Construction material inventory management device and program

Abstract

This reduces the possibility that a truck authorized to leave the warehouse may contain construction materials that were mistakenly loaded. [Solution] The outbound management device comprises a first acquisition unit 51A, a second acquisition unit 51B, and an outbound permission unit 51D. The first acquisition unit 51A acquires a predicted image representing the predicted loading configuration of multiple building materials (ordered materials) that are loaded onto trucks in the warehouse and scheduled to be outbound from the warehouse. Furthermore, the second acquisition unit 51B acquires an image capturing the actual loading configuration of the multiple building materials that have been loaded onto the trucks. The outbound permission unit 51D then performs a permission process to allow the truck loaded with multiple building materials to leave the warehouse if the similarity between the predicted loading configuration based on the predicted image and the image capture and the actual loading configuration is higher than a predetermined standard.

Description

Technical Field

[0001] The present invention relates to a delivery management device for construction materials, and a program.

Background Art

[0002] Structures such as buildings and temporary scaffolding are composed of a plurality of construction materials. For example, temporary scaffolding, which is a temporary structure, is constructed by assembling a plurality of temporary materials. Patent Document 1 discloses, as temporary materials, struts, horizontal members bridged between two struts, work scaffolding boards whose both ends are locked to two horizontal members, skirting boards arranged along the work scaffolding boards, and the like.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Construction materials such as temporary materials may be transported out of a predetermined warehouse by a truck for use. When construction materials are delivered out of the warehouse by truck, the delivered construction materials may not match the construction materials that are supposed to be transported out due to mistakes such as incorrect loading of the construction materials.

[0005] The present invention aims to reduce the possibility that a plurality of construction materials loaded on a truck permitted to exit the warehouse include incorrectly loaded construction materials.

Means for Solving the Problems

[0006] The construction material dispatch management device according to the present invention comprises: a first acquisition unit that acquires a predicted image representing the predicted loading configuration of a plurality of construction materials that are loaded onto a truck in a warehouse and are scheduled to be dispatched from the warehouse; a second acquisition unit that acquires an image capturing the actual loading configuration of the plurality of construction materials that have actually been loaded onto the truck; and a dispatch permission unit that performs a permission process to permit the truck carrying the plurality of construction materials to leave the warehouse when the similarity between the predicted loading configuration based on the predicted image and the image capturing and the actual loading configuration is higher than a predetermined standard.

[0007] The program according to the present invention causes a computer to function as the inventory management device. [Effects of the Invention]

[0008] According to the present invention, the possibility that a truck authorized to leave a warehouse may contain incorrectly loaded construction materials is reduced. [Brief explanation of the drawing]

[0009] [Figure 1] Figure 1 is a diagram showing the configuration of the outbound inventory management system, including the outbound inventory management device according to this embodiment. [Figure 2] Figure 2 shows the configuration of the outbound inventory management device according to this embodiment. [Figure 3] Figure 3 shows the flow chart for managing the dispatch of construction materials. [Figure 4] Figure 4 shows an example of the configuration of a building materials table. [Figure 5] Figure 5 shows an example of packaging for building materials. [Figure 6] Figures 6(a) to 6(c) illustrate the loading conditions when loading construction materials onto a truck. [Figure 7] Figure 7 shows an example of a predictive image representing the predicted packaging of building materials. [Figure 8] Figure 8 shows an image of a truck loaded with construction materials. [Figure 9]Figure 9 shows an example of the structure of an order information table. [Figure 10] Figure 10 shows another example of a predictive image representing the predicted packaging of building materials. [Figure 11] Figure 11 is a converted image of a captured image of a truck loaded with construction materials. [Modes for carrying out the invention]

[0010] Embodiments of the present invention will be described below with reference to the drawings. In each figure, the same or similar elements are denoted by the same reference numeral. In addition, in cases where multiple elements that should be denoted by the same reference numeral are denoted by the same reference numeral, only some of the elements may be denoted by the same reference numeral.

[0011] The inventory management system 10 according to this embodiment, shown in Figure 1, is configured to manage the dispatch of construction materials stored in the warehouse WH from the warehouse WH. After ordering the necessary construction materials, the user enters the warehouse WH in a truck TR and loads multiple construction materials related to the current order from among the multiple construction materials stored in storage areas P1 to P8 onto the truck TR. After that, the truck TR leaves the warehouse WH. The inventory management system 10 manages the dispatch of construction materials upon the departure of the truck TR. The construction materials here are materials for temporary scaffolding. Temporary scaffolding materials include support columns, horizontal members, scaffolding planks, and connectors for linking the support columns, etc. The construction materials may also be materials for buildings other than temporary scaffolding.

[0012] The outbound management system 10 includes an entry management device 31, an entry gate 32, a drive device 33, cameras 34A to 34D (collectively referred to as "34"), an exit management device 41, an exit gate 42, a drive device 43, cameras 44A to 44D (collectively referred to as "44"), and an outbound management device 50.

[0013] The entry management device 31 is composed of a computer or the like that manages entry of the truck TR into the warehouse WH. The drive device 33 is composed of a linear motor or the like, and opens and closes the entry gate 32. The entry gate 32 is normally closed. The entry management device 31 controls the drive device 33 to open the entry gate 32 when entry of the truck TR is permitted as described later. This allows the truck TR to enter the warehouse WH.

