Picking task generation method and device and picking system
Patent Information
- Application Number
- CN202510300306.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-13
- Publication Date
- 2026-09-15
Smart Images
Figure CN122759596A_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of intelligent warehousing and logistics, and in particular to a picking task generation method, apparatus and picking system. Background Technology
[0002] In warehousing and logistics, warehouses process numerous orders daily. Items from these orders are picked from the warehouse, and the items for each order are packaged. The application of computer technology in warehousing and logistics is becoming increasingly widespread, and robots are being introduced to assist in picking tasks. Summary of the Invention
[0003] This disclosure proposes an automatic picking task generation scheme. Based on input data such as current warehouse order data, bundled order constraint data, and partition data, as well as decision variables such as the variable of an order being assigned to a bundled order and the variable of a bundled order including its partitions, the scheme constructs the objectives and constraints for generating picking tasks for each order. Based on the solutions of the decision variables that satisfy the objectives and constraints, the scheme automatically generates the corresponding picking tasks for each order. One bundled order and one of its included partitions correspond to one picking task. This scheme automatically breaks down orders into picking tasks based on real-time information such as order demand and warehouse conditions, combined with specific decision variables. The generated picking tasks meet the predetermined objectives and constraints, optimize the picking process, and have strong executability.
[0004] This disclosure provides a method for generating picking tasks, including:
[0005] Determine the input data, which includes: current warehouse order data, collection order restriction data, and partition data;
[0006] Determine the decision variables, which include: the variable of order allocation to the collection order, and the variable of the collection order including the partitions, wherein a collection order and its included partitions correspond to a picking task;
[0007] Based on the input data and the decision variables, construct the objectives and constraints for generating picking tasks for the warehouse's orders;
[0008] Based on the solutions of the decision variables that satisfy the stated objective and the stated constraints, the corresponding picking tasks for the warehouse orders are generated.
[0009] In some embodiments, the objective includes a first objective and a second objective, and the objective for generating picking tasks for orders in the warehouse includes:
[0010] Based on the current order data and pooled order limit data of the warehouse, and combined with the variable of order allocation to pooled orders, construct the first objective that maximizes the ratio of the capacity of each pooled order to the maximum capacity of the pooled order.
[0011] Based on the constraints and partition data of the bundled orders, and combined with the variables of the bundled orders including partitions, a second objective is constructed that minimizes the number of picking tasks split into bundled orders.
[0012] In some embodiments, the current order data for the warehouse includes the current number of orders in the warehouse, and the aggregate order limit data includes the maximum number of aggregate orders and the maximum capacity of aggregate orders. Constructing a first objective that maximizes the ratio of the capacity of each aggregate order to the maximum capacity of the aggregate order includes:
[0013] Based on the current number of orders in the warehouse and the variables of orders assigned to collection orders, construct a feature representation that maximizes the ratio of the capacity of each collection order to the maximum capacity of the collection order;
[0014] Under the maximum number of sets, the feature representations that have the largest ratio between the capacity of each set and the maximum capacity of the set are summed to construct the first objective.
[0015] In some embodiments, the maximum capacity of a collection order includes one or more of the following: maximum volume of the collection order, maximum weight of the collection order, and maximum order quantity of the collection order. The feature representation that maximizes the ratio of the capacity of each collection order to the maximum capacity of the collection order includes:
[0016] Given the current number of orders in the warehouse, sum the variables of orders assigned to collection orders to obtain the feature representation of the order quantity of each collection order, and construct the first feature representation of the ratio of the feature representation of the order quantity of each collection order to the maximum order quantity of the collection order;
[0017] Given the current number of orders in the warehouse, the product of the variable that the order is assigned to a collection order and the volume of the order is summed to obtain the feature representation of the volume of each collection order. A second feature representation is constructed by comparing the feature representation of the volume of each collection order with the maximum volume of the collection order.
[0018] Given the current number of orders in the warehouse, sum the product of the variable that the order is assigned to a collection order and the weight of the order to obtain the feature representation of the weight of each collection order. Construct a third feature representation of the ratio of the feature representation of the weight of each collection order to the maximum weight of the collection order.
[0019] Based on the feature representation that maximizes one or more of the first feature representation, the second feature representation, and the third feature representation, construct the feature representation that maximizes the ratio of the capacity of each set to the maximum capacity of the set.
[0020] In some embodiments, the partition data includes the number of partitions, the aggregate order limit data includes the maximum number of aggregate orders, and the decision variable further includes a variable indicating whether the aggregate order is assigned an order. Constructing a second objective that minimizes the number of picking tasks resulting from splitting aggregate orders includes:
[0021] Given the number of partitions and the maximum number of collection orders, the variables of collection orders including partitions are summed to obtain a characteristic representation of the number of picking tasks;
[0022] Under the maximum number of set orders, sum the variables of whether set orders are assigned to orders to obtain a characteristic representation of the number of set orders;
[0023] The second objective is constructed based on the feature representation that minimizes the ratio of the feature representation of the number of picking tasks to the feature representation of the number of set orders.
[0024] In some embodiments, the current order data of the warehouse includes a binary variable of the current number of orders in the warehouse and whether the orders include partitions, the aggregate order limit data includes the maximum number of aggregate orders, and constructing the second objective includes:
[0025] Given the number of orders and the maximum number of aggregate orders, the variables representing the number of orders allocated to aggregate orders are summed to obtain a characteristic representation of the number of allocated orders.
[0026] Under the maximum number of set orders, first sum the variables of orders assigned to set orders, and under the maximum number of set orders and the number of partitions, sum the product of the binary variable of whether the order includes a partition and the summation result to obtain the characteristic representation of the total number of partitions of the assigned orders;
[0027] The second objective is constructed based on the feature representation that minimizes the ratio of the feature representation of the number of picking tasks to the feature representation of the number of collection orders, and the feature representation that minimizes the ratio of the feature representation of the total number of partitions for order allocation to the feature representation of the number of allocated orders.
[0028] In some embodiments, the objective further includes a third objective, the objective of which is constructed to generate picking tasks for orders in the warehouse, including:
[0029] Based on the set order constraint data and partition data, and combined with the variables of the set order including partitions, a third objective is constructed to balance the picking tasks of each partition.
