit F4c is configured to set routes and allocate carriers for each AGV2 using the method of the third embodiment. The first calculation unit F4a, the second calculation unit F4b, and the third calculation unit F4c may each be realized using a different processor. The processing device 11 may be equipped with multiple processors. The first calculation unit F4a, the second calculation unit F4b, and the third calculation unit F4c operate based on instructions from the management unit F5. The number of calculation units F4 equipped in the processing device 11 is not limited to three, and may be two, four, or more.

[0120] Priorities may be set in advance for the multiple calculation units F4 according to the expected accuracy of the output solution. The accuracy of the solution corresponds to the level of delivery efficiency. For example, the second calculation unit F4b, which treats the process from route assignment to pallet assignment as a single set packing problem, is expected to produce the most accurate solution. Therefore, the priority of the second calculation unit F4b may be set to the highest. Furthermore, the third embodiment, which separates the problem into route assignment and package assignment, and then optimizes each separately, is expected to produce the next most accurate solution after the second embodiment. Therefore, the priority of the third calculation unit F4c is set to the second highest. In the first embodiment, in which the route of the AGV2 is determined based on rules, the calculation time can be reduced, but the accuracy of the solution may vary depending on the rule settings. From the perspective of stability and reliability of the solution accuracy, the priority of the first calculation unit F4a corresponding to the first embodiment may be set to the third highest. Note that the priority settings described here are merely examples and may be changed as appropriate. The priority of the first calculation unit F4a may be set to second, and the priority of the third calculation unit F4c may be set to third.

[0121] In general, solving an optimization problem can take a calculation time that depends on the scale of the problem, specifically, the scale of input data such as the number of bases, the number of packages, and the number of AGVs 2. Compared to the first embodiment, which reduces the problem to a simple packing problem and allocates bins, the method of the second embodiment can take a long calculation time. The first calculation unit F4a and the third calculation unit F4c can be considered as backup means for when the calculation by the second calculation unit F4b is difficult.

[0122] The management unit F5 is configured to determine a final delivery plan based on one or more allocation results obtained by having the multiple calculation units F4 execute calculations related to bin allocation. The management unit F5 has a function to control the operation of the first calculation unit F4a, the second calculation unit F4b, and the third calculation unit F4c. For example, as shown in FIG. 25, when a predetermined start trigger is input (S201), the management unit F5 outputs a calculation start instruction to the multiple calculation units F4, causing them to start calculations (S202). That is, the first calculation unit F4a, the second calculation unit F4b, and the third calculation unit F4c start calculations in parallel based on the instruction from the management unit F5. The problem that each calculation unit F4 is addressing, i.e., the premise, such as the parcel data, is the same. The timing at which the calculation start instruction is output corresponds to the calculation start timing.

[0123] The management unit F5 then measures the elapsed time from the start of calculation and determines whether a predetermined waiting time has elapsed since the start of calculation (S203). This waiting time may be, for example, 5 or 10 minutes. If there is a constraint that AGV2 must not be stopped even while calculating the operation plan for AGV2, the waiting time may be set in units of seconds, such as 1 or 2 seconds. After a certain time has elapsed since the start of calculation, the management unit F5 determines whether at least one calculation result (in other words, a solution or allocation result) has been obtained (S205). If no solution has yet been obtained (NO in S204), i.e., if none of the multiple calculation units F4 have completed calculation, the management unit F5 may re-execute S204 after a certain time has elapsed. In this case, the waiting time may be less than half the waiting time in S203, such as 2 or 5 minutes.

[0124] If at least one solution has been obtained (S204 YES), the management unit F5 next determines whether only one solution has been obtained or multiple solutions have been obtained (S205). If only one solution has been obtained (S205 YES), the management unit F5 adopts that solution and generates delivery plan data (S206). On the other hand, if solutions have been output from multiple calculation units F4 (S205 NO), the management unit F5 adopts the solution output from the calculation unit F4 with the relatively highest priority from among the multiple obtained solutions as the final solution and generates delivery plan data (S207).