[0014] Each of the cameras 34A to 34D is a fixed-point camera, and captures images of the truck TR stopped in the region R1 near the entrance of the warehouse WH from mutually different directions. The truck TR imaged here is a truck TR with an empty loading bed TR1 before being loaded with building materials. The use of the captured images will be described later.

[0015] The exit management device 41 is composed of a computer or the like that manages exit of the truck TR from the warehouse WH. The drive device 43 is composed of a linear motor or the like, and opens and closes the exit gate 42. The exit gate 42 is normally closed. The exit management device 41 controls the drive device 43 to open the exit gate 42 when exit of the truck TR is permitted as described later. This allows the truck TR to exit from the warehouse WH.

[0016] Each of the cameras 44A to 44D is a fixed-point camera, and captures images of the truck TR stopped in the region R2 near the exit of the warehouse WH from mutually different directions. The truck TR imaged here is the truck TR after being loaded with building materials. The relative positions and orientations of the cameras 34A to 34D with respect to the region R1 are the same as the relative positions and orientations of the cameras 44A to 44D with respect to the region R2. Therefore, the four captured images taken by the cameras 34A to 34D and the four captured images taken by the cameras 44A to 44D are images of the truck TR captured from the same direction and at the same distance between the corresponding captured images.

[0017] The entry management device 31, the exit management device 41, and the cameras 34 and 44 are configured to be capable of communicating with the warehousing management device 50 via a network NW such as the Internet or a local area network (LAN).

[0018] The warehousing management device 50 is composed of a server computer or the like. The warehousing management device 50 includes a processor 51 such as a CPU (Central Processing Unit), a storage 52 which is a non-volatile storage device that stores various programs and various data executed or used by the processor 51, and a RAM (Random Access Memory) 53 that functions as the main memory of the processor 51. The warehousing management device 50 further includes a communication interface 54 such as a network interface card connected to the network NW, and a user interface 55 such as a display device and an operation device. The storage 52 stores a building material table and an order information table. These tables will be described later.

[0019] By executing the programs stored in the storage 52, the processor 51 operates as a first acquisition unit 51A, a second acquisition unit 51B, an entry permission unit 51C, and an exit permission unit 51D shown in Figure 2. Each of the units 51A to 51B cooperates with the entry management device 31, the exit management device 41, and one or more user terminals UT operated by a user (for example, a personal computer, a smartphone, a tablet, etc.) to realize the order and warehousing flow of building materials shown in Figure 3. Hereinafter, the flow shown in Figure 3 will be described with reference to Figures 1 to 3.

[0020] In the flow of Figure 3, a user who wishes to procure a desired building material operates their own user terminal UT to access the website managed by the warehousing management device 50, and places an order for the building material (Step S11). The order for building materials includes specification of the type and quantity of the building materials to be ordered.

[0021] In step S11, the user terminal UT communicates with the inventory management device 50, for example, to obtain the material name, product code, and specifications (below (A) to (C)) of each building material registered in the building material table described later, and displays them in a list on the display device of the user terminal UT. The user refers to the information of each building material displayed in the list and specifies the type and quantity of building materials to order (for example, 50 support columns of material name A, 50 horizontal members of material name D, and 40 scaffolding planks of material name R). These ordered building materials will be referred to as ordered materials below, and the type and quantity of building materials specified by the user, i.e., ordered materials, will be referred to as ordered material information below. The type of ordered material is identified by the material name and product code.

[0022] The user terminal UT holds user identification information that identifies the user, and truck identification information that identifies the truck TR (in particular, the specifications of the truck TR), which are pre-entered by the user. User identification information includes the user ID. Truck information includes information such as the maximum load capacity of the truck TR, the dimensions of the truck TR's cargo bed TR1, and the height limit of the truck TR. The dimensions of the cargo bed TR1 may include the dimensions of the floor, side panels, and rear panel of the cargo bed TR1. User identification information and truck identification information may be entered when ordering the above-mentioned building materials.

[0023] The user terminal UT transmits the ordered material information entered by the user, along with the user identification information and truck identification information, to the outbound management device 50 (step S12).

[0024] The ordered material information, user identification information, and truck identification information transmitted to the outbound management device 50 are acquired by the first acquisition unit 51A via the network NW and communication interface 54. Upon acquiring this information, the first acquisition unit 51A issues an order ID that uniquely identifies the order (step S21).

[0025] Subsequently, the first acquisition unit 51A, based on the ordered materials information, refers to the building materials table and generates a predicted image (see Figure 7; details will be described later) that shows the predicted packaging of the multiple ordered materials when they are loaded onto the truck TR (step S22).

[0026] Here, we will explain the building materials table used to generate predictive images. As shown in Figure 4, the building materials table contains information on each building material that can be ordered by the user. In the building materials table, the following information (A) to (I) is associated with each building material and registered. This information (A) to (I) includes information that identifies the building material (below (A) to (C)) as well as loading conditions when loading the building material onto a truck (below (D) to (I)). Note that packaging includes binding with binding material. Multiple building materials after packaging are also called packaging materials. (A) Material name (B) Product Code (C) Specifications, i.e., dimensions (height, width, and depth) and weight (D) Packaging material information: When packaging building materials of the same type (same material name and product code) to form a packaging material, (D-1) the quantity of building materials required for one packaging material, and (D-2) the height, width, depth, and weight of the packaging material. (E) Loading order: The higher the order, the earlier the item is loaded onto the TR1 cargo bed. (F) Number of a group that can be packaged together: A number that identifies a group that can be packaged together with different types of building materials to form a packaged material. (G) 90-degree rotation possible: Defines whether loading can be done sideways. (H) Basket insertion feasibility: Defines whether packaging by placing the item in a basket is possible. If packaging is possible, the predicted package will include a basket. (I) Sub-packaging and bundling pattern: Defines how the packaging shape changes from the normal packaging shape when packaging a number of building materials that is less than the number of building materials required for the normal packaging described in (D) above.