[0030] In some embodiments, the aggregate order limit data includes the maximum number of aggregate orders, the partition data includes the number of picking tasks currently pending in each partition, and the third objective of balancing picking tasks across partitions includes:
[0031] Under the maximum number of batch orders, the variables of the batch orders, including the partitions, are summed to obtain a characteristic representation of the number of picking tasks to be assigned to each partition;
[0032] The feature representation of the number of picking tasks to be assigned in a partition and the number of picking tasks to be picked up in the partition are summed to obtain the feature representation of the total number of picking tasks in the partition.
[0033] Based on the feature representation of the total picking task volume of each partition, construct the feature representation of the total picking task volume of each partition with the minimum variance.
[0034] The third objective is constructed based on the feature representation that minimizes the variance of the total picking tasks in each partition.
[0035] In some embodiments, the objective includes a first objective and a third objective, and the objective for generating picking tasks for orders in the warehouse includes:
[0036] Based on the current order data and pooled order limit data of the warehouse, and combined with the variable of order allocation to pooled orders, construct the first objective that maximizes the ratio of the capacity of each pooled order to the maximum capacity of the pooled order.
[0037] Based on the set order constraint data and partition data, and combined with the variables of the set order including partitions, a third objective is constructed to balance the picking tasks of each partition.
[0038] In some embodiments, the constraints include one or more of a first constraint, a second constraint, and a third constraint; the aggregate order limit data includes the maximum number of aggregate orders and the maximum capacity of aggregate orders; and the constraints for generating picking tasks for orders in the warehouse include:
[0039] Under the maximum number of batch orders, the variables of the orders assigned to the batch orders are summed, and the summation result is constrained to form the first constraint condition for each order to be assigned to a batch order;
[0040] Given the current number of orders in the warehouse, calculate the capacity of each collection order based on the variable of order allocation to collection orders, and use the maximum capacity of collection orders to constrain the capacity of each collection order, so as to form a second constraint condition that limits the capacity of collection orders.
[0041] For two specified orders, the variables of the order allocation to the same set of orders are summed, and the summation result is constrained to form a third constraint condition that the two specified orders are not allocated to the same set of orders.
[0042] In some embodiments, the maximum capacity of a bundled order includes one or more of the maximum volume of a bundled order, the maximum weight of a bundled order, and the maximum order quantity of a bundled order. The second constraint limiting the capacity of a bundled order includes one or more of the following:
[0043] Given the current number of orders in the warehouse, sum the variables of orders assigned to the collection order to obtain the characteristic representation of the order quantity of each collection order. Using the maximum order quantity of the collection order, constrain the characteristic representation of the order quantity of each collection order to form a second constraint condition that limits the order quantity of the collection order.
[0044] Given the current number of orders in the warehouse, the product of the variable that the orders are assigned to the collection order and the volume of the orders is summed to obtain the characteristic representation of the volume of each collection order. Using the maximum volume of the collection order, the characteristic representation of the volume of each collection order is constrained to form a second constraint condition that limits the volume of the collection order.
[0045] Given the current number of orders in the warehouse, the product of the variable that assigns the order to the collection order and the weight of the order is summed to obtain the characteristic representation of the weight of each collection order. Using the maximum weight of the collection order, the characteristic representation of the weight of each collection order is constrained to form a second constraint condition that limits the weight of the collection order.
[0046] In some embodiments, the solution of the decision variables that satisfy the objective and the constraints is obtained using an adaptive large-scale neighborhood search algorithm.
[0047] Some embodiments of this disclosure provide a picking task generation apparatus, including: a memory; and a processor coupled to the memory, the processor being configured to execute a picking task generation method based on instructions stored in the memory.
[0048] This disclosure provides an apparatus for generating picking tasks, comprising: a module for executing a picking task generation method.
[0049] This disclosure provides some embodiments of a picking system, including:
[0050] The picking task generation device is configured to execute the picking task generation method and send the generated picking tasks to the picking equipment;
[0051] The picking equipment is configured to perform picking based on the picking tasks issued.
[0052] Some embodiments of this disclosure provide a computer-readable storage medium having computer instructions stored thereon, which, when executed by a processor, implement a picking task generation method.
[0053] Some embodiments of this disclosure provide a computer program product including computer instructions that, when executed by a processor, implement a picking task generation method. Attached Figure Description
[0054] The accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. This disclosure can be more clearly understood from the following detailed description with reference to the accompanying drawings.
[0055] Obviously, the accompanying drawings described below are merely some embodiments of this disclosure. Those skilled in the art can obtain other drawings based on these drawings without any creative effort.
[0056] Figure 1 A flowchart illustrating a picking task generation method according to some embodiments of this disclosure is shown.
[0057] Figure 2 A schematic diagram of the structure of a picking task generation apparatus according to some embodiments of the present disclosure is shown.
[0058] Figure 3 A schematic diagram of the structure of a picking task generation apparatus according to some embodiments of the present disclosure is shown.
[0059] Figure 4 A schematic diagram of a picking system according to some embodiments of the present disclosure is shown. Detailed Implementation
[0060] It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values of the components and steps set forth in these embodiments do not limit the scope of this disclosure.
[0061] Those skilled in the art will understand that the terms "first," "second," etc., in the embodiments of this disclosure are only used to distinguish different steps, devices, or modules, and do not represent any specific technical meaning, nor do they indicate a necessary logical order between them.
[0062] It should also be understood that in the embodiments disclosed herein, "a plurality of" may refer to two or more, and "at least one" may refer to one, two or more.
[0063] It should also be understood that any component, data or structure mentioned in the embodiments of this disclosure can generally be understood as one or more unless expressly defined or given to the contrary in the context.
[0064] Furthermore, the term "and / or" in this disclosure is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this disclosure generally indicates that the preceding and following related objects have an "or" relationship.
[0065] It should also be understood that the description of the various embodiments in this disclosure emphasizes the differences between the various embodiments, and the similarities or similarities can be referred to each other. For the sake of brevity, they will not be described in detail.
[0066] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.
[0067] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit this disclosure or its application or use.
[0068] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.
[0069] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.
