JP2022066820A - Delivery instruction setting method in sorting device and delivery instruction system - Google Patents

Abstract

To increase the efficiency of setting a delivery instruction of high delivery performance in a sorting device.SOLUTION: In a method for setting a delivery instruction including delivery time of each of cargoes in a sorting device, the cargoes have been classified into groups in advance, initial delivery time corresponding to each cargo is derived by adding, to a predetermined reference time for each of the cargoes, each cargo addition time within a predetermined numerical range, and shifting time for which a value varies from one group to another is set, and an initial value for setting the delivery instruction is generated. the cargoes classified into the same group are continuously delivered, and the delivery instruction is generated from the initial value by using an optimization algorithm so that delivery performance for sequentially delivering the cargoes for each group can be equal to or higher than predetermined reference performance.SELECTED DRAWING: Figure 5

Description

The present disclosure relates to a method of setting a delivery instruction in a sorting device and a delivery instruction system.

Conventionally, various configurations are known as a device for loading and unloading a load. For example, in Patent Document 1, in an automated warehouse provided with a plurality of stages of luggage racks, when an empty space is created in the luggage rack due to shipping, the load is rearranged so as to fill the empty space, and the utilization rate of the storage space is obtained. The configuration for improving the above is disclosed. Further, in Patent Document 2, in an automated warehouse provided with a rotary rack, a forwarding direction pattern and a load selection pattern of the tray row are selected so that the forwarding distance of the tray row is the shortest in response to a series of delivery requests. , A configuration that enhances delivery efficiency is disclosed. Further, in Patent Document 3, when the group of articles to be transported is sorted by group, the article unit is divided by the sorting destination so that the bias of the transport amount does not become large in the chute to which the sorting destination corresponding to the group is assigned. Disclosed is a configuration in which a cutting unit is formed and an article is transported for each cutting unit.

Japanese Unexamined Patent Publication No. 2005-138955 Japanese Unexamined Patent Publication No. 2014-51391 Japanese Unexamined Patent Publication No. 2008-150191

When a plurality of warehousing packages are sorted by group and delivered in the sorted order, it is necessary to change the order of the warehousing packages in the sorting device so as to be in the order of the group. In such a sorting device, the sorting efficiency at the time of unloading the cargo in the order of the group changes depending on the order and timing of issuing the warehousing instruction for each of the plurality of warehousing. In order to improve the efficiency of warehousing, it is sufficient to set the warehousing instruction including the warehousing time in advance so that the warehousing efficiency becomes higher. The amount of calculation for setting instructions can be enormous. Therefore, there has been a demand for a technique for improving the efficiency of setting a delivery instruction having high delivery performance.

The present disclosure can be realized in the following forms.
(1) One form of the present disclosure is a warehousing instruction for setting a warehousing instruction including a warehousing time of each of the said wares in a sorting device for sorting and warehousing by changing the order of a plurality of warehousing. In the setting method, the plurality of loads are classified into a plurality of groups in advance, and for each of the above-mentioned loads, addition by cargo within a predetermined numerical range with respect to a predetermined reference time. By adding the time and the shift time for which different values are set for each group, the initial delivery time corresponding to each of the shipments is derived, and the initial value for setting the delivery instruction is generated. Then, an optimization algorithm is used so that the shipping performance of the cargoes classified into the same group is continuously delivered and the shipping performance of the plurality of cargoes being sequentially delivered for each group is equal to or higher than the predetermined standard performance. It is used to generate the shipping instruction from the initial value.
According to the setting method of the delivery instruction in this form, the load addition time within the predetermined numerical range and the shift time with different values set for each group are set with respect to the predetermined reference time. Since the initial delivery time is derived by addition to generate the initial value and the delivery instruction is generated from the initial value using the optimization algorithm, the effect of increasing the efficiency of setting the delivery instruction with high delivery performance can be obtained.
(2) In the method of setting the delivery instruction in the above form, the order of delivery is predetermined for the plurality of groups, and the shift time is set longer as the order of delivery is later. May be. According to this form of the delivery instruction setting method, the efficiency of setting the delivery instruction with high delivery performance can be further improved.
(3) In the method of setting the delivery instruction in the above embodiment, in the initial value, the initial delivery time of the load having the latest initial delivery time among the packages belonging to the first group among the plurality of groups is the above. Among the shipments belonging to the second group to be delivered next to the first group, the initial delivery time may be set earlier than the initial delivery time of the earliest shipment. According to this form of the delivery instruction setting method, the efficiency of setting the delivery instruction with high delivery performance can be further improved.
(4) In the method of setting the delivery instruction in the above-described embodiment, the optimization algorithm is a genetic algorithm, and the load-specific addition time is within a numerical range limited to the number of the packages included in each group. It may be derived at random. According to this form of the delivery instruction setting method, inconveniences such as initial convergence can be suppressed, and optimization of the delivery instruction becomes easy.
(5) In the method of setting the delivery instruction in the above-described embodiment, the reference performance belongs to different groups among the packages belonging to the same group that are continuously delivered in the order of delivery of the plurality of packages. It is also possible that the number of deliveries per unit time is set to be equal to or greater than a predetermined standard number of deliveries without the load being mixed. According to this form of the delivery instruction setting method, it is possible to improve the delivery efficiency while suppressing the delivery in which the group order does not match.
(6) In the method of setting the delivery instruction in the above-described embodiment, the initial value includes the initial delivery time corresponding to the individual load and the initial information of the information indicating the storage location of the individual load in the sorting device. As the initial information of the information indicating the storage location including the storage location, the storage location determined load for which the storage location in the sorting device has already been determined when the initial value is generated indicates the determined storage location. For a storage location undecided load for which the storage location in the sorting device has not been determined when the information is used to generate the initial value, the individual storage location has not been determined prior to the storage of the storage location undecided load. Using the information tentatively set for the load, using the optimization algorithm, the delivery instruction includes information indicating the storage location of the undetermined storage location in addition to the delivery time of each individual load. May be generated. According to this form of the delivery instruction setting method, even if there are a load whose storage location has been determined and a load whose storage location has not been determined when the operation for setting the delivery instruction is started, it is highly efficient and appropriate. It is possible to set various shipping instructions.
(7) In the method of setting the delivery instruction in the above-described embodiment, the sorting device holds a plurality of storage shelves capable of storing the plurality of loads and the load stored in any position of the plurality of storage shelves. At the delivery time specified in the delivery instruction corresponding to the load, the take-out unit to be taken out from the storage shelf and the load to be sequentially taken out from the plurality of storage shelves by the take-out unit are sequentially taken out by the take-out unit according to a predetermined discharge algorithm. The operation of generating the shipping instruction from the initial value is provided with a shipping unit that sequentially discharges from the sorting device, and the shipping performance is the reference performance when the shipping unit discharges the load according to the discharging algorithm. It may be performed so as to be as described above. According to this form of the delivery instruction setting method, the delivery instruction can be set according to the discharge algorithm of the delivery unit to improve the delivery performance.
The present disclosure can be realized in various forms other than the above. For example, a delivery instruction system for setting a delivery instruction, a delivery instruction setting algorithm used in a sorting device, and a delivery instruction setting algorithm are used to give a delivery instruction. It can be realized in the form of a computer program that causes a computer to execute the process of setting the computer, a non-temporary recording medium on which the computer program is recorded, or the like.