[0027] Regarding (G) above, for example as shown in Figure 5, since the support columns 101 and 102, which are of different lengths, can be packaged together as packaging material 100, the support columns 101 and 102 belong to the same group that can be packaged together and are assigned the same group number (for example, A-1).

[0028] Examples of cases defined in (J) above include those shown in Figures 6(a) to (c). Figure 6(a) shows a case where, due to a reduction in building materials, two sides (e.g., width and height) of the normally packaged packaging material 111 are reduced by the same proportion, as shown in packaging material 112. Examples of such building materials include support columns and horizontal members. Figure 6(b) shows a case where, due to a reduction in building materials, only one side (e.g., height) of the normally packaged packaging material 121 is reduced, resulting in packaging material 122. Examples of such building materials include scaffolding planks. In the case of Figure 6(c), the shape of the packaging material 131 when packaged together with other types of building materials is defined. Packaging material 131 may include, for example, packaging material 131A corresponding to the case in Figure 6(a) and packaging material 131B corresponding to the case in Figure 6(b).

[0029] Referring again to Figures 1-3, in step S22, the first acquisition unit 51A extracts the material name and product code of the ordered materials included in the ordered materials information. The first acquisition unit 51A uses the extracted material name and product code as keys to refer to the building materials table and acquires the information (C) to (H) above that is associated with those keys. The first acquisition unit 51A performs this process for each of the multiple types of ordered materials.

[0030] In step S22, the first acquisition unit 51A generates a predicted image representing the predicted packaging when the ordered materials indicated by the ordered materials information are loaded onto the truck TR, using the type and quantity of each ordered material in the ordered materials information and the information obtained from the building materials table, so as to satisfy the dimensions and maximum load weight indicated by the truck information. The predicted image is generated as a three-dimensional image. The predicted image is an image in which the packaging materials are approximated and combined with three-dimensional shapes such as polygonal prisms, cylinders, and parts of cylinders, as shown in the predicted image 70 in Figure 7. Parts of cylinders include semi-cylinders and arched prisms. Circles include ellipses. The predicted image 70 includes a cubic-shaped packaging material 91 approximating multiple scaffolding planks that are stacked and bundled, a cubic-shaped packaging material 92 approximating multiple bundled support columns and / or horizontal members, and two semi-cylindrical packaging materials 93 approximating multiple bundled support columns. Note that the actual package, which consists of the same ordered materials as the predicted package in predicted image 70, includes packaging materials 91-93 as shown in Figure 8.

[0031] Referring again to Figures 1-3, the first acquisition unit 51A associates the order ID, user information, ordered material information (type and quantity of each ordered material), and predicted image with each other and stores them as order information in the order information table (step S23). As shown in Figure 9, the order information table is composed of the associated order ID, user information, ordered material information, and predicted image as one record. The predicted image data may also include correspondence relationship data showing the correspondence between the images of each packaging material that make up the predicted image and the types and quantities of one or more ordered materials that make up that packaging material (breakdown of packaging materials).

[0032] Referring again to Figures 1-3, after step S23, the first acquisition unit 51A returns the above order information to the user terminal UT as confirmation of receipt of the order received (step S24). At this time, the order ID and user information are encoded into various codes such as barcodes or two-dimensional codes and sent to the user terminal UT as an entry pass.

[0033] After receiving order information via the user terminal UT, the user heads to the warehouse WH with the empty truck TR1 (step S13). The user then presents the entry pass (more specifically, the order ID and user information code displayed on the user terminal UT) included in the order information received by the user terminal UT to the entry management device 31 located near the entrance of the warehouse WH (step S14). The entry management device 31, which includes a code reader, reads the presented entry pass and obtains the order ID and user information. The outbound management system 10 then performs entry processing based on the order ID and user information (step S25).

[0034] In the entry process in step S25, the entry management device 31 first transmits the order ID and user information to the outbound management device 50. The order ID and user information are obtained by the entry permission unit 51C of the outbound management device 50. The entry permission unit 51C determines whether the obtained combination of order ID and user information is registered in the order information table. If it is registered, it determines that the truck TR can enter; if it is not registered, it determines that the truck TR cannot enter. The entry permission unit 51C supplies the determination result to the entry management device 31 and presents the determination result to the user. The entry management device 31 controls the drive unit 33 and opens the entry gate 32 only if the determination result is that entry is possible. This allows the truck TR to enter the warehouse WH.

[0035] The truck TR enters the warehouse WH and stops in area R1 (Figure 1) near the entrance (step S15). The stopping of the truck TR at the predetermined position is detected by a sensor (not shown) or the like. Upon this detection, cameras 34A to 34D each take an image of the truck TR (step S26). The captured images are transmitted to the outbound management device 50. These four images are of the truck TR with its cargo bed TR1 empty, taken from different directions. The images transmitted to the outbound management device 50 are acquired as entry images by the second acquisition unit 51B via the network NW and communication interface 54.