[0070] Furthermore, to avoid obscuring this disclosure with unnecessary detail, only processing steps and / or apparatus structures closely related to at least the solutions according to this disclosure are shown in the accompanying drawings, while other details not closely related to this disclosure are omitted. It should also be noted that similar reference numerals and letters in the drawings indicate similar items, and therefore once an item is defined in one drawing, it need not be discussed again in subsequent drawings.
[0071] This disclosure proposes an automatic picking task generation scheme. Based on real-time information such as order requirements and warehouse conditions, and combined with specific decision variables, the scheme automatically breaks down orders into picking tasks. The generated picking tasks can meet the predetermined goals and constraints, optimize the picking tasks, and have strong executability.
[0072] Figure 1 The diagram illustrates a flowchart of a picking task generation method according to some embodiments of this disclosure. This picking task generation method can be executed, for example, by a picking task generation device, which can be, for example, a warehouse intelligent planning system or a module thereof. Figure 1 As shown, the picking task generation method of this embodiment includes the following steps.
[0073] In step 110, the orders entering the warehouse and their order information are obtained.
[0074] Orders are the starting point of the entire picking process; orders entering the warehouse are those awaiting picking. Order information includes, for example, a unique identifier for each order, order creation time, customer information, order priority, and information about the items included in the order. Item information may include, for example, SKUs (Stock Keeping Units). The accuracy and timeliness of order information are crucial, directly impacting the planning of subsequent consolidation orders and picking tasks.
[0075] Below are some examples of order information.
[0076] order_id: A unique identifier for an order, used to distinguish different orders.
[0077] priority: The priority of the order. Possible values include "high", "medium", etc., indicating the urgency of the order processing.
[0078] Items: Contains information for each item in the order, such as item ID, location, weight, and volume. Item_id: A unique identifier for the item. Location: The item's storage location in the warehouse, which can be represented by a shelf number. Weight: The item's weight, which can be used to calculate picking difficulty or time. Volume: The item's volume, ensuring that picking does not exceed the warehouse aisle and equipment capacity.
[0079] In step 120, the input data for planning the picking task is determined.
[0080] The input data includes: current warehouse order data, aggregate order restriction data, and partition data.
[0081] The warehouse's current order data includes a binary variable indicating the current number of orders and whether the orders include a partition. The aggregate order limit data includes the maximum number of aggregate orders and the maximum capacity of aggregate orders. The maximum capacity of an aggregate order includes one or more of the following: maximum volume, maximum weight, and maximum order quantity. The partition data includes the number of partitions and the number of picking tasks currently pending in each partition.
[0082] n: Order quantity is a basic indicator for measuring the current workload and directly affects the generation of consolidation orders and the allocation of picking tasks. The current order quantity in the warehouse can be determined based on the orders entering the warehouse. For example, the order quantity can be determined based on all orders entering the warehouse, or based on the orders entering the warehouse that need to be processed on the same day.
[0083] : A binary variable indicating whether order i includes partition s, used to represent whether order i involves partition s. If order i includes partition s, If the value is equal to 1, otherwise, if order i does not include partition s, =0. A partition is a logical area where the warehouse is divided. This variable helps determine the geographical distribution of orders and the composition of a collection of orders. Based on the items involved in the order and the partition of the warehouse where the items are stored, the value of the binary variable determining whether order i includes partition s is determined. For example, if order 01 involves items 001 and 002, and item 001 is stored in partition 1 of the warehouse, and item 002 is stored in partition 2 of the warehouse, then... =1, =1, =0. This relates to order details and warehouse storage status.
[0084] m: The maximum number of batch orders generated this time. This limits the number of batch orders generated each time, ensuring that the processing of orders does not exceed the capacity of the warehouse and equipment.
[0085] Q: The maximum order quantity for a batch order limits the number of orders in each batch order to prevent exceeding the processing capacity of operators or equipment.
[0086] V: Maximum volume of the bulk order limits the volume of each bulk order, ensuring that it does not exceed the physical limits of warehouse aisles and equipment during the picking process.
[0087] W: Maximum weight of the bundled order. This limits the weight of each bundled order to ensure that the bundled order does not exceed the equipment's carrying capacity during transportation and picking.
[0088] k: Number of zones, also known as the number of logical zones in the warehouse. It defines the warehouse's zoning structure and helps in planning and allocating picking tasks. k is related to the warehouse layout.
[0089] The current number of picking tasks awaiting pickup in partition s reflects the task load of each partition, helping to optimize task allocation to balance workload.
[0090] In step 130, the decision variables used to plan the picking task are determined.
[0091] Decision variables are used in decision models to represent different decision choices and states, thereby finding the optimal decision under complex constraints.
[0092] The decision variables include: variables for assigning orders to a collection order, and variables for the collection order comprising partitions. If necessary, the decision variables may also include a variable for whether a collection order is assigned an order. The value of the variable for whether a collection order is assigned an order can be determined based on the value of the variable for assigning orders to a collection order. One collection order and one of its constituent partitions correspond to one picking task.
[0093] The variable for assigning an order to a collection order indicates whether order i is assigned to collection order j, and is a crucial decision in the collection order generation process. This variable is a binary variable, such as a 0-1 variable; if order i is assigned to collection order j, ... If order i is not assigned to set j, .
[0094] The variable `j` indicates whether a pickup order (j) includes partitions (s). This helps determine the coverage of the pickup order and ensures the rationality of the picking task. This variable is a binary variable, such as a 0-1 variable; if pickup order j includes partition s, ... If set j does not include partition s, .
[0095] This variable indicates whether a batch order (j) has been assigned an order. It's used to determine the activity level of a batch order and helps optimize resource utilization. This variable is binary, such as a 0-1 variable. If batch order j is assigned an order, then batch order j is active. If order j in a collection is not assigned any orders, then order j in a collection is inactive. . The value can be determined according to The value is determined.
[0096] In step 140, based on the input data and the decision variables, the objectives and constraints for generating picking tasks for the warehouse orders are constructed to build a decision model.
[0097] Based on the input data and decision variables, the decision model can comprehensively describe the complexity of picking tasks and, in combination with predetermined goals and constraints, find a reasonable allocation of picking tasks to improve the overall operational efficiency of the warehouse.
[0098] The established goals may include, for example, a first goal, a second goal, a third goal, etc. For example, the goal may include a first goal and a second goal, or a first goal and a third goal, or a first goal, a second goal, and a third goal, but is not limited to the examples given.