[0125] With this configuration, if the second calculation unit F4b can output a solution within a certain time, a delivery plan is generated based on the solution of the second calculation unit F4b. In other words, a delivery plan including platoon allocation is generated based on the optimal solution and reflected in the actual operation of the AGV2. This makes it possible to optimize delivery efficiency. Also, if the input data is large and the second calculation unit F4b cannot output a solution within the operational time limit, a delivery plan is generated based on the output solution of another calculation unit F4. This reduces the risk of the AGV2 being unable to deliver due to waiting for the generation of a delivery plan. In other words, the delivery control system 1 can flexibly respond to the number of packages, etc.

[0126] As a modification of the fifth embodiment, the management unit F5 may be configured to determine the complexity of the allocation plan based on the contents of the prerequisites and select the calculation unit F4 to be used for calculating the bin allocation according to the complexity. The complexity of the allocation plan corresponds to the scale of the input data. The management unit F5 may determine the complexity based on at least one of the total number of packages, the number of bases, the total number of delivery ports, the number of AGVs 2, the total number of bins, and the number of candidate routes. The complexity may be evaluated using a predetermined number of levels, such as two levels (high and low) or three levels (high, medium, and low). Of course, it may also be expressed as a score from 1 to 100 or 0.1 to 1.0.

[0127] For example, the management unit F5 may determine that the complexity is high when the total number of packages is equal to or greater than a predetermined value. The management unit F5 may determine that the complexity is high when the number of bases or the total number of delivery ports is equal to or greater than a predetermined value. The management unit F5 may also determine that the complexity is high when the number of AGVs 2 or the total number of platoons is equal to or greater than a predetermined value. The complexity may also be determined in consideration of the number of candidate routes that the AGV 2 can take. Of course, the management unit F5 may determine the complexity by combining multiple types of items. For example, the complexity may be determined to be high based on whether the sum or product of the total number of packages, the total number of delivery ports, and the total number of platoons is equal to or greater than a predetermined value. The complexity determination algorithm may be designed as appropriate.

[0128] The management unit F5 may create a delivery plan using the first calculation unit F4a if it determines that the complexity is high, create a delivery plan using the third calculation unit F4c if it determines that the complexity is medium, and create a delivery plan using the second calculation unit F4b if it determines that the complexity is low. Alternatively, if the management unit F5 determines that the complexity is not high, it may create a delivery plan based on the results of parallel calculations by all three calculation units F4, as in the fifth embodiment. If it determines that the complexity is high, the management unit F5 may create a delivery plan based on the results of parallel calculations by the first calculation unit F4a and the third calculation unit F4c. The management unit F5 may change the combination of calculation methods used to calculate the problem depending on the complexity.

[0129] <Additional remarks (1)> The AGV2 may be a magnetically guided AGV2 that uses a magnetic sensor to detect a magnetic tape embedded in the floor and travels along the route. The AGV2 may also be an optically guided AGV2 that has an optical sensor and travels while detecting the reflection of a guide tape attached to the floor. The AGV2 may be an image recognition AGV2 that uses a camera to read a 2D code or an AR marker, recognizes its position, and travels accordingly. The AGV2 may be a SLAM (Simultaneous Localization and Mapping) AGV2 that uses information obtained by a camera, laser, etc. to estimate its current location and create an environmental map to travel accordingly.

[0130] The AGV 2 may be, like a train, a combination of multiple vehicles (hereinafter also referred to as carts) each having one or two loading platforms 28. When the AGV 2 is configured by coupling carts, the number of loading platforms 28 provided on the AGV 2 can be varied depending on the number of couplings.

[0131] The AGV2 described above is configured to transfer cargo using rollers and motors on the platform, but the method for realizing automatic transfer is not limited to this. Automatic transfer can also be achieved using an arm robot.

[0132] The present disclosure may be applied not only to delivery robots such as AGV2, but also to trucks and the like. When applied to trucks and the like, there may be advantages such as eliminating the need to consolidate shipping areas in one location, and thus increasing the flexibility of facility layout. In addition, the receiving port 51a at the base 3 may have a width equivalent to the width of multiple loading bays 28. The role of each receiving port may not be fixed, but may be variable depending on the settings. In this case, after loading bay allocation is completed, the role (receiving / shipping) of the delivery port at the base 3 may be determined to match the loading bay allocation, in other words, to reduce movement within the base. The processing device 11 may have a function to optimize the role allocation of the delivery ports at the base 3.