[0036] The completion of the above imaging is notified to the user by any means (e.g., voice guidance). Upon receiving this notification, the user moves the truck TR and sequentially loads the ordered building materials, which are the building materials for the current order, onto the truck TR from among the multiple building materials stored in storage locations P1 to P8 within the warehouse WH (step S16). At this time, the user terminal UT may, under the user's operation, display the predicted image included in the order information received above on the display device of the user terminal UT. The predicted image displayed on the user terminal UT may include callouts F1 to F3 that add the breakdown (type and quantity) of the ordered building materials for each of the three-dimensional shapes constituting the predicted packaging, as shown in Figure 10. The user terminal UT may receive image data capable of displaying such a predicted image. Callouts F1 to F3 are displayed using the above correspondence data. By referring to such a predicted image, the user can easily load the ordered building materials onto the truck TR.

[0037] Referring again to Figures 1-3, the user stops the truck TR in area R2 (Figure 1) near the exit of the warehouse WH after the loading of the ordered materials is complete (step S17). The user then presents an entry pass (order ID and user information code) to the exit management device 41 located near the exit of the warehouse WH (step S17). The exit management device 41, which includes a code reader, reads the presented entry pass and obtains the order ID and user information (step S27). Furthermore, the exit management device 41 uses cameras 44A-44D to image the truck TR with the ordered materials actually loaded (step S27). The four captured images, along with the order ID and user information, are transmitted to the outbound management device 50. The captured images transmitted to the outbound management device 50 are acquired as outbound images by the second acquisition unit 51B via the network NW and communication interface 54. The order ID and user information transmitted to the outbound management device 50 are acquired by the first acquisition unit 51A. The first acquisition unit 51A refers to the predicted image table based on the order ID and user information and acquires the corresponding predicted image.

[0038] Through the above process, the first acquisition unit 51A acquires a predicted image. The second acquisition unit 51B acquires the image captured at the time of entry and the image captured at the time of exit.

[0039] After acquiring each image, the exit permission unit 51D performs exit processing based on the predicted image acquired by the first acquisition unit 51A and the exit image acquired by the second acquisition unit 51B (step S28). In the exit processing, the exit permission unit 51D, based on the predicted image and the exit image, grants permission for the truck TR to exit if the similarity between the predicted cargo shape in the predicted image and the actual cargo shape in the exit image is higher than a predetermined standard. A specific example of this process is described below.

[0040] The exit permission unit 51D first compares the four exit images with the four entry images acquired when the truck TR entered. In the entry images, the cargo bed TR1 is empty, while in the exit images, the ordered materials are loaded onto the cargo bed TR1. Therefore, the exit permission unit 51D compares the corresponding exit and entry images to identify the portion that is commonly visible in the images being compared as the truck portion shown in those images. Since the position of the cargo relative to the truck TR is roughly fixed, the exit permission unit 51D identifies the image with the background removed that is in a predetermined positional relationship with the identified truck portion in the exit images as the cargo image showing the actual cargo configuration. Background removal may be performed before or during the comparison of the exit and entry images. If one of the exit images is the image shown in Figure 8, the portion indicated by reference numerals 91 to 93 becomes the cargo image. Furthermore, if the area containing the actual package configuration can be identified from the exit image alone, the entry image is unnecessary. For example, the exit permission unit 51D can identify the truck portion and background from the exit image using image recognition, and then identify the image excluding these parts from the exit image as the package configuration image.

[0041] Subsequently, the exit permission unit 51D converts the previously identified package image into a two-dimensional image represented by the contours of the subjects (packaging materials) depicted in the package image, using image processing such as contour detection. This two-dimensional image consists of, for example, image G2 shown by the solid line in Figure 11. Images G1 and G3 shown by the dashed lines in Figure 11 represent the contours of the driver's cab and cargo bed TR1 of the truck TR, respectively. At least one of these two contours may be included in the two-dimensional image. In this case, the package image also includes the corresponding driver's cab or cargo bed TR12. Correspondingly, the prediction image may also include an image representing the driver's cab and / or cargo bed TR12 (for example, a simply rendered graphic image). As shown in Figure 11, the two-dimensional image can also be described as an image that represents the actual package as depicted in the package image as a three-dimensional graphic for each packaging unit (i.e., each piece of packaging material). The exit permission unit 51D compares the four two-dimensional images converted in this way with the prediction image, which is a three-dimensional image. For example, the exit permission unit 51D generates a 2D image from the 3D predicted image, viewed from the same direction as the imaging direction of the cameras 44A to 44D relative to the truck TR. The exit permission unit 51D compares the four 2D images based on the captured image and the four 2D images based on the predicted image, comparing images that are viewed from the same direction as the imaging direction, and derives four similarity values. If all four similarity values ​​are higher than a predetermined threshold, the exit permission unit 51D determines that the similarity between the actual cargo configuration captured in the exit image and the predicted cargo configuration represented by the predicted image is higher than a predetermined standard. Here, the cargo configuration of construction materials is almost determined if the composition of the construction materials is the same. Therefore, if the similarity between the actual cargo configuration and the predicted cargo configuration is high, the actual cargo configuration can be said to be the configuration when multiple ordered materials are properly loaded onto the truck. For this reason, if the similarity between the actual cargo configuration and the predicted cargo configuration is high, it can be said that there are no errors in loading the construction materials.