An explanatory diagram showing a schematic configuration of a delivery instruction system. An explanatory diagram schematically showing the outline of the internal configuration of the elevator on the delivery side. Explanatory drawing which showed the control part by the functional block. A flowchart showing a delivery instruction setting processing routine. Explanatory diagram showing the range of initial shipping times generated for each container. Explanatory drawing which shows an example of the initial value of a delivery time. An explanatory diagram showing the relationship between the number of optimization generations and the objective function. Explanatory diagram showing the initial delivery time range generated for each container in the comparative example. Explanatory drawing which shows an example of the initial value of the delivery time of a comparative example. An explanatory diagram showing the relationship between the shift time and the container sorting amount. The explanatory view which showed the control part of 2nd Embodiment by the functional block. A flowchart showing a delivery instruction setting processing routine. The explanatory view which showed the control part of 3rd Embodiment by the functional block. A flowchart showing a delivery instruction setting processing routine.

A. First Embodiment:
(A-1) Configuration of sorting device:
FIG. 1 is an explanatory diagram showing a schematic configuration of the delivery instruction system 10 of the first embodiment. The delivery instruction system 10 includes a sorting device 20 and a control unit 60 that controls the sorting operation in the sorting device 20. In addition, FIG. 1 and FIG. 2 described later show XYZ axes orthogonal to each other in order to specify the transport direction in the sorting device 20. The X-axis, Y-axis, and Z-axis shown in each figure represent the same orientation. In the present specification, the Z axis indicates the vertical direction, and the X axis and the Y axis indicate the horizontal direction.

The sorting device 20 is a device in which a plurality of containers 22 classified in advance into a plurality of groups are stored, and the plurality of stored containers 22 are sorted by group and sequentially delivered to each group. Specifically, the order of the containers 22 that have been received in a state where the containers 22 that belong to different groups coexist is changed, the containers 22 that belong to the same group are continuously delivered, and the containers are placed in the order of the predetermined group. It is a device to issue 22. The sorting device 20 is installed in a distribution center as a distribution base, for example, and stores a plurality of containers 22 having different delivery destinations in the order received by the distribution center, sorts them by delivery destination, and has the same delivery destination. It can be used to ship the containers 22 together. At this time, the delivery destination corresponds to the above-mentioned group. The container 22 is also referred to as a "load".

The sorting device 20 includes a warehousing conveyor 32, a warehousing side elevator 34, storage shelves 40a to 40d, a warehousing side elevator 50, and a warehousing conveyor 51. The warehousing conveyor 32 is a device that sequentially conveys the container 22 sent from the previous process, for example, the container 22 carried into the distribution center in which the sorting device 20 is installed, toward the inside of the sorting device 20.

In the present embodiment, the warehousing conveyor 32 is provided with a group identification unit 30. The group identification unit 30 identifies the groups to which the individual containers 22 conveyed by the warehousing conveyor 32 belong, in the order of passing through the group identification unit 30. The group identification unit 30 may be a device for identifying the group to which each container 22 belongs by reading the label attached to each container 22 to indicate the delivery destination of each container 22, for example. The group identification unit 30 may have a different configuration as long as the group of each container 22 can be identified. For example, the worker reads the delivery destination of each container 22 and the worker inputs the read information from the input device. May be. The information identified by the group identification unit 30 is transmitted to the control unit 60.

The warehousing-side elevator 34 is provided with an elevating mechanism, and the container 22 conveyed by the warehousing conveyor 32 is conveyed in the vertical direction (Z-axis direction) and distributed to any of the storage shelves 40a to 40d. In the present embodiment, the warehousing-side elevator 34 distributes the container 22 based on the warehousing instruction transmitted from the control unit 60. The "warehousing instruction" transmitted from the control unit 60 is information indicating a storage location in the sorting device 20 of each container 22, and for each container 22, it is placed in any of the storage shelves 40a to 40d. Contains information that indicates whether it should be distributed.

The storage shelves 40a to 40d have a structure for temporarily storing the container 22 stored in the sorting device 20. The storage shelves 40a to 40d have a four-story structure in which they are stacked vertically upward (in the + Z direction) in this order. The number of storage shelves may be appropriately set according to the number of containers 22 to be sorted. Hereinafter, the structure of the storage shelves 40d on the fourth floor will be described with reference to FIG. 1, but each of the storage shelves 40a to 40d has the same structure.

The storage shelf 40d includes a left shelf 42 provided on the −Y direction side, a right shelf 44 provided on the + Y direction side, and a discharge conveyor 46 provided between the left shelf 42 and the right shelf 44. Be prepared. The left shelf 42, the right shelf 44, and the discharge conveyor 46 are all provided so as to extend in the X-axis direction, and the container 22 can be conveyed in the −X direction.

The storage shelf 40d is further provided with a branching device 36 at the end in the + X direction, which is a connection portion with the warehousing side elevator 34. The container 22 transported to the 4th floor by the warehousing elevator 34 is distributed to one of the left shelf 42 and the right shelf 44 by the branching device 36. The above-mentioned warehousing instruction transmitted from the control unit 60 further includes information instructing which storage shelf of the left shelf 42 and the right shelf 44 should be distributed for each container 22. The branching device 36 distributes the container 22 to either the left shelf 42 or the right shelf 44 according to the warehousing instruction.

The left shelf 42 is divided into a plurality of divided shelves 43 arranged continuously in the X-axis direction. The individual divided shelves 43 are formed in a size that allows the containers 22 to be placed one by one. Further, each of the divided shelves 43 is formed as an independent conveyor, and the container 22 can be conveyed to the divided shelves 43 adjacent to the downstream side (-X direction). The container 22 distributed to the left shelf 42 is conveyed from the upstream side (+ X direction) to the downstream side (-X direction) by the individual split shelves 43, for example, the end portion on the downstream side (-X direction). It is stored in the state of being packed in order from the beginning. Similarly, the right shelf 44 is also divided into a plurality of divided shelves 45 arranged continuously in the X-axis direction, and the container 22 distributed to the right shelf 44 is conveyed by the individual divided shelves 45, for example. , It is stored in a packed state in order from the end on the downstream side (-X direction).

The storage shelf 40d further includes a cutting device 48. FIG. 1 shows a state in which the cutting device 48 is provided along the right shelf 44, but a similar cutting device 48 is also provided on the −Y direction side of the left shelf 42 symmetrically with the right shelf 44 side. ing. The cutting device 48 includes a shaft 48a extending in the X-axis direction and an arm 48b that moves along the shaft 48a in the X-axis direction and expands and contracts in the Y-axis direction. The cutting device 48 moves the container 22 stored at a specific position on the left shelf 42 or the right shelf 44 to the discharge conveyor 46 according to the delivery instruction transmitted from the control unit 60. The shipping instruction includes the shipping time for each container 22. The cutting device 48 moves the arm 48b of any of the cutting devices 48 to the position corresponding to the container 22 specified by the shipping instruction according to the shipping instruction, and extends the arm 48b to extend the designated container 22. , Push toward the discharge conveyor 46 at the delivery time specified by the delivery instruction. The cutting device 48 is also referred to as a “take-out unit”.

After the specific container 22 is moved to the discharge conveyor 46 by the cutting device 48, the container 22 arranged on the upstream side (+ X direction side) of the moved container 22 moves to the −X direction side. It may be allowed to be packed on the downstream side, or the divided shelves 43 and 45 on which the moved container 22 is arranged may be left empty. The container 22 extruded to the discharge conveyor 46 by the cutting device 48 is conveyed in the −X direction toward the delivery side elevator 50 by the discharge conveyor 46.

After being temporarily stored in any of the storage shelves 40a to 40d, the container 22 discharged from the discharge conveyor 46 of the storage shelves 40a to 40d is conveyed in the vertical direction (Z-axis direction) by the delivery side elevator 50 provided with an elevating mechanism. It is discharged to the delivery conveyor 51. In the present embodiment, the delivery-side elevator 50 is driven and controlled by the control unit 60, and operates according to a preset discharge algorithm for discharging the container 22 carried into the delivery-side elevator 50 faster. As an example, in the discharge algorithm of the present embodiment, the time until each container 22 carried into the delivery side elevator 50 is discharged from the delivery side elevator 50 without considering the classification of the group to which each container 22 belongs. However, the operation pattern in the delivery-side elevator 50 is defined so that the discharge amount of the container 22 per unit time in the delivery-side elevator 50 is as large as possible within a range not exceeding a predetermined upper limit time. The delivery-side elevator 50 is also referred to as a “delivery unit”.