[0133] <Additional remarks (2)> The various flowcharts shown in this disclosure are all examples, and the number of steps constituting the flowcharts and the execution order of the processes can be changed as appropriate. The controls shown in each flowchart may be combined / executed in parallel to the extent that there is no contradiction. Expressions such as acquisition, determination, detection, generation, and calculation may be used interchangeably. The acquisition of certain data by a certain device also includes the device generating the data based on a signal input from another device / sensor. The processor may be any computing core, such as a CPU, MPU, GPU, or DFP (Data Flow Processor). Some or all of the functions of the processing device 11 may be realized as hardware. [Explanation of symbols]

[0134] 1 Delivery control system, 2 AGV (delivery vehicle), 3 base, 28 loading platform, 11 processing device (delivery plan creation device), 12 input device, 14 communication device, 15 database, 51 delivery port, 51a receiving port, 51b shipping port, 111 processor (arithmetic device), 112 memory, F1 condition registration unit, F2 reading unit, F4 calculation unit, F4a first calculation unit, F4b second calculation unit, F4c third calculation unit, F5 management unit

Claims

1. A delivery plan creation device that creates a delivery plan for at least one delivery vehicle (2), which is equipped with a plurality of loading platforms (28) and is configured to be able to transfer cargo by stopping the vehicle so that the loading platform to be transferred is located at a location corresponding to a delivery port at a base, at least one computing device (111); The computing device Acquire route data indicating the route of the delivery vehicle, capacity data indicating the number of loading bays provided in the delivery vehicle, layout data indicating the number and positions of the delivery ports at the base, and parcel data indicating the shipping location and delivery destination for each parcel, A delivery plan creation device configured to create the delivery plan including cargo bay allocation based on the route data, the capacity data, the layout data, and the cargo data.

2. A delivery plan creation device that creates the delivery plans for a plurality of the delivery vehicles, The computing device Execute a process of acquiring the route data for each delivery vehicle and the parcel data for the parcels assigned to each delivery vehicle as input data; 2. The delivery plan creation device according to claim 1, configured to determine the loading bay allocation for each delivery vehicle based on the acquired route data and the package data of the package in charge.

3. The computing device generating data listing a plurality of executable delivery plans for each of the delivery vehicles based on the capacity data for each of the delivery vehicles and the cargo data for the cargo in charge that are acquired as the input data; calculating an optimization problem for identifying a delivery plan that maximizes or minimizes a value of a predetermined objective function from among the delivery plans based on data of the delivery plans; The delivery plan creation device according to claim 2 , configured to determine the allocation of loading bays for each of the delivery vehicles based on a solution to the optimization problem.

4. The computing device 3. The delivery plan creation device according to claim 2, wherein the delivery plan creation device is configured to determine the allocation of loading bays for each of the delivery vehicles by solving the allocation of loading bays for each of the delivery vehicles as a two-dimensional packing problem.

5. A delivery plan creation device that creates the delivery plans for a plurality of the delivery vehicles, The computing device Read out, from a predetermined database, movement cost data indicating the movement cost between the bases and transfer cost data indicating the transfer cost, which is the cost required for one transfer; Implementing a process of determining a route and assigned cargo for each delivery vehicle based on the movement cost data, the transfer cost data, the cargo data, and the capacity data for each delivery vehicle; 2. The delivery plan creation device according to claim 1, configured to determine the loading bay allocation for each delivery vehicle based on the route for each delivery vehicle and the assigned cargo determined in the above process.

6. 6. The delivery plan creation device according to claim 5, wherein the calculation device is configured to determine the route and the loads to be carried by each delivery vehicle by solving a predetermined optimization problem including constraints that the number of loads that the delivery vehicle can carry when fully loaded is equal to or less than the number of bays, and that the loads pass the shipping location before the delivery destination.