[0042] If the similarity between the predicted and actual cargo configurations is not higher than a predetermined standard, the exit permission unit 51D displays a message on the exit management device 41 indicating that there is a loading error and informs the user of this. In this case, the exit gate 42 will not open. As a result, the construction materials will be rearranged. If the similarity is higher than a predetermined standard, the exit permission unit 51D permits the truck TR to exit as there is no loading error. In this case, the exit permission unit 51D transmits an exit path, which is an encoded message granting permission to exit, to the user terminal UT. The exit permission unit 51D may also display a message on the exit management device 41 indicating that the truck is permitted to exit and informs the user of this.

[0043] The user displays the exit pass on the user terminal UT and presents the exit pass to the exit management device 41 (step S18). The exit management device 41, which includes a code reader, reads the presented exit pass. When the exit management device 41 reads the exit pass, it determines that the truck TR is permitted to exit and controls the drive unit 43 to open the exit gate 42 (step S29). As a result, the truck TR exits the warehouse WH (step S18).

[0044] As described above, the outbound management device 50 according to this embodiment includes a first acquisition unit 51A, a second acquisition unit 51B, and an outbound permission unit 51D. The first acquisition unit 51A acquires a predicted image representing the predicted loading configuration of multiple building materials (ordered materials) that are to be loaded onto a truck TR in the warehouse WH and then dispatched from the warehouse WH. Furthermore, the second acquisition unit 51B acquires an image capturing the actual loading configuration of the multiple building materials that have actually been loaded onto the truck TR. The outbound permission unit 51D then performs permission processing to allow the truck TR loaded with multiple building materials to leave the warehouse WH if the similarity between the predicted loading configuration based on the predicted image and the image capture and the actual loading configuration is higher than a predetermined standard.Here, if the similarity between the predicted loading configuration and the actual loading configuration in both the predicted image and the image capture is high, it is highly likely that the correct building materials have been loaded onto the truck TR.In other words, it is unlikely that building materials different from the ordered materials ordered by the user have been loaded onto the truck TR. Thus, with the above configuration, the possibility that a mistakenly loaded construction material is included among the multiple construction materials loaded onto a truck (TR) permitted to leave the warehouse (WH) is reduced.

[0045] Furthermore, if the similarity is below the predetermined standard, the exit permit unit 51D will perform a denial process, rather than the permit process, which denies the truck TR's exit from the warehouse WH. This prevents truck TR from releasing construction materials different from the ordered materials from the warehouse.

[0046] Furthermore, the exit authorization unit 51D, as part of the authorization process, transmits an exit pass to the user terminal UT of a user who is departing multiple construction materials by truck TR, which is an exit pass that authorizes the departure and opens the exit gate 42 of the warehouse WH. As a result, the departure of the truck TR is permitted on the condition that the exit pass is transmitted. The exit gate may be a manually operated door, the exit management device 41 may control the locking / unlocking of the exit gate, and the exit pass may be a pass that unlocks the exit gate when read by the exit management device 41. The specific act of opening the exit gate 42 includes unlocking such an exit gate.

[0047] As another modification, the exit permission unit 51D may, as at least part of the permission process, directly control the drive unit for the exit gate or the device for unlocking / locking the exit gate to open the exit gate. This allows the exit gate to be opened with a simple process.

[0048] In the above embodiment, the exit permission unit 51D derives a similarity score by image processing based on the predicted image and the captured image, and performs the permission process if the derived similarity score is higher than a predetermined standard. This makes the permission process automatic. The method for deriving the similarity score by image processing based on the predicted image and the captured image is arbitrary. Template matching using the predicted image as a template image may be used. The captured image may be subjected to predetermined processing as described above to facilitate comparison with the predicted image. Predetermined processing includes, for example, extracting the contour lines of the actual package captured in the captured image by comparing them with the lines of the three-dimensional shape in the predicted image. The similarity score may also be derived by a machine learning model that takes the predicted image and the captured image as input and outputs the similarity score between the two. This derivation by model is also one form of deriving the similarity score by image processing. Machine learning is performed using multiple training data including the predicted image, the captured image, and the similarity score as the correct answer.

[0049] Furthermore, the exit permit unit 51D identifies the actual cargo configuration captured in the exit image based on the entry image taken by the cargo bed TR1 of an empty truck TR, and derives the similarity score based on the identified actual cargo configuration and the predicted image. This allows for accurate identification of the actual cargo configuration and improves the accuracy of the similarity score derivation.

[0050] As a variation of the above embodiment, the exit permission unit 51D may present the predicted image and the captured image to the operator by displaying them, for example, on the display device of the user interface 55. Here, the predicted image is preferably a predicted image that allows for the identification of the breakdown of packaging materials, as shown in Figure 10. The operator compares the predicted image and the captured image to visually determine whether the correct building materials are loaded on the truck TR, and if correct, inputs that fact (which can also be said to be that the similarity is higher than the predetermined standard) into the user interface 55. The input of this fact is then input into the exit permission unit 51D. When this information is input, the exit permission unit 51D may determine that the similarity is higher than the predetermined standard and perform the permission process. With this configuration as well, the exit permission unit 51D can perform the permission process when the similarity between the predicted packaging based on the predicted image and the captured image and the actual packaging is higher than the predetermined standard, thereby reducing the possibility that incorrectly loaded building materials are included among the multiple building materials loaded on the truck TR that is permitted to exit from the warehouse WH.