FIG. 2 is an explanatory diagram schematically showing an outline of the internal configuration of the delivery side elevator 50. The delivery-side elevator 50 includes an upstream-side elevator box 52 and a downstream-side elevator box 54 that individually perform an elevator-elevation operation in the Z-axis direction. The upstream elevating box 52 is formed side by side in the Z-axis direction, and includes four transfer chambers 53a to 53d that can store one container 22 at a time. As the upstream elevating box 52 moves up and down, the transfer chamber 53a can accept the container 22 conveyed to the discharge conveyor 46 of the storage shelf 40a, and the transfer chamber 53b conveys to the discharge conveyor 46 of the storage shelf 40a or 40b. The transferred container 22 can be accepted, the transfer chamber 53c can accept the container 22 conveyed to the discharge conveyor 46 of the storage shelf 40b or 40c, and the transfer chamber 53d can accept the container 22 conveyed to the discharge conveyor 46 of the storage shelf 40c or 40d. The container 22 is accepted.

The downstream elevating box 54 is formed side by side in the Z-axis direction, and includes four transfer chambers 55a to 55d that can store one container 22 at a time. The downstream elevating box 54 exchanges the container 22 with the upstream elevating box 52. Specifically, as the downstream elevating box 54 and the upstream elevating box 52 move up and down, the transfer room 55a can exchange the container 22 with the transfer room 53a or 53b, and the transfer room 55b becomes the transfer room 53a. , 53b, or 53c can exchange the container 22, the transfer room 55c can exchange the container 22 with the transfer room 53b, 53c, or 53d, and the transfer room 55d can exchange the container 22 with the transfer room 53c or 53d. The container 22 can be exchanged with.

In the delivery-side elevator 50, the container 22 that has moved to the transfer chamber 55a or 55b of the downstream-side elevator box 54 is discharged to the delivery conveyor 51 from the outlet portion 56 provided in the delivery-side elevator 50. For example, when the container 22 carried from the storage shelf 40d into the transfer chamber 53d of the upstream elevating box 52 moves by the shortest route, it first moves to the transfer chamber 55d or 55c of the downstream elevating box 54. Further, it moves to the transfer chamber 53b or 53c of the upstream lift box 52, then reaches the transfer chamber 55a or 55b of the downstream lift box 54, and is discharged to the delivery conveyor 51. However, for example, if another container 22 already exists in the transfer room to be moved to, even if the upstream elevating box 52 or the downstream elevating box 54 moves up and down, it cannot move to the next transfer room, and the next There may be a waiting time waiting for the transfer room to become available. In the discharge algorithm described above, the operation of sending and receiving between the upstream elevator box 52 and the downstream elevator box 54 is performed so that such a waiting time is as short as possible and the delivery efficiency in the delivery side elevator 50 is as high as possible. The rules are predetermined. When controlling the operation of the delivery-side elevator 50, the information indicating from which storage shelf the container 22 was carried into the delivery-side elevator 50 is the container storage location and delivery indicated by the delivery instruction for each container 22. It may be obtained from the time, and a sensor may be provided at the connection portion of the elevator 50 on the delivery side with each storage shelf to detect from which storage shelf the container 22 is carried in.

The container 22 discharged to the delivery conveyor 51 is conveyed to the delivery conveyor 51 and used for the next process. In FIG. 1, a plurality of containers 22 belonging to any of the four groups A, B, C, and D are stored in the sorting device 20 in a disjointed order, sorted by group, and then in a predetermined group. It shows how the goods are delivered in the order of groups A, B, C, and D, which is the order. In the next step, for example, each container 22 may be sequentially loaded on a truck prepared for each group having a different delivery destination and waiting in the order of the predetermined group.

The control unit 60 is composed of a so-called microcomputer equipped with a CPU, a ROM, a RAM, and the like for executing logical operations, and drives and controls the entire sorting device 20. The control unit 60 acquires a detection signal from a sensor provided in each unit of the sorting device 20 such as a group identification unit 30 and an input signal from an input device provided in the sorting device 20, and performs necessary calculations. Further, a drive signal is output to each part of the warehousing conveyor 32, the warehousing side elevator 34, the storage shelves 40a to 40d, the warehousing side elevator 50, and the warehousing conveyor 51 to control the operation of each part.

FIG. 3 is an explanatory diagram showing the control unit 60 by a functional block representing a part of the function executed by the control unit 60. The control unit 60 includes a warehousing order information acquisition unit 62, an initial value generation unit 64, a warehousing instruction generation unit 66, a discharge algorithm storage unit 68, and a warehousing side elevator drive unit 69 as functional blocks. The operation of these functional blocks will be described in detail later.

(A-2) How to optimize delivery instructions:
In the present embodiment, the delivery instructions are optimized and set by using the optimization algorithm for the containers 22 that have been stored separately. The operation of setting such a delivery instruction will be described below. In the above description, the delivery instruction including the delivery time for each container 22 and the warehousing instruction which is information indicating the storage location in the sorting device 20 of each container 22 are distinguished. However, since the storage location of each container 22 affects the result of the optimization of the delivery time, in the following description, the "delivery instruction" is the delivery time of each container 22 in addition to the delivery time of each container 22. Information indicating a storage location in the sorting device 20 is included.

FIG. 4 is a flowchart showing a delivery instruction setting processing routine executed by the control unit 60. This routine is started when the operation of setting the delivery instruction by the control unit 60 can be started. Specifically, for example, the transfer of a plurality of containers 22 from the previous process to the warehousing conveyor 32 is started, and the last container 22 among the sorting target containers passes through the place where the group identification unit 30 is installed. It is started when the group is identified. In the present embodiment, when the last container 22 passes through the group identification unit 30 and the delivery instruction setting processing routine is activated, all the containers 22 including the first container 22 are carried into the warehousing side elevator 34. Not.

When this routine is activated, the warehousing order information acquisition unit 62 of the control unit 60 acquires information (hereinafter, also referred to as “receipt order information”) indicating the classification of the group to which each container 22 belongs in the warehousing order (step). S100). For example, the warehousing order information identified and acquired by the group identification unit 30 is stored in a storage device in the control unit 60, and in step S100, the warehousing order information acquisition unit 62 sorts all the items from the storage device. It suffices to acquire the receipt order information regarding the target container. The warehousing order information acquired by the warehousing order information acquisition unit 62 is sent to the initial value generation unit 64.

Next, the initial value generation unit 64 of the control unit 60 creates a plurality of sets of initial values of the delivery time for the container group for which the warehousing order information has been acquired (step S110). The delivery time generated for each container 22 in step S110 is referred to as "initial delivery time", and the set of initial delivery times generated for all containers 22 is referred to as "initial value of delivery time". The initial value of the delivery time created in step S110 is an initial value used for optimizing and setting the delivery instruction using the optimization algorithm.

FIG. 5 is an explanatory diagram showing a possible range of initial delivery times generated for each container 22. In the present embodiment, the initial delivery time is generated at a time within a limited range for each group to which each container 22 belongs. Specifically, in the present embodiment, the initial delivery time t _ini of each container 22 is set by the following equation (1).