[0099] First objective: Based on the current order data and aggregate order limit data of the warehouse, and combined with the variable of order allocation to aggregate orders, construct a first objective that maximizes the ratio of the capacity of each aggregate order to the maximum capacity of the aggregate order.
[0100] Based on the current number of orders in the warehouse and the variables of orders allocated to collection orders, construct a feature representation that maximizes the ratio of the capacity of each collection order to the maximum capacity of the collection order; under the maximum number of collection orders, sum the feature representations that maximize the ratio of the capacity of each collection order to the maximum capacity of the collection order to construct the first objective.
[0101] The process of constructing the feature representation that maximizes the ratio of the capacity of each aggregate order to the maximum capacity of the aggregate order includes: summing the variables of orders allocated to aggregate orders under the current order quantity in the warehouse to obtain the feature representation of the order quantity of each aggregate order, and constructing a first feature representation that represents the ratio of the feature representation of the order quantity of each aggregate order to the maximum order quantity of the aggregate order; summing the product of the variable of orders allocated to aggregate orders and the volume of orders under the current order quantity in the warehouse to obtain the feature representation of the volume of each aggregate order, and constructing a second feature representation that represents the ratio of the feature representation of the volume of each aggregate order to the maximum volume of the aggregate order; summing the product of the variable of orders allocated to aggregate orders and the weight of orders under the current order quantity in the warehouse to obtain the feature representation of the weight of each aggregate order, and constructing a third feature representation that represents the ratio of the feature representation of the weight of each aggregate order to the maximum weight of the aggregate order; and constructing a feature representation that maximizes one or more of the first, second, and third feature representations to maximize the ratio of the capacity of each aggregate order to the maximum capacity of the aggregate order.
[0102] The primary goal is: Where min represents minimizing the first objective function , ,in, It is a characteristic representation of the order quantity of each collection order. Maximum order quantity of aggregated orders The first characteristic of the proportion is represented by, This represents the volume of order i. It is a characteristic representation of the volume of each set unit. Maximum volume of a single set The second characteristic of the proportion is represented by, This represents the weight of order i. It is a characteristic representation of the weight of each set. Maximum weight of a single collection The third characteristic of the proportion is represented by max, which means maximization. This indicates the weight of the first objective in the overall optimization objective; the specific value can be configured. The meanings of other symbols are described in steps 120 and 130 above, and will not be repeated here.
[0103] The primary objective is to maximize the order volume of bundled orders. Given a fixed number of orders, this reduces the number of bundled orders, thereby decreasing the number of trips picking equipment (such as robots) makes within the warehouse and reducing congestion. Specifically, maximizing the order volume of each bundled order by ensuring its volume, weight, and order quantity are close to its maximum capacity (V, W, Q) improves the efficiency of each picking operation.
[0104] Second objective: Based on the constraints and partition data of the bundled orders, and combined with the variables of the bundled orders including partitions, construct a second objective that minimizes the number of picking tasks split into bundled orders.
[0105] An exemplary approach: Under the conditions of the number of partitions and the maximum number of collection orders, sum the variables of collection orders including partitions to obtain a feature representation of the number of picking tasks; under the condition of the maximum number of collection orders, sum the variables of whether collection orders are assigned to orders to obtain a feature representation of the number of collection orders; construct the second objective based on the feature representation that minimizes the ratio of the feature representation of the number of picking tasks to the feature representation of the number of collection orders.
[0106] Another exemplary approach: Under the conditions of order quantity and maximum number of collection orders, sum the variables of orders assigned to collection orders to obtain a feature representation of the number of assigned orders; under the condition of maximum number of collection orders, first sum the variables of orders assigned to collection orders, and under the conditions of maximum number of collection orders and number of partitions, sum the product of the binary variable of whether an order includes a partition and the summation result to obtain a feature representation of the total number of partitions of the assigned orders; based on the feature representation that minimizes the ratio of the feature representation of the number of picking tasks to the feature representation of the number of collection orders, and the feature representation that minimizes the ratio of the feature representation of the total number of partitions of the assigned orders to the feature representation of the number of assigned orders, the second objective is constructed.
[0107] The second objective is: Where min represents minimizing the second objective function , ,in, You can optionally enable or disable this setting. This is a characteristic representation of the quantity of picking tasks. It is a characteristic representation of the single quantity of a set, by minimizing Optimize the number of cross-zones in a batch order and reduce the number of picking tasks split into batch orders, so as to reduce the number of zone crossings and dwell times within zones and improve picking efficiency. This is a characteristic representation of the total number of partitions for allocating orders. It is a characteristic representation of the number of assigned orders, by minimizing By reducing the number of cross-regional orders within a single order, the number of cross-regional orders in a bundled order is reduced, thus decreasing the number of picking tasks that a bundled order requires to be split. This optimizes the cross-regional characteristics of bundled orders by adjusting the allocation of orders across different regions, ensuring the rationality of bundled orders. This indicates the weight of the second objective in the overall optimization objective; the specific value can be configured. The meanings of other symbols are described in steps 120 and 130 above, and will not be repeated here.
[0108] Third objective: Based on the bound data and partition data of the collection order, and combined with the variables of the collection order including the partition, construct a third objective that balances the picking tasks of each partition.
[0109] Under the maximum number of collection orders, the variables of the collection orders including the partitions are summed to obtain the feature representation of the number of picking tasks to be assigned to the partition; the feature representation of the number of picking tasks to be assigned to the partition and the number of picking tasks to be picked up in the partition are summed to obtain the feature representation of the total number of picking tasks in the partition; based on the feature representation of the total number of picking tasks in each partition, a feature representation with the minimum variance of the total number of picking tasks in each partition is constructed; based on the feature representation with the minimum variance of the total number of picking tasks in each partition, the third objective is constructed.
[0110] The third objective is: Where min represents minimizing the third objective function , ,in, This is a characteristic representation of the number of picking tasks to be assigned in partition s. This represents the current number of picking tasks awaiting pickup in partition s. This is a characteristic representation of the total picking tasks in partition s. This represents the variance of the total picking tasks in each partition s. This indicates the weight of the third objective in the overall optimization objective; the specific value can be configured. The meanings of other symbols are described in steps 120 and 130 above, and will not be repeated here. By minimizing the variance of the total picking tasks in each partition, the picking task volume in each partition is made more balanced, improving the overall picking efficiency of the warehouse.