[0051] Furthermore, in the above embodiment, the predicted packaging is a packaging that approximates multiple building materials as three-dimensional figures for each packaging unit (each packaging material). The exit permit unit 51D generates an image (for example, image G2 in Figure 11) that represents the multiple building materials captured in the image as three-dimensional figures for each packaging unit, based on the captured image, and derives the similarity between the three-dimensional figures represented by the generated image and the three-dimensional figures of the predicted packaging (Figure 7) as the similarity between the predicted packaging and the actual packaging. With this configuration, the predicted image representing the predicted packaging is simplified, the amount of data in the predicted image is reduced, and the processing load using the predicted image is reduced.

[0052] The specific form and generation method of the predicted image are arbitrary. As another example of such generation, the predicted image may be an image that combines the images of each of the multiple building materials that constitute the predicted packaging. In such a case, a three-dimensional image of the building material is prepared for each type of building material (material name and product code), and the predicted image is generated by combining these three-dimensional images. In such a case, the exit permit unit 51D may derive the similarity score by matching the predicted image with the captured image. With the above configuration, the accuracy of the similarity score is improved compared to when the three-dimensional figure is used.

[0053] The predicted images may be manually generated by a designated operator while referring to the type and quantity of each ordered material in the ordered material information. Alternatively, they may be generated by a machine learning-generated prediction model that takes the ordered material information and truck information as input and outputs the predicted images.

[0054] Furthermore, in the above, the predicted image is a three-dimensional image, and the captured images include the first to nth captured images of the actual package configuration taken from different first to nth directions (where N is an integer of 2 or more). The exit permission unit 51D compares the three-dimensional image with the first to nth captured images to derive the similarity between the predicted package configuration and the actual package configuration. The predicted image may be a two-dimensional image, and the captured images may consist only of images taken from one direction, but it is preferable to use a three-dimensional image for the predicted image and the first to nth captured images. This reduces blind spots during imaging and improves the accuracy of deriving the similarity.

[0055] Furthermore, as described above, the first acquisition unit 51A acquires order material information specified by the user when ordering multiple building materials, including the type of building material to be ordered and the quantity of each type of building material. In addition, the first acquisition unit 51A acquires a predicted image by generating a predicted image representing the predicted packaging based on the acquired order material information. Thus, the predicted image is generated automatically. Loading multiple building materials onto a truck TR properly requires know-how, but the automatic generation of the predicted image makes it easy to obtain a predicted image even without know-how.

[0056] The first acquisition unit 51A may acquire the dimensions of the truck bed TR1 and generate a predictive image based on the acquired dimensions and material specification information. This generates a predictive image with good accuracy that takes into account the dimensions of the truck bed TR1. The first acquisition unit 51A may also generate a predictive image without acquiring the dimensions of the truck bed TR1. In such a case, multiple predictive images based on multiple truck bed TR1s with typical dimensions may be generated and selected as appropriate by the user.

[0057] In the above configuration, information (C) to (I) above) is set for each type of building material to be considered when loading it onto the truck bed TR1, and the first acquisition unit 51A generates a predictive image by referring to the above information. Note that the above information may consist of, for example, at least one of (C) to (I) above. For example, the information may include at least one of the following: the dimensions and weight of the building material to be set (C) above, the loading order indicating the loading order of the building material to be set (E) above, the packaging type specifying the packaging method of the building material to be set (D), (F), (H), (I) above, and the loading method in which the building material to be set can be loaded (G) above). With such a configuration, a predictive image can be generated with high accuracy.

[0058] Furthermore, as described above, when the first acquisition unit 51A acquires order material information, it sends an entry pass to the user terminal UT of a user who is dispatching multiple construction materials by truck TR, allowing them to enter the warehouse WH. This allows entry to the warehouse WH when an order for construction materials is placed.

[0059] In the above configuration, the first acquisition unit 51A transmits a predicted image to the user terminal UT of a user who is retrieving multiple construction materials from a truck TR. This allows the user to refer to the predicted image when loading the construction materials, thereby reducing the possibility of the user making loading errors.

[0060] Furthermore, the predicted image transmitted to the user terminal UT should be configured to further show the breakdown of building materials (each type of building material and the quantity of each type) for each of the packaging materials that make up the predicted packaging, each of which is composed of one or more building materials (see, for example, Figure 10). This further reduces the possibility of the user making a mistake in loading the building materials.

[0061] (modified version) The present invention is not limited to the embodiments described above. Modifications of the above embodiments will be described below.

[0062] The hardware configuration of each of the above parts 51A to 51D is arbitrary. For example, the entry permission unit 51C may be provided in the entry management device 31. The outbound management device 50 may be composed of multiple computers that can communicate via a network or the like. Each of the above parts 51A to 51D may be distributed and provided in multiple devices. For example, the first acquisition unit 51A and the exit permission unit 51D may be provided in different devices. Also, the exit permission unit 51D may be distributed and provided in multiple devices. For this reason, the above processor 51 that operates as each of the above parts 51A to 51D may be composed of one or more IC chips. Examples of IC chips include CPUs, ASICs (Application Specific Integrated Circuits), FPGAs (Field-Programmable Gate Arrays), DSPs (Digital Signal Processing), microcomputers, and GPUs (Graphics Processing Units). The warehouse may be an indoor warehouse or an outdoor warehouse, for example, a facility without a roof where the storage area for building materials is surrounded by a fence.