Initial delivery time t _ini = reference time t _stn + addition time by cargo t _add + shift time t _s … (1)

Here, the reference time t _stn is a predetermined time as a reference for determining the delivery time. In the present embodiment, for example, the time is set as the time when the last container 22 passes through the group identification unit 30. The cargo addition time t _add is a value derived within a predetermined numerical range for each container 22. In the present embodiment, the cargo addition time t _add is set to a value within a numerical range limited to the number of containers 22 included in the group to which each container 22 belongs. Specifically, the cargo addition time t _add is a number randomly selected from a numerical range from 0 to the number of containers 22 included in the group to which the container 22 belongs. The shift time t _s is set to a different value for each group. In the present embodiment, as described above, the goods are delivered from the sorting device 20 in the order of groups A, B, C, and D, but the shift time t _s increases in the order of groups A, B, C, and D. It is stipulated as. The shift time t _s used to generate the initial delivery time t _ini of the containers 22 belonging to each of the groups A to D is set to the shift time t _sA , the shift time t _sB , the shift time t _sC , and the shift time t _sD , respectively. Then, for example, the shift time t _sB is set to twice the shift time t _sA , the shift time t _sC is set to three times the shift time t _sA , and the shift time t _sD is set to four times the shift time t _sA . can do. However, the method of setting the shift time _ts for each group may be different from the above.

By generating the initial delivery time t _ini as described above, the range that the initial delivery time t _ini can take is within the range limited to the number of containers in the group, as shown in FIG. The range of is shifted in the order of groups A, B, C, and D so that the time is delayed. FIG. 5 further shows how a gap time is provided between the groups that are continuously issued. A gap time G1 is provided between the latest time that can be taken as the initial delivery time t _ini of the container 22 in the group A and the earliest time that can be taken as the initial delivery time t _ini of the container 22 in the group B. ing. Similarly, a gap time G2 is provided between the group B and the group C, and a gap time G3 is provided between the group C and the group D. In the example of FIG. 5, a gap time G0 is provided between the reference time t _stn and the earliest possible time as the initial delivery time t _ini of the container 22 in the group A.

FIG. 6 is an explanatory diagram showing an example of the initial value of the delivery time, which is a set of the initial delivery time t _ini generated for each container 22. FIG. 6 shows each initial delivery time t _ini generated for each group regardless of the order of receipt in the entire container. “Container No.” shown in FIG. 6 indicates the order of warehousing within the group. As described above, since the cargo addition time t _add in Eq. (1) is a number randomly selected from the numerical range from 0 to the number of containers in the group, the initial delivery time t _ini is shown in FIG. It is randomly generated within the possible range of the initial delivery time for each group shown in.

Returning to FIG. 4, in step S110, when a plurality of “initial values of the delivery time” which are a set of the initial delivery time t _ini as shown in FIG. 6 are created, the initial value generation unit 64 of the control unit 60 determines the delivery time. An "initial value of the warehousing instruction" is given for each initial value, and an "initial value of the warehousing instruction" is created by combining the initial value of the warehousing time and the initial value of the warehousing instruction (step S120). Regarding the warehousing instruction, that is, which of the storage shelves 40a to 40d should be distributed, and which of the left shelf 42 and the right shelf 44 should be distributed in the distribution destination storage shelf. In the present embodiment, the initial value of the instruction is given at random using a random number.

The initial value of the delivery instruction created in step S120 is an initial value of a solution candidate for optimization using an optimization algorithm, and is also referred to as a “first generation solution candidate”. The number (number of individuals) of solution candidates of each generation generated in the process of optimization using the optimization algorithm, including the solution candidates of the first generation, can be, for example, tens to hundreds. The number of solution candidates for each generation is preferably 20 or more, and more preferably 30 or more, from the viewpoint of sufficient optimization. Further, the number of solution candidates for each generation is preferably 300 or less, and more preferably 100 or less, from the viewpoint of reducing the optimization process. Further, the number of solution candidates may be the same for all generations, or may be different for each generation.

When a plurality of initial values of solution candidates are created in step S120, the result is obtained when the delivery instruction generation unit 66 of the control unit 60 controls the operation of warehousing and shipping of the container according to the initial values of each created solution candidate. The number of discrepancies in the container order and the shipping time (hereinafter, these are also referred to as “the shipping status of the container”) are calculated (step S130). That is, all the containers 22 are stored in any of the storage shelves 40a to 40d in the sorting device 20 based on the initial value of the warehousing instruction included in the initial value of each solution candidate. Then, at the delivery time included in the initial value of the solution candidate, the container 22 corresponding to the delivery time is moved to the discharge conveyor 46 by using the cutting device 48, and the delivery side elevator operates according to the discharge algorithm described above. When the containers 22 are sequentially carried out from 50, the order of the containers 22 discharged from the delivery side elevator 50 and the time required for the containers 22 to be delivered are calculated. Information indicating the discharge algorithm that the delivery side elevator 50 follows during operation is stored in the discharge algorithm storage unit 68 of the control unit 60 (see FIG. 3), and the delivery instruction generation unit 66 is stored in the discharge algorithm storage unit 68. With reference to the above information, the shipping status of the container is calculated. When the sorting device 20 actually performs the operation of sorting the containers 22, the delivery-side elevator drive unit 69 of the control unit 60 refers to the above information stored in the discharge algorithm storage unit 68, and discharges the algorithm. According to this, the delivery side elevator 50 is driven and controlled.

The "number of mismatches in container order" in the shipping status of containers means different groups among the containers 22 belonging to the same group that are delivered consecutively when the containers 22 are delivered in a predetermined group order. Refers to the number of containers 22 belonging to. That is, it refers to the number of containers 22 that are out of the predetermined shipping order of groups, such as groups A, B, C, and D, and are mixed and shipped in different groups. Of the shipping states of the containers, the "delivery time" is the time required to deliver all the containers 22, and in the present embodiment, the last container after the last container 22 passes through the group identification unit 30. 22 is the time until the elevator 50 on the delivery side is discharged. It should be noted that the starting point of the warehousing time is the time when the last container 22 is warehousing, such as when the last container 22 has passed through the group identification unit 30, since the first container 22 is warehousing. This is a case where the evaluation result of the warehousing performance fluctuates due to the existence of a waiting time for the container 22 to be brought in before the container 22 is warehousing. Therefore, when all the containers 22 are continuously received at regular time intervals, for example, the time when the first container 22 is stored may be used as the starting point of the delivery time.

When the shipping state of the container is calculated for the initial value of each solution candidate in step S120, the shipping instruction generation unit 66 of the control unit 60 substitutes the calculated shipping state of the container into the objective function, and the value of the objective function and the value of the objective function in advance. Compare with the determined reference value (step S140). The objective function can be expressed by the following equation (2).

Objective function = delivery time + number of mismatches in container order x 200 ... (2)

In the above equation (2), the value 200 to be multiplied by the number of discrepancies in the container order is a value set so that the solution in which the disagreement occurs in the container order can be excluded with high accuracy, and different values may be set. Further, as the reference value used in step S140, there is no discrepancy in the container order according to the number of containers to be shipped and the number of containers for each group, and it is possible to select a solution candidate having sufficiently high shipping efficiency. It may be set appropriately. In step S140, when the value of the objective function is equal to or less than the reference value, it is determined that the warehousing performance of the sorting device 20 sequentially issuing containers for each group is equal to or higher than the predetermined reference performance. That is, the above-mentioned standard performance means that the delivery time is equal to or less than the above-mentioned reference value and the number of shipments per unit time is equal to or more than the predetermined standard number of shipments without causing a mismatch in the container order. The reference number of shipments is a value obtained by dividing the total number of containers 22 used for sorting by the sorting device 20 by the reference value.