[0111] Constraints may include, for example, a first constraint on order allocation, a second constraint on the single capacity of the set, and a third constraint on the mutual exclusivity of orders.
[0112] These constraints ensure that each order is appropriately allocated to the collection order, while meeting limitations such as capacity, weight, and order quantity, and avoiding conflicts between orders. These constraints work together to ensure the feasibility of the picking task.
[0113] The first constraint on order allocation is as follows: Given the maximum number of orders in a set, sum the variables for allocating orders to that set, and then constrain the summation result to form the first constraint for allocating each order to a set. The formula is expressed below; the meaning of each symbol is explained in steps 120 and 130 above, and will not be repeated here.
[0114]
[0115] This ensures that each order is assigned to, and only to, a single set of orders, avoiding omissions and duplicate assignments.
[0116] The second constraint on the capacity of a collection order: Given the current number of orders in the warehouse, calculate the capacity of each collection order based on the variable of orders allocated to collection orders, and use the maximum capacity of the collection order to constrain the capacity of each collection order, thus forming the second constraint limiting the capacity of the collection order. The second constraint may include one or more of the following:
[0117] Given the current number of orders in the warehouse, the variables for allocating orders to aggregate orders are summed to obtain a characteristic representation of the order quantity for each aggregate order. Using the maximum order quantity for an aggregate order, the characteristic representation of the order quantity for each aggregate order is constrained to form a second constraint condition limiting the order quantity of an aggregate order. The formula is expressed below; the meaning of each symbol is described in steps 120 and 130 above, and will not be repeated here.
[0118]
[0119] The number of orders in each pool order cannot exceed the maximum limit (Q). This helps control the size of each pool order and avoids overloading.
[0120] Given the current number of orders in the warehouse, the product of the variable assigning the order to a collection order and the volume of the order is summed to obtain a characteristic representation of the volume of each collection order. Using the maximum volume of the collection order, the characteristic representation of the volume of each collection order is constrained to form a second constraint condition limiting the volume of the collection order. The formula is expressed as follows, and the meaning of each symbol is described in steps 120-140 above, which will not be repeated here.
[0121]
[0122] The total volume of each collection order must not exceed the maximum volume (V) of the picking container of the picking equipment (robot) to ensure that the robot can accommodate all the assigned items during the picking process.
[0123] Given the current number of orders in the warehouse, the product of the variable assigning the order to the collection order and the order's weight is summed to obtain a characteristic representation of the weight of each collection order. Using the maximum weight of the collection order, the characteristic representation of the weight of each collection order is constrained to form a second constraint condition limiting the weight of the collection order. The formula is expressed below; the meaning of each symbol is described in steps 120-140 above, and will not be repeated here.
[0124]
[0125] The total weight of each assembly must not exceed the maximum weight limit (W) to ensure that the robot does not exceed its load capacity during operation.
[0126] The third constraint on order mutual exclusion: For two specified orders, sum the variables related to the order allocation to the same set of orders, and then constrain the sum to form a third constraint that prevents the two specified orders from being allocated to the same set of orders. The formula is expressed as follows:
[0127]
[0128] in, This represents the current order set in the warehouse. Indicates order Whether it is assigned to set j, Indicates order Whether it is assigned to set j, and the meaning of other symbols are described in steps 120-140 above, and will not be repeated here.
[0129] Some orders are mutually exclusive and cannot be assigned to the same set of orders. This constraint prevents potentially conflicting or incompatible orders from being processed simultaneously.
[0130] These constraints ensure the rationality and effectiveness of task allocation, supporting the achievement of overall optimization goals.
[0131] In step 150, the solution for the decision variables that satisfy the objective and the constraints is obtained.
[0132] For example, the Adaptive Large Neighborhood Search (ALNS) algorithm can be used to solve the problem and obtain the optimal solution for the decision variables that satisfies the objective and the constraints. The ALNS algorithm is a powerful metaheuristic algorithm that combines the advantages of local search and large neighborhood search. It can dynamically adjust the size and structure of the search neighborhood to efficiently find the optimal solution.
[0133] The ALNS algorithm starts with a seed algorithm to construct an initial solution. This initial solution can be generated using a simple heuristic method, ensuring its feasibility and rationality. Next, the ALNS algorithm enters the optimization phase. In the optimization phase, the algorithm improves the initial solution through a series of neighborhood operations. Neighborhood operations include removing and inserting tasks, swapping path segments, etc. These operations aim to explore different regions of the solution space and break the local optima of the current solution.
[0134] The ALNS algorithm possesses an adaptive mechanism. Based on the quality of the current solution and feedback during the search process, ALNS dynamically adjusts the probability of neighborhood selection and the search strategy. This adaptability enables the ALNS algorithm to effectively explore different regions of the solution space, thereby improving its global search capability. Through continuous iteration and optimization, the ALNS algorithm gradually approaches the global optimum of the problem.
[0135] The initial solution algorithm is described below.
[0136] The initial solution algorithm provides a valid starting point for subsequent optimization processes through a reasonable order allocation strategy. A seed algorithm based on a gain function is employed to construct the initial solution. This process begins by calculating the potential gain between each order and the set of orders. Specifically, the potential gain of assigning each order to different sets of orders is evaluated, and these gain values are stored in a set for decision-making.
[0137] The algorithm first checks if the set of orders to be assigned is empty. If all orders have been successfully assigned, the algorithm terminates and returns the current initial solution. If there are still orders to be assigned, the algorithm continues to evaluate whether there are valid options for the gain value, i.e., whether there are any combinations that can bring positive gain. If there is no valid gain, the algorithm checks if there is an empty set of orders. If there is an empty set of orders, it selects an order as a seed order and assigns it to that set of orders, thus initiating the construction process of that set of orders.
[0138] When an effective gain is available, the algorithm selects the order with the largest gain and combines it with the corresponding set of orders, then assigns the order to the set of orders. After each assignment, the set of orders, the order set, and the gain set are updated to reflect the latest state. This process is repeated until all orders have been assigned.