[0063] (Scope of the present invention) The present invention is not limited to the embodiments described above. For example, the present invention includes various modifications to the above embodiments and modifications that can be understood by those skilled in the art within the scope of the technical concept of the present invention. The configurations listed in the above embodiments and modifications can be combined as appropriate to the extent that they do not contradict each other. It is also possible to delete any of the above configurations.

[0064] (Note) The following examples illustrate configurations based on the above embodiments and modifications. Any partial configuration of the above embodiments and modifications may be applied to each appendix. Furthermore, parts of each appendix may be combined. (Note 1) A first acquisition unit acquires a predictive image representing the predicted loading configuration of multiple construction materials that are to be loaded onto trucks in the warehouse and to be shipped out of the warehouse, A second acquisition unit acquires images of the actual packaging of multiple construction materials actually loaded onto the truck, A departure permit unit that, when the similarity between the predicted packaging based on the predicted image and the captured image and the actual packaging is higher than a predetermined standard, performs a permit process to allow the truck loaded with the plurality of construction materials to leave the warehouse, A construction materials inventory management device equipped with [specific features / features]. (Note 2) The exit permit unit, when the similarity is below the predetermined standard, performs a denial process instead of the permit process, which denies the truck's exit from the warehouse. The inventory management device described in Appendix 1. (Note 3) The authorization process includes a process of transmitting an exit pass to a terminal of a user who is departing the plurality of construction materials by truck, which authorizes the departure, and which opens the exit gate of the warehouse. The inventory management device described in Appendix 1 or 2. (Note 4) The aforementioned authorization process includes the process of opening the exit gate of the warehouse. The inventory management device described in any of the appendices 1 to 3. (Note 5) The entry permit unit derives the similarity score by image processing based on the predicted image and the captured image, and performs the permit process if the derived similarity score is higher than the predetermined standard. The inventory management device described in any of the appendices 1 to 4. (Note 6) The exit permit unit identifies the actual cargo configuration as captured in the image obtained by the second acquisition unit, based on the image of the truck with an empty cargo bed, and derives the similarity based on the identified actual cargo configuration and the predicted image. The inventory management device described in Appendix 5. (Note 7) The exit permission unit presents the predicted image and the captured image to the operator, and when the operator inputs information to the exit permission unit indicating that the similarity is higher than the predetermined standard, it determines that the similarity is higher than the predetermined standard and performs the permission process. The inventory management device described in any of the appendices 1 to 6. (Note 8) The predicted packaging is a packaging that approximates the multiple building materials into a three-dimensional shape for each packaging unit. The exit permit unit generates an image representing the multiple building materials captured in the image as three-dimensional figures for each packaging unit, based on the captured image, and derives the degree of similarity between the three-dimensional figures represented in the generated image and the three-dimensional figures of the predicted packaging as the degree of similarity between the predicted packaging and the actual packaging. The inventory management device described in any of the appendices 1 to 7. (Note 9) The predicted packaging is composed of a combination of images of each of the multiple building materials. The aforementioned entry permit unit derives the similarity by matching the predicted image with the captured image. The inventory management device described in any of the appendices 1 to 8. (Note 10) The aforementioned predicted image is a three-dimensional image. The captured image includes the first to nth captured images taken from different first to nth directions of the actual packaging (where N is an integer of 2 or more), The aforementioned entry permit unit derives the similarity by comparing the three-dimensional image with the first to nth captured images. The inventory management device described in any of the appendices 1 to 9. (Note 11) The first acquisition unit is, When ordering the aforementioned multiple building materials, the system obtains order material information specified by the user, which includes the type of building material to be ordered and the quantity of each type of building material. The predicted image is obtained by generating the predicted image representing the predicted packaging based on the acquired order material information. The inventory management device described in any of the appendices 1 to 10. (Note 12) The first acquisition unit acquires the dimensions of the truck bed and generates the predicted image based on the acquired dimensions and the material specification information. The inventory management device described in Appendix 11. (Note 13) Each of the aforementioned building materials has information set for it that is taken into consideration when loading it onto the truck bed. The first acquisition unit generates the predicted image by referring to the information, The information includes at least one of the following: the dimensions and weight of the building materials to be specified; the loading order indicating the order in which the building materials to be specified are loaded; the packaging configuration specifying the packaging configuration of the building materials to be specified; and the loading configurations in which the building materials to be specified can be loaded. The inventory management device described in Appendix 11 or 12. (Note 14) When the first acquisition unit acquires the ordered materials information, it transmits an entry pass to the terminal of the user who is departing the multiple construction materials by truck, enabling them to enter the warehouse. The inventory management device described in any of the appendices 11 to 13. (Note 15) The first acquisition unit transmits the predicted image to the terminal of the user who is retrieving the multiple building materials by truck. The inventory management device described in any of the appendices 1 to 14. (Note 16) The predicted image transmitted to the terminal further shows the breakdown of the building materials in each of the packaging materials that constitute the predicted packaging, each of which is composed of one or more building materials packed together. The inventory management device described in Appendix 14 or 15. (Note 17) A program that, when executed by a computer, causes the computer to function as a stock dispatch management device as described in any of the appendices 1 to 16. [Explanation of Symbols]

