Optimization Method for Order Picking Operations of Multi-Rotary Shelf Devices

Abstract

The present invention relates to the technical field of picking optimization of multi-rotary shelf devices, and discloses a picking operation optimization method for multi-rotary shelf devices, which includes the following steps: obtaining the optimal picking order of a single rotary shelf, calculating the picking time set of each rotary shelf, performing serialization conversion to obtain a picking time sequence, calculating and obtaining a collaborative picking order based on the transfer time interval and the picking time sequence, searching for the rotary shelves and goods corresponding to each element in the collaborative picking order, determining the set to which the goods belong, calculating the transfer time corresponding to the set to which the goods belong, accumulating to obtain the total operation time, and using it as an evaluation index to determine the optimal collaborative picking order. The picking operation optimization method for multi-rotary shelf devices of the present invention improves the collaborative operation efficiency of multi-rotary shelf devices, and enables them to obtain a picking path with the optimal execution efficiency and timeliness under complex constraint conditions such as picking batches, the number of goods in batches, and transfer time intervals.

Description

Technical Field

[0001] The present invention relates to the technical field of picking optimization of multi-rotary shelf devices, and particularly relates to a method for optimizing the picking operation of multi-rotary shelf devices. Background Art

[0002] With the improvement of people's living standards and the rapid development of modern logistics services, efficient warehousing devices and systems have become a new market demand and are applicable to complex logistics management. Rotary shelves are often configured in modern logistics warehousing systems due to their advantages such as fast turnover speed, high storage density, and high shipping efficiency, and are widely used for the scheduling and management of a large variety and quantity of goods. At the same time, when rotary shelves cooperate in operations, optimizing the picking path of goods is of great significance for improving the operation efficiency of rotary shelves and the management efficiency of warehousing systems.

[0003] The problem of optimizing the picking operation of rotary shelves is similar to the dynamic non-closed traveling salesman problem, and can be divided into single rotary shelves and multi-rotary shelves according to the research object. Domestic and foreign scholars have studied the picking problem of single rotary shelves by establishing mathematical models under different constraint conditions and using ant colony algorithms, particle swarm algorithms, or heuristic fusion algorithms, etc., and have achieved many meaningful results. With the increase in the demand for the storage volume of goods in the industrial field and the expansion of the single rotary shelf warehousing system, the shipping efficiency has been greatly reduced, which has greatly restricted the application of such warehousing systems.

[0004] The multi-rotary shelf device is composed of several rotary shelves combined and arranged side by side, allowing various types of goods to be mixed and loaded, and can rotate in any direction, forward or backward. Each shelf has the same-sized storage locations but different quantities. It adopts an overall drive method, and the drive mechanism is located in the internal elliptical area, and a corresponding stacker is fixed at the outbound end. The goods are loaded on trays with a fixed number of goods, and the horizontal transfer is completed by hanging the tray on the rotary chain device. When performing the picking operation, various types of goods corresponding to each tray are allowed to be mixed and loaded in the shelf. The console plans the picking path according to the numbers of the goods to be picked, controls the operation of the rotary shelves and the stacker, and sequentially completes the picking operation at the stacker. Regarding the research on the picking operation under multi-rotary shelves, on the one hand, when obtaining the optimal picking path, the main purpose is to improve the picking efficiency, and the constraint conditions are single, such as the number of stackers, the types of goods, etc. However, in actual picking operations, ensuring the classification accuracy of a large number of goods is equally important, and a regular operation mode of batches and fixed intervals can reduce errors during goods classification; on the other hand, most of the planning methods adopted have many random factors and a high time complexity, making it difficult to meet the timeliness requirements.

[0005] In summary, in the prior art, the execution efficiency of goods picking by multi-rotary shelf devices is low and the timeliness is poor. Summary of the Invention

[0006] The object of the present invention is to provide an optimization method for the picking operation of a multi-rotary shelf device in view of the above-mentioned deficiencies in the technology, so as to improve the collaborative operation efficiency of the multi-rotary shelf device, and to obtain a picking path with optimal execution efficiency and timeliness under complex constraint conditions such as picking batches, the number of goods in each batch, and transfer time intervals.