[0139] This gain-maximizing strategy not only considers the rationality of order allocation in constructing the initial solution but also provides a solid foundation for subsequent optimization. This strategy ensures the validity and feasibility of the initial solution in the global search, thus laying the groundwork for improving the overall efficiency of the picking task.
[0140] The initial solution algorithm, also known as the seed algorithm, involves the following steps to construct an initial solution.
[0141] Input: Collection Single Orders pending allocation Gain function ,in All are empty.
[0142] Step 1: Calculate all effective gains Stored as a gain set .
[0143] Effective gain is calculated using a gain function. The gain function quantifies the specific benefits of allocating a particular order to a specific batch of orders. These benefits may manifest in multiple ways, such as reducing delivery costs, improving delivery efficiency, and reducing waiting time. Effective gain can be a comprehensive indicator, calculated by combining multiple factors.
[0144] An example of a gain function is:
[0145] ,
[0146] in, This is the cost of the current allocation plan. This is the new cost after allocating orders to a certain batch of orders. This reflects cost savings. This is the total time for the current plan (e.g., picking time). This is the total time for the new plan. This reflects an improvement in time efficiency. This refers to resource utilization rate (such as the utilization rate of equipment or personnel), which can be a positive value indicating an improvement in resource utilization. , , It is a weighting coefficient used to adjust the degree of influence of different factors on the gain.
[0147] Step 2: Determine if the order to be assigned is empty. If the order to be assigned is empty, i.e. If the condition is met, proceed to step 7; otherwise, proceed to step 3.
[0148] Step 3: Determine if an effective gain exists. If an effective gain exists, i.e. If the condition is met, proceed to step 6; otherwise, proceed to step 4.
[0149] Step 4: Determine if an empty set exists. If an empty set exists, i.e. If the condition is met, proceed to step 5; otherwise, proceed to step 6.
[0150] Step 5: From Select a seed order and assign it to an empty set. ,renew , , , Return and execute step 2.
[0151] Step 6: Calculate the gain pair with the maximum gain, i.e. , order Add to collection list In progress, updates , , , Return and execute step 2.
[0152] Step 7: Return the set of orders and the orders to be assigned as the initial solution.
[0153] The ALNS algorithm is introduced below. The ALNS algorithm is used to optimize the initial solution of warehouse picking tasks.
[0154] The ALNS algorithm sets a feasible initial solution as the current optimal solution. Simultaneously, it assigns initial equal weights to both the destruction and reconstruction operations.
[0155] In each iteration, the ALNS algorithm randomly selects a destruction operation based on these weights to partially destroy the current solution, thereby removing some orders or changing their allocation. Next, the algorithm selects a reconstruction operation to repair the destroyed solution and generate a new candidate solution.
[0156] This new solution will be evaluated, and if it outperforms the current solution or meets certain acceptance criteria (such as simulated annealing), then the new solution will replace the current solution. Conversely, if the new solution is better than the current best solution, then the best solution will be updated.
[0157] To improve the efficiency and adaptability of the ALNS algorithm, the weights of the destruction and reconstruction operations are dynamically adjusted based on the performance of the solutions. This adjustment allows the algorithm to explore different solution spaces more effectively.
[0158] The iterative process continues until certain exit conditions are met, such as reaching the maximum computation time or the optimal solution no longer being updated after multiple iterations. Ultimately, the algorithm outputs an optimized best solution, providing an efficient solution for warehouse picking tasks.
[0159] The iterative optimization process of the ALNS algorithm is as follows:
[0160] First, construct a feasible solution. .
[0161] Next, let the optimal solution be... , To disrupt operator weights, To reconstruct operator weights.
[0162] Then, the following loop is executed iteratively until the exit condition is met (such as computation time, the optimal solution not being updated after a certain number of iterations, etc.).
[0163] The loop is as follows:
[0164] By weight Randomly select sabotage operation By weight Select rebuild operation Candidate solutions are obtained ;
[0165] If accepted (For example, accepting when the objective function is smaller, or using simulated annealing to determine acceptance), let ;
[0166] if The objective function is less than ,make ;
[0167] based on( Update weights , .
[0168] Finally, after the loop ends, return the current best solution. .
[0169] In step 160, based on the solutions of the decision variables that satisfy the objective and the constraints, the picking tasks corresponding to the warehouse orders are generated.
[0170] Based on the optimal solution of decision variables such as the variable of the order being assigned to the collection order and the variable of the collection order including the partitions, it can be determined which collection order a certain order is specifically assigned to, which partitions a certain collection order includes, and a collection order and its included partitions correspond to a picking task, thereby generating the corresponding picking tasks for the warehouse orders.
[0171] The generated picking task includes information such as the picking task identifier, the partitions and items involved in the picking task, the task time requirements, priority, and the assigned object (a picking robot).
[0172] After generating a picking task, the task can be sent to the picking equipment, instructing it to perform picking based on the task.
[0173] This embodiment automatically breaks down orders into picking tasks based on real-time information such as order requirements and warehouse conditions, combined with specific decision variables. The generated picking tasks meet predetermined goals and constraints, thus optimizing the picking process and demonstrating strong executability. Setting different goals allows for optimization of the picking tasks from different perspectives; for example, see the optimization effects of the aforementioned first, second, and third goals.
[0174] Figure 2 A schematic diagram of the structure of a picking task generation apparatus according to some embodiments of this disclosure is shown. For example... Figure 2 As shown, the picking task generation apparatus 200 of this embodiment includes: a memory 210 and a processor 220 coupled to the memory 210. The processor 220 is configured to execute the picking task generation method in the aforementioned embodiments based on instructions stored in the memory 210.
[0175] The picking task generation device 200 may also include an input / output interface 230, a network interface 240, a storage interface 250, etc. These interfaces 230, 240, 250, as well as the memory 210 and the processor 220, can be connected, for example, via a bus 260.
[0176] The memory 210 may include, for example, system memory, fixed non-volatile storage media, etc. The system memory stores, for example, the operating system, application programs, boot loader, and other programs.
[0177] The processor 220 can be implemented using a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gates, or transistors, or other discrete hardware components.