[0065] 10...Outbound management system, 31...Entry management device, 32...Entry gate, 33...Drive unit, 34,34A~34D...Camera, 41...Exit management device, 42...Exit gate, 43...Drive unit, 44,44A~44D...Camera, 50...Outbound management device, 51...Processor, 51A...First acquisition unit, 51B...Second acquisition unit, 51C...Entry permission unit, 51D...Exit permission unit, 52...Storage, 54...Communication interface, 5 5...User interface, 70...Predictive image, 91-93...Packaging materials, 100...Packaging materials, 101-102...Support column, 111-112...Packaging materials, 121-122...Packaging materials, 131, 131A-131B...Packaging materials, F1-F3...Blowout, G1-G3...Image, NW...Network, P1-P8...Storage location, R1...Area, R2...Area, TR...Truck, TR1...Loading bed, UT...User terminal, WH...Warehouse.

Claims

1. A first acquisition unit acquires a predictive image representing the predicted loading configuration of multiple construction materials that are to be loaded onto trucks in the warehouse and to be shipped out of the warehouse, A second acquisition unit acquires images of the actual packaging of multiple construction materials actually loaded onto the truck, A departure permit unit that, when the similarity between the predicted packaging based on the predicted image and the captured image and the actual packaging is higher than a predetermined standard, performs a permit process to allow the truck loaded with the plurality of construction materials to leave the warehouse, A construction materials inventory management device equipped with [specific features / features].

2. The exit permit unit, when the similarity is below the predetermined standard, performs a denial process instead of the permit process, which denies the truck's exit from the warehouse. The inventory management device according to claim 1.

3. The authorization process includes a process of transmitting an exit pass to a terminal of a user who is departing the plurality of construction materials by truck, which authorizes the departure, and which opens the exit gate of the warehouse. The inventory management device according to claim 1.

4. The aforementioned authorization process includes the process of opening the exit gate of the warehouse. The inventory management device according to claim 1.

5. The entry permit unit derives the similarity score by image processing based on the predicted image and the captured image, and performs the permit process if the derived similarity score is higher than the predetermined standard. The inventory management device according to claim 1.

6. The exit permit unit identifies the actual cargo configuration as captured in the image obtained by the second acquisition unit, based on the image of the truck with an empty cargo bed, and derives the similarity based on the identified actual cargo configuration and the predicted image. The inventory management device according to claim 5.

7. The exit permission unit presents the predicted image and the captured image to the operator, and when the operator inputs information to the exit permission unit indicating that the similarity is higher than the predetermined standard, it determines that the similarity is higher than the predetermined standard and performs the permission process. The inventory management device according to claim 1.

8. The predicted packaging is a packaging that approximates the multiple building materials into a three-dimensional shape for each packaging unit. The exit permit unit generates an image representing the multiple building materials captured in the image as three-dimensional figures for each packaging unit, based on the captured image, and derives the degree of similarity between the three-dimensional figures represented in the generated image and the three-dimensional figures of the predicted packaging as the degree of similarity between the predicted packaging and the actual packaging. The inventory management device according to claim 1.

9. The predicted packaging is composed of a combination of images of each of the multiple building materials. The aforementioned entry permit unit derives the similarity by matching the predicted image with the captured image. The inventory management device according to claim 1.

10. The aforementioned predicted image is a three-dimensional image. The captured image includes first to n captured images taken from different first to n directions of the actual packaging (wherein N is an integer of 2 or more). The aforementioned entry permit unit derives the similarity by comparing the three-dimensional image with the first to the Nth captured images. The inventory management device according to claim 1.

11. The first acquisition unit is, When ordering the aforementioned multiple building materials, the system obtains order material information specified by the user, which includes the type of building material to be ordered and the quantity of each type of building material. The predicted image is obtained by generating the predicted image representing the predicted packaging based on the acquired order material information. The inventory management device according to claim 1.

12. The first acquisition unit acquires the dimensions of the truck bed and generates the predicted image based on the acquired dimensions and the material specification information. The inventory management device according to claim 11.

13. Each of the aforementioned building materials has information set for it that is taken into consideration when loading it onto the truck bed. The first acquisition unit generates the predicted image by referring to the information, The information includes at least one of the following: the dimensions and weight of the building materials to be specified; the loading order indicating the order in which the building materials to be specified are loaded; the packaging configuration specifying the packaging method of the building materials to be specified; and the loading configurations in which the building materials to be specified can be loaded. The inventory management device according to claim 11.

14. When the first acquisition unit acquires the ordered material information, it transmits an entry pass to the terminal of the user who is departing the multiple construction materials by truck, enabling them to enter the warehouse. The inventory management device according to claim 11.

15. The first acquisition unit transmits the predicted image to the terminal of the user who is retrieving the multiple building materials by truck. The inventory management device according to claim 1.

16. The predicted image transmitted to the terminal further shows the breakdown of the building materials in each of the packaging materials that constitute the predicted packaging, each of which is composed of one or more building materials packed together. The inventory management device according to claim 15.

17. A program that, when executed by a computer, causes the computer to function as the inventory management device described in claim 1.

Images