7. generating data listing a plurality of executable delivery plans for each of the delivery vehicles based on the capacity data for each of the delivery vehicles and the cargo data for the cargo in charge; calculating an optimization problem for identifying one of the plurality of delivery plans that maximizes or minimizes a value of a predetermined objective function based on data of the plurality of delivery plans; The delivery plan creation device according to claim 5 , configured to determine a route, a load, and a loading bay assignment for each of the delivery vehicles based on a solution to the optimization problem.

8. The computing device 6. The delivery plan creation device according to claim 5, configured to determine the allocation of loading bays for each of the delivery vehicles by solving the allocation of loading bays for each of the delivery vehicles as a two-dimensional packing problem.

9. A delivery plan creation device that creates the delivery plans for a plurality of the delivery vehicles, the plurality of delivery vehicles include a first vehicle and a second vehicle; The computing device determining a route for the first vehicle based on a predetermined rule before determining a route for the second vehicle; extracting, as candidates for handling, luggage that can be handled by the first vehicle based on the luggage data for each luggage and the route of the first vehicle; determining the loads and load bay allocations of the first vehicle by solving a two-dimensional packing problem to select loads that can actually be delivered by the first vehicle and allocate load bays based on the number of load bays of the first vehicle and the load data of the candidate; The delivery plan creation device according to claim 1 , configured to determine a route for the second vehicle so as to deliver part or all of the packages that cannot be delivered by the first vehicle.

10. A delivery plan creation device that creates the delivery plans for a plurality of the delivery vehicles, The computing device The layout data, the capacity data for each delivery vehicle, the cargo data for each cargo, movement cost data indicating the movement cost between the bases, and transfer cost data indicating the transfer cost which is the cost required for one loading and unloading are taken in as input data, generating data listing a plurality of executable delivery plans for each of the delivery vehicles based on the capacity data for each of the delivery vehicles and the parcel data for each of the parcels; calculating an optimization problem for identifying one of the plurality of delivery plans that maximizes or minimizes a value of a predetermined objective function based on data of the plurality of delivery plans; The delivery plan creation device according to claim 1 , configured to determine a route, a load, and a loading bay assignment for each of the delivery vehicles based on a solution to the optimization problem.

11. A delivery plan creation device that creates the delivery plans for a plurality of the delivery vehicles, The computing device A plurality of calculation units (F4a, F4b, F4c) each calculating a delivery plan including a bin allocation using a different method; a management unit (F5) that controls the plurality of calculation units, A priority is preset for each of the plurality of calculation units, The management unit operating the plurality of calculation units in parallel at a predetermined calculation start timing; When a predetermined time has elapsed since the start of the calculation, if there is only one calculation unit outputting a calculation result, the calculation result is adopted; 2. The delivery plan creation device according to claim 1, wherein, when there are a plurality of calculation units outputting the calculation results at a timing when a predetermined time has elapsed from the calculation start timing, the calculation result to be finally adopted is selected based on the priority of the calculation units outputting the calculation results.

12. A delivery plan creation device that creates a delivery plan for a plurality of delivery vehicles, The computing device A plurality of calculation units (F4a, F4b, F4c) that calculate routes, assigned packages, and loading bay allocations for each delivery vehicle using different methods; a management unit (F5) that controls the plurality of calculation units, The management unit determining the complexity of the bin allocation based on at least one of the total number of packages, the number of base stations, the total number of delivery ports, the number of delivery vehicles, the total number of bins, and the number of route candidates that the delivery vehicles can take; The delivery plan creation device according to claim 1 , configured to select the calculation unit to be used in determining the loading bin allocation in accordance with the complexity.

13. The computing device data indicating a stop position corresponding to the transfer port is taken in as input data; Based on the data of the stop positions corresponding to the delivery port, the delivery plan for the delivery vehicle including stop positions for transferring the cargo is created; The delivery plan creation device according to claim 1 , configured to transmit an instruction signal to the delivery vehicle to cause the delivery vehicle to travel in accordance with the delivery plan.

Images