When it is determined in step S140 that the value of the objective function is equal to or less than the reference value, the issue instruction generation unit 66 sets a solution candidate whose objective function satisfies the reference value or less as the issue instruction (step S150), and sets this routine. finish. If there are a plurality of solution candidates whose objective function value is equal to or less than the reference value among the plurality of solution candidates for which the objective function value has been calculated, in step S150, for example, the best solution candidate, that is, the objective function The solution candidate with the smallest value may be set as the delivery instruction.

When it is determined in step S140 that the value of the objective function exceeds the reference value, the issue instruction generation unit 66 of the control unit 60 has a smaller value of the objective function from the initial values of the plurality of solution candidates created in step S120. Four solution candidates are selected in order from the one (step S160). In this step S160, a solution candidate used for creating a next-generation solution candidate for optimization using a genetic algorithm described later is selected. The number of solution candidates selected in step S160 may be other than 4, and may be any natural number smaller than the number of next-generation solution candidates. When the number of next-generation solution candidates is about 30 to 300, the number of solution candidates selected in step S160 is preferably 2 or more, preferably 3 or more, from the viewpoint of improving the efficiency of optimization. Is more desirable. Further, when the number of solution candidates of the next generation is about 30 to 300, the number of solution candidates selected in step S160 is preferably 10 or less from the viewpoint of reducing the optimization process, and is 5 or less. Is more desirable.

As the method of selecting solution candidates in step S160, as described above, in addition to the method of selecting a predetermined number of upper solution candidates having a small value of the objective function, another method may be adopted. .. For example, a solution candidate is selected by roulette selection (specifically, the smaller the value of the objective function, the higher the probability of remaining in the next generation), which determines the probability of remaining in the next generation in proportion to the evaluation value of the solution candidate. You may choose.

After that, the delivery instruction generation unit 66 of the control unit 60 uses a genetic algorithm, which is an optimization algorithm, based on the solution candidates selected in step S160, and sets a preset number of solutions as next-generation solution candidates. Create a candidate (step S170). A part of the next-generation solution candidates may be created, for example, by elite storage in which the solution candidates selected in step S160 are left as they are. Further, the other part of the next-generation solution candidate may be created by crossing or mutating using the solution candidate selected in step S160.

When a preset number of solution candidates are created as next-generation solution candidates in step S170, the delivery instruction generation unit 66 of the control unit 60 returns to step S130 and repeats the processes after step S130. That is, for each of the created next-generation solution candidates, the number of mismatches in the container order and the delivery time are calculated (step S130), the value of the objective function is obtained, and the value of the objective function is compared with the reference value (step S130). Step S140). Then, until a solution candidate whose objective function value is lower than the reference value is obtained in step S150, a next-generation solution candidate is created using a genetic algorithm (steps S160 and S170), and the objective function value and the reference value are combined. Repeat the operation of comparing.

When the warehousing instruction is set in step S150, the control unit 60 transmits a drive signal to each unit including the warehousing side elevator 34 and the branching device 36 according to the information indicating the storage location in the sorting device 20 included in the warehousing instruction. The individual containers 22 are housed in the sorting device 20. Then, the control unit 60 moves the container 22 corresponding to the delivery time to the discharge conveyor 46 by using the cutting device 48 according to the information indicating the delivery time included in the delivery instruction, and operates according to the discharge algorithm described above. Carry it into the side elevator 50.

According to the delivery instruction system 10 and the delivery instruction setting method of the present embodiment configured as described above, when the delivery instruction including the delivery time of each container 22 is set in the sorting device 20, the individual container 22 is set. For each, to the predetermined reference time t _stn , add the cargo addition time t _add within the predetermined numerical range and the shift time t _s with different values for each group. Then, the initial delivery time t _ini corresponding to each container 22 is derived, and the initial value for setting the delivery instruction is generated. Then, a shipping instruction is generated from the above initial value by using a genetic algorithm which is an optimization algorithm so that the shipping performance of sequentially issuing a plurality of containers 22 for each group becomes equal to or higher than a predetermined reference performance. Therefore, it is possible to improve the efficiency of setting a delivery instruction having high delivery performance for sorting a plurality of received containers 22 into groups and sequentially issuing them. That is, when generating the initial value for setting the issue instruction, the initial value of the issue time in the group is set within a limited range, and the range of the initial value of the issue time for each group is set. , Since the order of delivery of the group is shifted, it is easy to optimize the delivery instruction from the initial value.

Further, in the present embodiment, a gap time is provided between the groups to be continuously delivered at the initial delivery time t _ini (see FIG. 5). Therefore, in the optimization using the genetic algorithm, it is possible to obtain a solution with a shorter number of generations, no disagreement in the container order, and high delivery performance. For example, depending on the shift time t _s of the equation (1), the range that can be taken as the initial delivery time t _ini overlaps between at least one of the groups that are continuously issued, and a gap time is provided. It may not be available. However, it is easier to reduce the number of generations required to obtain the desired solution if the overlap of the range of the initial delivery time t _ini is small between the group to be shipped first and the group to be shipped later. Is desirable. Hereinafter, the effects of this embodiment will be described with reference to specific examples.

FIG. 7 is an explanatory diagram showing the relationship between the number of generations of optimization using the genetic algorithm and the objective function (delivery time + number of discrepancies in container order × 200). In FIG. 7, the graph (α) shows the case where the total number of containers to be sorted is 50, and the shift time t _{sA described} above corresponding to the equation (1) is 10 seconds, and the graph (β) shows the case. , The case where the total number of containers is 150 and the shift time t _sA is 15 seconds is shown. Further, the graph (γ) shows a comparative example, in which the total number of containers is 50, and in step S110 of FIG. 4, a random initial value of the delivery time is generated instead of the initial value of the delivery time of the embodiment. Represents the case. FIG. 7 shows a simulation result in which a control unit of a sorting device is given a predetermined warehousing order information as a process corresponding to step S100 of FIG. 4 to execute a process of optimizing a warehousing instruction. Hereinafter, a comparative example of the graph (γ) will be further described. In the comparative example, the delivery instruction is set by an operation similar to that of the embodiment, and only the operation of step S110 is different.

FIG. 8 is an explanatory diagram showing a possible range of the initial delivery time generated for each container 22 in the comparative example in the same manner as in FIG. In the comparative example, the initial delivery time t _ini of each container 22 is randomly set within a certain numerical range regardless of the group to which the container 22 belongs. The initial delivery time t _ini of each container 22 in the comparative example is set by the following equation (3).

Initial delivery time t _ini = reference time t _stn + additional time for each shipment t _add … (3)

In the comparative example, the load addition time t _add is a number randomly selected from a numerical range from 0 to the total number of containers. Further, in the comparative example, unlike the equation (1) of the embodiment, the shift time _ts is not added. Therefore, as shown in FIG. 8, the initial delivery time t _ini is randomly set for the container 22 belonging to any of the groups within the same wide numerical range.

FIG. 9 is an explanatory diagram showing an example of the initial value of the delivery time, which is a set of the initial delivery time t _ini generated for each container 22 of the comparative example, in the same manner as in FIG. FIG. 9 shows each initial warehousing time t _ini generated for each group according to the warehousing order within the group, regardless of the overall warehousing order. As described above, since the cargo addition time t _add in Eq. (2) is a number randomly selected from the numerical range from 0 to the total number of containers, the initial delivery time t _ini is the group shown in FIG. It is randomly generated within the possible range of the initial delivery time that is common to all.

In FIG. 7, the value of the “objective function (delivery time + number of discrepancies in container order × 200)” is shown on the left side of the vertical axis. Further, on the right side of the vertical axis, "the number of discrepancies in the container order" is shown. Further, each graph in FIG. 7 shows the result for the solution candidate having the smallest value of the objective function in each generation of the optimization process. The graphs (α) and (β) show the relationship between the number of generations and the value of the “objective function (delivery time + number of discrepancies in container order × 200)”, and the graph (γ) shows the number of generations and “container order”. The relationship with "the number of discrepancies" is shown.