[0178] The input / output interface 230 provides a connection interface for input / output devices such as monitors, mice, keyboards, and touchscreens. The network interface 240 provides a connection interface for various networked devices. The storage interface 250 provides a connection interface for external storage devices such as SD cards and USB flash drives. The bus 260 can use any bus architecture from various bus structures. For example, bus architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MCA) bus, and Peripheral Component Interconnect (PCI) bus.
[0179] Figure 3 A schematic diagram of the structure of a picking task generation apparatus according to some embodiments of this disclosure is shown. For example... Figure 3 As shown, the picking task generation apparatus 300 of this embodiment includes a module for executing a picking task generation method.
[0180] The input data module 310 is configured to determine input data, which includes: current warehouse order data, collection order restriction data, and partition data;
[0181] The decision variable module 320 is configured to determine decision variables, which include: variables for order allocation to a collection order and variables for the collection order including partitions, wherein a collection order and its included partitions correspond to a picking task;
[0182] The construction module 330 is configured to construct the objectives and constraints for generating picking tasks for the warehouse orders based on the input data and the decision variables.
[0183] The task generation module 340 is configured to generate picking tasks corresponding to the orders in the warehouse based on the solutions of the decision variables that satisfy the objective and the constraints.
[0184] The construction module 330 includes a target construction unit 331 and a constraint construction unit 332.
[0185] The objective building unit 331 is configured to, based on the current order data and bundled order limit data of the warehouse, and combined with the variable of order allocation to bundled orders, construct a first objective that maximizes the ratio of the capacity of each bundled order to the maximum capacity of the bundled order; and / or, based on the bundled order limit data and partition data, and combined with the variable of bundled orders including partitions, construct a second objective that minimizes the number of picking tasks split from the bundled order; and / or, based on the bundled order limit data and partition data, and combined with the variable of bundled orders including partitions, construct a third objective that balances the picking tasks of each partition.
[0186] The constraint construction unit 332 is configured to, under the maximum number of aggregate orders, sum the variables of orders assigned to aggregate orders and constrain the summation result to form a first constraint condition that each order is assigned to an aggregate order; and / or, under the current number of orders in the warehouse, calculate the capacity of each aggregate order based on the variables of orders assigned to aggregate orders, and constrain the capacity of each aggregate order using the maximum capacity of aggregate orders to form a second constraint condition that limits the capacity of aggregate orders; and / or, for two specified orders, sum the variables of orders assigned to aggregate orders and constrain the summation result to form a third constraint condition that the two specified orders are not assigned to the same aggregate order.
[0187] Figure 4 Schematic diagrams of picking systems according to some embodiments of this disclosure are shown. Figure 4 As shown, the picking system 400 of this embodiment includes: a picking task generation device 410, configured to execute a picking task generation method and send the generated picking tasks to picking equipment; and a picking equipment 420, configured to perform picking according to the issued picking tasks. The picking task generation device 410 is, for example, a picking task generation device 200 or 300. The picking equipment 420 is, for example, a picking robot or other picking equipment.
[0188] The technical solution disclosed herein involves the collection, updating, analysis, processing, use, transmission, and storage of user personal information (such as phone number, delivery address, etc.) related to orders, all of which comply with relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. Necessary measures are taken to prevent unauthorized access to user personal information data and to safeguard user personal information security, network security, and national security.
[0189] Those skilled in the art will understand that embodiments of this disclosure can be provided as methods, systems, or computer program products. Therefore, this disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this disclosure can take the form of a computer program product embodied on one or more (non-transitory) computer-readable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, cloud storage, etc.) containing computer program code. A computer program product should be understood as a software product that primarily implements its solution through a computer program.
[0190] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0191] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0192] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
Claims
1. A method for generating picking tasks, comprising: Determine the input data, which includes: current warehouse order data, collection order restriction data, and partition data; Determine the decision variables, which include: the variable of order allocation to the collection order, and the variable of the collection order including the partitions, wherein a collection order and its included partitions correspond to a picking task; Based on the input data and the decision variables, construct the objectives and constraints for generating picking tasks for the warehouse's orders; Based on the solutions of the decision variables that satisfy the stated objective and the stated constraints, the corresponding picking tasks for the warehouse orders are generated.
2. The picking task generation method according to claim 1, wherein, The objectives include a first objective and a second objective, and the objectives for generating picking tasks for orders in the warehouse include: Based on the current order data and pooled order limit data of the warehouse, and combined with the variable of order allocation to pooled orders, construct the first objective that maximizes the ratio of the capacity of each pooled order to the maximum capacity of the pooled order. Based on the constraints and partition data of the bundled orders, and combined with the variables of the bundled orders including partitions, a second objective is constructed that minimizes the number of picking tasks when the bundled orders are split.
3. The picking task generation method according to claim 2, wherein, The current order data for the warehouse includes the current number of orders in the warehouse, and the aggregate order limit data includes the maximum number of aggregate orders and the maximum capacity of aggregate orders. The first objective, which maximizes the ratio of the capacity of each aggregate order to the maximum capacity of the aggregate order, includes: Based on the current number of orders in the warehouse and the variables of orders assigned to collection orders, construct a feature representation that maximizes the ratio of the capacity of each collection order to the maximum capacity of the collection order; Under the maximum number of sets, the feature representations that have the largest ratio between the capacity of each set and the maximum capacity of the set are summed to construct the first objective.
4. The picking task generation method according to claim 3, wherein, The maximum capacity of a collection order includes one or more of the following: maximum volume of the collection order, maximum weight of the collection order, and maximum order quantity of the collection order. The feature representation that maximizes the ratio of each collection order's capacity to its maximum capacity includes: Given the current number of orders in the warehouse, sum the variables of orders assigned to collection orders to obtain the feature representation of the order quantity of each collection order, and construct the first feature representation of the ratio of the feature representation of the order quantity of each collection order to the maximum order quantity of the collection order; Given the current number of orders in the warehouse, the product of the variable that the order is assigned to a collection order and the volume of the order is summed to obtain the feature representation of the volume of each collection order. A second feature representation is constructed by comparing the feature representation of the volume of each collection order with the maximum volume of the collection order. Given the current number of orders in the warehouse, sum the product of the variable that the order is assigned to a collection order and the weight of the order to obtain the feature representation of the weight of each collection order. Construct a third feature representation of the ratio of the feature representation of the weight of each collection order to the maximum weight of the collection order. Based on the feature representation that maximizes one or more of the first feature representation, the second feature representation, and the third feature representation, construct the feature representation that maximizes the ratio of the capacity of each set to the maximum capacity of the set.