As shown in FIG. 7, in the graph (α) in which the total number of containers is 50 and the shift time t _sA is 10 seconds, the disagreement in the container order is eliminated by the calculation of the 3rd generation, and the delivery performance exceeding the standard performance is achieved. Obtained. Further, in the graph (β) in which the total number of containers is 150 and the shift time t _sA is 15 seconds, the disagreement in the container order was eliminated by the calculation of the 5th generation, and the delivery performance higher than the standard performance was obtained. On the other hand, in the graph (γ) of the comparative example, the discrepancy in the container order could not be resolved even if the calculation was performed up to the 50th generation.

FIG. 10 is an explanatory diagram showing the relationship between the shift time t _sA and the container sorting amount. The container sorting amount is a value obtained by dividing the total number of containers 22 used for sorting by the sorting device 20 by the delivery time. In FIG. 10, the horizontal axis shows the shift time t _sA , and the vertical axis shows the container sorting amount. FIG. 10 shows a calculation result when the total number of containers is 100 and the number of solution candidates for each generation is 100.

In FIG. 10, the result of the first generation solution candidate, which is the initial value of the delivery instruction for optimization using the optimization algorithm, is shown in the graph (a), and the result of the tenth generation solution candidate is shown. , Shown in graph (b). Each graph shows the result of the solution candidate (hereinafter, also referred to as the best solution candidate) having the largest container sorting amount among the solution candidates belonging to each generation. FIG. 10 shows only the results of the two generations, the first generation and the tenth generation, but the results of the generations between the first generation and the tenth generation are sandwiched between the two graphs (a) and (b). As the number of generations increases, the graph becomes closer to the graph (a) from the graph (a) (data is not shown). In FIG. 10, the region in which no solution candidate for which the disagreement in the container order has been resolved can be obtained, including the generations between them, is shown as the “container order disagreement region”.

In the example shown in FIG. 10, in the first generation, by setting the shift time t _sA to 14 seconds or more, an individual (solution candidate) in which the disagreement in the container order was eliminated was obtained. Further, in the first generation, when the shift time t _sA was in the range of 15 to 18 seconds, a particularly large value was obtained as the container sorting amount. In the 10th generation, by setting the shift time t _sA to 10 seconds or more, an individual (solution candidate) in which the disagreement in the container order was eliminated was obtained. In the 10th generation, if the shift time t _sA is 10 seconds or more, the amount of container sorting is larger than that in the 1st generation in the entire range up to the calculated shift time t _sA 25 seconds. In the 10th generation, unlike the 1st generation, the container sorting amount did not decrease even if the shift time t _sA was lengthened, and showed a stable high level. Further, when the best solution candidates of the 1st generation and the 10th generation are compared and examined, the average value of the gap times in the solution candidates (the average values of the gap times G1, G2, etc. shown in FIG. 5) is 18 to 12 seconds. Was shortened to (data not shown). This is because as the number of generations progresses while performing genetic operations such as crossing and mutation, the degree of optimization of solution candidates increases, and the amount of container sorting (number of shipments per unit time) increases, resulting in a gap time. Is thought to have become shorter.

As described above, by setting the initial delivery time t _ini using the equation (1) from FIGS. 7 and 10, the efficiency of setting the delivery instruction with high delivery efficiency is improved while suppressing the disagreement of the container order. It was confirmed that the effect was obtained.

The shift time _ts required to eliminate the mismatch in the container order is, for example, the discharge algorithm in the elevator 50 on the delivery side, the usage status of the sorting device 20 (for example, the total number of containers to be sorted, the group to which the container belongs). It may vary depending on the number, the number of containers contained in each group, etc. (see, for example, graphs (α) (β) in FIG. 7 and FIG. 10). Therefore, the shift time _ts may be appropriately set for each of these conditions according to the discharge algorithm in the delivery side elevator 50 and the usage status of the sorting device 20.

B. Second embodiment:
FIG. 11 is an explanatory diagram showing the control unit 160 included in the delivery instruction system 10 of the second embodiment by a functional block representing a part of the function executed by the control unit 160. In the delivery instruction system 10 of the second embodiment, similarly to the first embodiment, a plurality of containers 22 classified in advance into a plurality of groups are stored, and the stored plurality of containers 22 are sorted by group. It is a device that sequentially issues items for each group. In the delivery instruction system 10 of the second embodiment, the same reference number is assigned to the portion common to the delivery instruction system 10 of the first embodiment. As shown in FIG. 11, the control unit 160 includes an initial value generation unit 64, a delivery instruction generation unit 66, an discharge algorithm storage unit 68, and a delivery side elevator drive unit 69 as functional blocks, as in the first embodiment. In addition, a storage location acquisition unit 61 is provided in place of the warehousing order information acquisition unit 62.

In the first embodiment, when the last container among the sorting target containers passes through the group identification unit 30 and the control unit 60 starts the optimization operation, all the sorting target containers are set to the warehousing side elevator 34. The shipping instructions that have not been brought in and are set as a result of the optimization include the shipping instructions for all the containers to be sorted. On the other hand, in the second embodiment, when the control unit 160 starts the optimization operation, all the sorting target containers are already on the left shelf 42 or in any of the storage shelves 40a to 40d. It is stored in the right shelf 44. Therefore, the warehousing instruction set in the second embodiment includes the warehousing time of each container 22, but does not include the warehousing instruction.

FIG. 12 is a flowchart showing a delivery instruction setting processing routine executed by the control unit 160 of the delivery instruction system 10 of the second embodiment. In this routine, the warehousing of the container 22 into the sorting device 20 is started, and the warehousing instruction for all the containers to be sorted (in which of the storage shelves 40a to 40d, the left shelf 42 or the right shelf 44 is stored. Is activated when the instruction) is determined.

The storage location of the container 22 may be determined according to some predetermined algorithm. For example, the storage shelves 40a, 40b, 40c, and 40d may be sequentially allocated in this order. Further, the storage destinations may be alternately allocated between the left shelf 42 and the right shelf 44. Alternatively, the storage shelves to be stored may be randomly selected from the storage shelves 40a to 40d, and the storage destination may be randomly selected between the left shelf 42 and the right shelf 44. Further, at this time, of the storage shelves 40a to 40d and the left shelf 42 and the right shelf 44 of each storage shelf, the smaller the number of containers already stored, the more weighted and stored the container 22 is. You may decide the destination.

When the delivery instruction setting processing routine of FIG. 12 is activated, the storage location acquisition unit 61 of the control unit 160 acquires information indicating the storage location of the stored container 22 (step S200). The control unit 160 stores the result of determining the storage destination of the container to be sorted, and the storage location acquisition unit 61 uses the stored result described above to indicate information indicating the storage location of each container 22, specifically. Acquires a storage place (which division shelf 43 or division shelf 45 is stored) in the left shelf 42 or the right shelf 44 among the storage shelves 40a to 40d.

After that, the initial value generation unit 64 of the control unit 160 creates a plurality of initial values of the delivery time of each container 22 as initial values of solution candidates (step S210). The process of step S210 is the same process as step S110 of the first embodiment described above, and is a set of initial issue time t _ini obtained by the equation (1) as shown in FIG. Create multiple "values" as "initial values for shipping instructions". In the second embodiment, the warehousing order information obtained from the group identification unit 30 is not required when setting the warehousing instruction based on FIG. Therefore, the reference time t _stn used when setting the initial delivery time t _ini of each container 22 by the equation (1) is, for example, the last time instead of the time when the last container 22 passes through the group identification unit 30. The storage location of the container 22 can be set to a determined time.

After that, the delivery instruction generation unit 66 of the control unit 160 sets the delivery instruction by optimization using a genetic algorithm using the initial value of the solution candidate created in step S210 (steps S230 to S270). The processes of steps S230 to S270 are the same as those of steps S130 to S170 of FIG.

With such a configuration, the initial delivery time t _ini is generated using Eq. (1), the initial value of the delivery instruction is created, and the delivery instruction is set by optimization using a genetic algorithm. Therefore, even when a plurality of containers for which warehousing has already been completed are set as sorting target containers, the effect of increasing the efficiency of setting a warehousing instruction having high warehousing performance can be obtained as in the first embodiment.

C. Third embodiment:
FIG. 13 is an explanatory diagram showing the control unit 260 included in the delivery instruction system 10 of the third embodiment by a functional block representing a part of the functions executed by the control unit 260. In the delivery instruction system 10 of the third embodiment, similarly to the first embodiment, a plurality of containers 22 classified in advance into a plurality of groups are stored, and the stored plurality of containers 22 are sorted by group. It is a device that sequentially issues items for each group. In the delivery instruction system 10 of the third embodiment, the same reference number is assigned to the portion common to the delivery instruction system 10 of the first embodiment. As the first embodiment, the control unit 260 includes a warehousing order information acquisition unit 62, an initial value generation unit 64, a warehousing instruction generation unit 66, an discharge algorithm storage unit 68, and a warehousing side elevator drive unit 69 as functional blocks. Further, as in the second embodiment, the storage location acquisition unit 61 is provided.

In the first embodiment, the shipping instruction set as a result of optimization includes a warehousing instruction for all the sorting target containers, and in the second embodiment, all the shipping instructions set as a result of optimization are included. Does not include a warehousing instruction for the container to be sorted. On the other hand, in the third embodiment, when the last container 22 among the sorting target shelves passes through the group identification unit 30 and starts the optimization operation, some containers including the first container 22 are included. With respect to 22, the storage location in the sorting device 20 has already been determined, and at least a part of the containers 22 thereof is stored in the left shelf 42 or the right shelf 44 in any of the storage shelves 40a to 40d. ing. Therefore, in the third embodiment, when the delivery instruction including the warehousing instruction can be set, specifically, for example, when the last container 22 passes through the group identification unit 30, the storage in the sorting device 20 is performed. The delivery instruction including the warehousing instruction is set only for the container 22 whose location has not been determined, specifically, for example, the container 22 that has not been carried into the warehousing side elevator 34. The container 22 in which the storage location in the sorting device 20 is determined when the process of optimizing the delivery instruction is started is also called a "storage location determined container", and the storage location in the sorting device 20 is determined. The non-existing container 22 is also referred to as a “storage location undecided container”.

FIG. 14 is a flowchart showing a delivery instruction setting processing routine executed by the control unit 260 of the delivery instruction system 10 of the third embodiment. This routine is started when the operation of setting the delivery instruction by the control unit 260 can be started. Specifically, for example, the transfer of the plurality of containers 22 from the previous process to the warehousing conveyor 32 is started, and the container 22 is activated when the last container 22 among the sorting target containers passes through the group identification unit 30. In the present embodiment, as described above, some containers including the first container 22 are included before the time when the last container 22 passes through the group identification unit 30 and the delivery instruction setting processing routine is started. With respect to, a warehousing instruction is set, and the operation of warehousing to each of the storage shelves 40a to 40d is started via the warehousing side elevator 34. For the container whose storage location has been determined, the storage location determined before the delivery instruction setting processing routine is started may be determined according to some predetermined algorithm as in the second embodiment.

When this routine is activated, the storage location acquisition unit 61 of the control unit 260 acquires information indicating the storage location of the container whose storage location has been determined, and the warehousing order information acquisition unit 62 acquires the warehousing order of the individual containers 22. Acquire information (step S300). The storage location determined for the storage location determined container is stored in the storage device included in the control unit 260. Further, the warehousing order information acquired by the warehousing order information acquisition unit 62 is the warehousing order information for the storage location undecided container to be warehousing after the storage location determined container, and is identified by the group identification unit 30 and controlled by the control unit. It is the warehousing order information stored in the storage device in 260. The information indicating the storage location and the warehousing order information acquired in step S300 are sent to the initial value generation unit 64.

Next, the initial value generation unit 64 of the control unit 260 creates a plurality of sets of initial values of the delivery time for the sorting target container (step S310). The process of step S310 is the same as the process of step S110 of the first embodiment. That is, a plurality of "initial values of the delivery time", which is a set of the initial delivery time t _ini obtained by the equation (1) as shown in FIG. 6, are created. In the third embodiment, the reference time t _stn used when setting the initial delivery time t _ini of each container 22 by the equation (1) in step S310 is, for example, the last container as in the first embodiment. It may be the time when 22 has passed the group identification unit 30.

When a plurality of "initial values of the shipping time", which is a set of the initial shipping time t _ini , are created in step S310, the initial value generating unit 64 of the control unit 260 stores the storage acquired in step S300 for each initial value of the shipping time. The information indicating the location or the "initial value of the warehousing instruction" is given to create the "initial value of the warehousing instruction" (step S320). In step S320, for the container whose storage location has been determined, the information indicating the storage location acquired in step S300, that is, the already determined warehousing instruction is given. Further, for the storage location undecided container to be stored after the storage location determined container, the initial value of the warehousing instruction tentatively set for each storage location undecided load in the same manner as in step S120 of the first embodiment. Is given.

After that, the delivery instruction generation unit 66 of the control unit 260 sets the delivery instruction by optimization using a genetic algorithm using the initial value of the solution candidate created in step S310 (steps S330 to S370). The processes of steps S330 to S370 are the same as those of steps S130 to S170 of FIG. However, in the third embodiment, when creating the next-generation solution candidate, the warehousing instruction of the container whose storage location has been determined is unchanged even if the number of generations is repeated.

With such a configuration, the initial delivery time t _ini is generated using Eq. (1), the initial value of the delivery instruction is created, and the delivery instruction is set by optimization using a genetic algorithm. Therefore, even when a plurality of containers for which warehousing has already been completed are set as sorting target containers for some containers, the efficiency of setting a warehousing instruction with high warehousing performance is achieved as in the first embodiment. The effect of enhancing is obtained.

D. Other embodiments:
(D-1) About cargo:
In the above embodiment, the load sorted by the sorting device 20 is a container 22 containing articles, but a different configuration may be used. The load may be, for example, an individual article, or may be a single unit of transfer.

(D-2) Addition time by cargo t _add :
In the above embodiment, the cargo addition time t _add used when generating the initial delivery time t _ini using the equation (1) is a numerical range from 0 to the number of containers 22 included in the group to which the container 22 belongs. The number is randomly selected from, but it may have a different configuration. For example, the cargo addition time t _add may be a number obtained by multiplying a number randomly selected as described above by a predetermined coefficient. At this time, the above-mentioned coefficient may be a different number for each group. Alternatively, the cargo addition time t _add does not have to be set randomly, and for example, the warehousing order within the group of individual containers 22 may be used as the cargo addition time t _add . However, it is desirable to randomly set the load-based addition time t _add because it is easy to suppress inconveniences such as initial convergence and increase the efficiency of reaching a desired solution by optimization using a genetic algorithm. Further, the cargo addition time t _add does not have to be within the numerical range limited to the number of containers included in the group to which each container 22 belongs, and may be within a preset numerical range.

(D-3) About the shift time t _s :
In the above embodiment, the shift time t _s used when generating the initial delivery time t _ini using the equation (1) is set to be longer in the order of delivery in the later group, but may have a different configuration. When generating the initial delivery time t _ini , the delivery time for each group is calculated by adding the shift time t _s with different values for each group together with the addition time t _add for each shipment in a predetermined numerical range. By generating the initial delivery time t _ini so as to shift from each other, the effect of reducing the processing for optimizing the delivery instruction can be obtained. However, in order to enhance the effect of reducing the process for optimizing the delivery instruction, it is desirable to set the shift time _ts longer as the delivery order is later in the group as in the embodiment.