5. The picking task generation method according to claim 2, wherein, The partition data includes the number of partitions, the aggregate order limit data includes the maximum number of aggregate orders, and the decision variables also include whether the aggregate order is assigned an order. Constructing a second objective that minimizes the number of picking tasks resulting from splitting aggregate orders includes: Given the number of partitions and the maximum number of collection orders, the variables of collection orders including partitions are summed to obtain a characteristic representation of the number of picking tasks; Under the maximum number of set orders, sum the variables of whether set orders are assigned to orders to obtain a characteristic representation of the number of set orders; The second objective is constructed based on the feature representation that minimizes the ratio of the feature representation of the number of picking tasks to the feature representation of the number of set orders.
6. The picking task generation method according to claim 5, wherein, The warehouse's current order data includes a binary variable indicating the current number of orders in the warehouse and whether the orders include partitions. The aggregate order limit data includes the maximum number of aggregate orders. Constructing the second objective includes: Given the number of orders and the maximum number of aggregate orders, the variables representing the number of orders allocated to aggregate orders are summed to obtain a characteristic representation of the number of allocated orders. Under the maximum number of set orders, first sum the variables of orders assigned to set orders, and under the maximum number of set orders and the number of partitions, sum the product of the binary variable of whether the order includes a partition and the summation result to obtain the characteristic representation of the total number of partitions of the assigned orders; The second objective is constructed based on the feature representation that minimizes the ratio of the feature representation of the number of picking tasks to the feature representation of the number of set orders, and the feature representation that minimizes the ratio of the feature representation of the total number of partitions for order allocation to the feature representation of the number of allocated orders.
7. The picking task generation method according to claim 2, wherein, The objectives also include a third objective, which is to construct the objective of generating picking tasks for orders in the warehouse, including: Based on the set order constraint data and partition data, and combined with the variables of the set order including partitions, a third objective is constructed to balance the picking tasks of each partition.
8. The picking task generation method according to claim 7, wherein, The aggregate order limit data includes the maximum number of aggregate orders, and the partition data includes the number of picking tasks currently pending in each partition. The third objective of achieving a balanced picking workload across partitions includes: Under the maximum number of batch orders, the variables of the batch orders, including the partitions, are summed to obtain a characteristic representation of the number of picking tasks to be assigned to each partition; The feature representation of the number of picking tasks to be assigned in a partition and the number of picking tasks to be picked up in the partition are summed to obtain the feature representation of the total number of picking tasks in the partition. Based on the feature representation of the total picking task volume of each partition, construct the feature representation of the total picking task volume of each partition with the minimum variance. The third objective is constructed based on the feature representation that minimizes the variance of the total picking tasks in each partition.
9. The picking task generation method according to claim 1, wherein, The objectives include a first objective and a third objective, and the objectives for generating picking tasks for orders in the warehouse include: Based on the current order data and pooled order limit data of the warehouse, and combined with the variable of order allocation to pooled orders, construct the first objective that maximizes the ratio of the capacity of each pooled order to the maximum capacity of the pooled order. Based on the set order constraint data and partition data, and combined with the variables of the set order including partitions, a third objective is constructed to balance the picking tasks of each partition.
10. The picking task generation method according to any one of claims 1-9, wherein, The constraints include one or more of the first, second, and third constraints. The aggregate order limit data includes the maximum number of aggregate orders and the maximum capacity of aggregate orders. The constraints for generating picking tasks for orders in the warehouse include: Under the maximum number of batch orders, the variables of the orders assigned to the batch orders are summed, and the summation result is constrained to form the first constraint condition for each order to be assigned to a batch order; Given the current number of orders in the warehouse, calculate the capacity of each collection order based on the variable of order allocation to collection orders, and use the maximum capacity of collection orders to constrain the capacity of each collection order, so as to form a second constraint condition that limits the capacity of collection orders. For two specified orders, the variables of the order allocation to the same set of orders are summed, and the summation result is constrained to form a third constraint condition that the two specified orders are not allocated to the same set of orders.
11. The picking task generation method according to claim 10, wherein, The maximum capacity of a batch order includes one or more of the maximum volume, maximum weight, and maximum order quantity of a batch order. The second constraint limiting the capacity of a batch order includes one or more of the following: Given the current number of orders in the warehouse, sum the variables of orders assigned to the collection order to obtain the characteristic representation of the order quantity of each collection order. Using the maximum order quantity of the collection order, constrain the characteristic representation of the order quantity of each collection order to form a second constraint condition that limits the order quantity of the collection order. Given the current number of orders in the warehouse, the product of the variable that the orders are assigned to the collection order and the volume of the orders is summed to obtain the characteristic representation of the volume of each collection order. Using the maximum volume of the collection order, the characteristic representation of the volume of each collection order is constrained to form a second constraint condition that limits the volume of the collection order. Given the current number of orders in the warehouse, the product of the variable that assigns the order to the collection order and the weight of the order is summed to obtain the characteristic representation of the weight of each collection order. Using the maximum weight of the collection order, the characteristic representation of the weight of each collection order is constrained to form a second constraint condition that limits the weight of the collection order.
12. The picking task generation method according to any one of claims 1-9, wherein, The solution for the decision variables that satisfies the objective and the constraints is obtained using an adaptive large-scale neighborhood search algorithm.
13. A picking task generation device, comprising: Memory; And a processor coupled to the memory, characterized in that the processor is configured to execute the picking task generation method according to any one of claims 1-12 based on instructions stored in the memory.
14. A picking task generation device, comprising: A module for performing the picking task generation method according to any one of claims 1-12.
15. A picking system, comprising: The picking task generation device is configured to perform the picking task generation method according to any one of claims 1-12 and send the generated picking task to the picking equipment; The picking equipment is configured to perform picking based on the picking tasks issued.
16. A computer-readable storage medium storing computer instructions thereon, characterized in that, When executed by a processor, the computer instructions implement the picking task generation method as described in any one of claims 1-12.
17. A computer program product comprising computer instructions, characterized in that, When executed by a processor, the computer instructions implement the picking task generation method as described in any one of claims 1-12.