(D-4) Sorting device:
In the above embodiment, the sorting device 20 has the configuration shown in FIG. 1, but may have a different configuration. For example, the sorting device is "a plurality of storage shelves capable of storing a plurality of loads" and "a take-out unit for taking out the loads stored in any place of the plurality of storage shelves from the storage shelves at a specified delivery time". And "a warehousing unit that sequentially unloads loads sequentially taken out from a plurality of storage shelves by a unloading unit from a sorting device" may be provided. Specifically, for example, a stacker crane may be used instead of the warehousing-side elevator 34 that distributes the load to a plurality of storage shelves, the cutting-out device 48 that is the take-out section, and the warehousing-side elevator 50 that is the warehousing section.

(D-5) Emission algorithm:
In the above embodiment, the delivery-side elevator 50 sequentially receives the containers 22 conveyed to each discharge conveyor 46, and performs the delivery operation according to a predetermined discharge algorithm without considering the group to which the containers 22 belong. However, it may have a different configuration. For example, in the delivery-side elevator 50, the group to which the container 22 carried into the delivery-side elevator 50 belongs is identified, and the movement in the delivery-side elevator 50 is controlled by a discharge algorithm in consideration of each group, or the delivery-side elevator 50 is used. By making the container 22 stand by at the entrance of the elevator 50 on the delivery side prior to carrying in the elevator, the container 22 is actively replaced in the elevator 50 on the delivery side so that the order of the containers 22 follows the group order. It is possible. However, in the above embodiment, by using an emission algorithm that does not consider each group and suppressing the selection of solutions that cause group order disagreement in the optimization process, in addition to improving the efficiency of issuing instruction setting with high delivery performance, It is possible to reduce the processing for drive control of the elevator 50 on the delivery side.

(D-6) About the optimization algorithm:
In the above embodiment, the genetic algorithm is used as the optimization algorithm for optimizing and setting the delivery instruction, but a different configuration may be used. For example, various optimization algorithms that can be used for combination optimization such as tabu search (TS), simulated annealing (SA, simulated annealing), or simulated evolution (SimE) can be used. Even in such a case, when creating the initial value of optimization, by setting a different shift time _ts between groups as in Eq. (1), a desired delivery instruction with high delivery performance can be obtained. The efficiency of setting can be increased.

10 ... Delivery instruction system 20 ... Sorting device 22 ... Container 30 ... Group identification unit 32 ... Warehousing conveyor 34 ... Warehousing side elevator 36 ... Branching device 40a-40d ... Storage shelf 42 ... Left shelf 43, 45 ... Split shelf 44 ... Right shelf 46 ... Discharge conveyor 48 ... Cutting device 48a ... Shaft 48b ... Arm 50 ... Delivery side elevator 51 ... Delivery conveyor 52 ... Upstream side lifting box 53a to 53d ... Transfer room 54 ... Downstream side lifting box 55a to 55d ... Delivery room 56 ... Exit Units 60, 160, 260 ... Control unit 61 ... Storage location acquisition unit 62 ... Storage order information acquisition unit 64 ... Initial value generation unit 66 ... Delivery instruction generation unit 68 ... Discharge algorithm storage unit 69 ... Delivery side elevator drive unit

Claims (8)

It is a method of setting a warehousing instruction for setting a warehousing instruction including a warehousing time of each said ware in a sorting device that sorts and warehousing by changing the order of a plurality of warehousing.
The plurality of loads are previously classified into a plurality of groups, and the plurality of loads are classified into a plurality of groups.
By adding the load-specific addition time within the predetermined numerical range and the shift time with different values for each group to the predetermined reference time for each of the loads. , Derived the initial delivery time corresponding to the individual shipment, and generated the initial value for setting the delivery instruction.
An optimization algorithm is used so that the delivery performance of the shipments classified into the same group is continuously delivered and the delivery performance of the plurality of loads is sequentially delivered for each group to be equal to or higher than the predetermined standard performance. , A method of setting a shipping instruction to generate the shipping instruction from the initial value. The method for setting a shipping instruction according to claim 1.
The plurality of groups have a predetermined order of delivery, and the order of delivery is predetermined.
The shift time is a method of setting a delivery instruction in which the delivery order is set longer as the group is later. The method for setting a shipping instruction according to claim 2.
In the initial value, the initial delivery time of the cargo belonging to the first group among the plurality of groups and the latest initial delivery time is the second group to be delivered next to the first group. A method of setting a delivery instruction in which the initial delivery time is set earlier than the initial delivery time of the earliest load among the packages belonging to. The method for setting a shipping instruction according to any one of claims 1 to 3.
The optimization algorithm is a genetic algorithm and is
The loading addition time is a method of setting a shipping instruction that is randomly derived in a numerical range limited to the number of the loading included in each of the groups. The method for setting a shipping instruction according to any one of claims 1 to 4.
The reference performance is the number of shipments per unit time in the order of delivery of the plurality of shipments, without the shipments belonging to the same group being continuously issued and the shipments belonging to different groups being mixed. Is a setting method for issuing instructions that is set so that is greater than or equal to the predetermined standard number of issues. The method for setting a shipping instruction according to any one of claims 1 to 5.
The initial value includes the initial delivery time corresponding to the individual load and the initial information of the information indicating the storage location of the individual load in the sorting device.
As the initial information of the information indicating the storage location, for the storage location determined load for which the storage location in the sorting device has already been determined when the initial value is generated, the information indicating the determined storage location is used. For a storage location undecided load in which the storage location in the sorting device has not been determined when the initial value is generated, the individual storage location undecided load is set prior to the storage of the storage location undetermined load. On the other hand, using the temporarily set information,
A method of setting a warehousing instruction that uses the optimization algorithm to generate an instruction as the warehousing instruction that includes information indicating a storage location of the undetermined storage location in addition to the unloading time of each of the packages. The method for setting a shipping instruction according to any one of claims 1 to 6.
The sorting device is
Multiple storage shelves that can store multiple of the above loads,
A take-out unit that takes out the load stored in an arbitrary position of the plurality of storage shelves from the storage shelf at the delivery time specified in the delivery instruction corresponding to the load.
A delivery unit that sequentially discharges the load sequentially taken out from the plurality of storage shelves by the take-out unit from the sorting device according to a predetermined discharge algorithm.
Equipped with
The operation of generating the delivery instruction from the initial value is a method of setting a delivery instruction, which is performed so that the delivery performance becomes equal to or higher than the reference performance when the delivery unit discharges the load according to the discharge algorithm. It is a delivery instruction system that sets a delivery instruction including the delivery time of each said load in a sorting device that sorts and delivers by changing the order of a plurality of packages that have been received.
The plurality of loads are previously classified into a plurality of groups, and the plurality of loads are classified into a plurality of groups.
By adding the load-specific addition time within the predetermined numerical range and the shift time with different values for each group to the predetermined reference time for each of the above-mentioned packages. , An initial value generator that derives the initial delivery time corresponding to each of the shipments and generates an initial value for setting the delivery instruction.
An optimization algorithm is used so that the delivery performance of the shipments classified into the same group is continuously delivered and the delivery performance of the plurality of loads is sequentially delivered for each group to be equal to or higher than the predetermined standard performance. , A delivery instruction generation unit that generates the delivery instruction from the initial value,
A take-out unit that performs an operation of unloading each of the loads at the warehousing time included in the warehousing instruction.
Delivery instruction system equipped with.

Images