[0007] To achieve the above object, the optimization method for the picking operation of the multi-rotary shelf device involved in the present invention includes the following steps:

[0008] A) Obtain the optimal picking order of a single rotary shelf;

[0009] B) Calculate the picking time set of each rotary shelf according to the optimal picking order obtained in step A);

[0010] C) Perform serialization conversion according to the picking time set to obtain a picking time sequence;

[0011] D) Calculate and obtain the collaborative picking order based on the transfer time interval and the picking time sequence;

[0012] E) Search for the rotary shelves and goods corresponding to each element in the collaborative picking order, and determine the set to which the goods belong;

[0013] F) Calculate the transfer time corresponding to the set to which the goods belong, accumulate to obtain the total operation time, and use it as an evaluation index to determine the optimal collaborative picking order.

[0014] Preferably, in step A), from the initial coordinates G of the required goods on each rotary shelf n , search and determine the optimal picking order of each shelf where n is the shelf number, N is the number of shelves in the overall device, m is the number of goods of the target category on shelf n, and t m is the picking time of the last good on this shelf;

[0015] In step B), calculate the picking time set I n of each rotary shelf, I n ={t1,...,t a ,t b ,...,t m}, where t a ,t b are two adjacent moments;

[0016] In step C), if is an empty set, calculate the picking time sequence E n corresponding to the picking time set I n ;

[0017] In step D), calculate the collaborative picking order E coop , and combine them according to the transfer time interval and the picking time series to obtain

[0018] In step E), search for the collaborative picking order E coop For the corresponding carousel and goods of each element in it, determine the set S to which the goods G belongs n , and S n ←S n ∪G,

[0019] In step F), calculate the transfer time corresponding to the set S n , and accumulate it to obtain the total operation time T turn , and use it as an evaluation index. If T turn (S best )>T turn (S * ), then min T turn =T turn (S * ), thereby determining the optimal collaborative picking order, S best ←S * , where S * is the comparison sequence, and S best is the optimal sequence.

[0020] Preferably, in step F), determine the optimal collaborative picking order, update the picking task S. If length(S)≥U, where U is the total number of goods required for the picking task, end, otherwise return to step B).

[0021] Preferably, in step C), according to the single optimal picking order calculate the picking operation time without time interval constraints, denoted as the shortest picking time. When planning the picking operation, introduce a waiting time between two adjacent goods in sequence, extend the picking time interval between them, and it cannot be less than the minimum time interval. After referring to the transfer time interval t0, give a time interval △t. According to the corresponding goods picking time, calculate an equidistant sequence E = [e1, e2, e3,..., e m , and the time interval between the times corresponding to two adjacent elements in E is △t. The i-th element e i represents the goods quantity that can be picked at the time of (i - 1)△t, that is, at the time of t(e i ). There is e i ≥0, and when e iWhen it is ≥2, it means that multi-goods shipping can be carried out at this moment, thus converting I n into a picking time series E with positive integer element values n , where e1 corresponds to the initial picking moment t(e i ), t a , t b The time interval △t between two adjacent moments satisfies a given time interval △t. If this transfer time interval △t is greater than the required transfer time interval t0, and meets L = [(t a -t b ) / t0], it means that between t a , t b L goods can also be picked at t0, so L zeros need to be added to E n to indicate that it can be combined with other rotary shelves to meet collaborative transfer.

[0022] Preferably, in step D), after the picking time set I n is converted into a picking time series E n , according to the given transfer time interval t0, the number of goods in batches a0, the situations of collaborative combination of E n include: First, the interval Δt in the time series satisfies Δt ≤ t0; Second, the element value e in the time series satisfies e ≤ a0. Thus, the picking time series of multiple rotary shelf devices are traversed and combined to obtain a set of collaborative picking orders E coop , denoted as The length of this sequence is the number of picking times, and each element value is greater than or equal to the number of goods in batches.

[0023] Compared with the prior art, the present invention has the following advantages:

[0024] 1. It improves the collaborative operation efficiency of multiple rotary shelf devices, enabling it to obtain a picking path with optimal execution efficiency and timeliness under complex constraint conditions such as picking batches, the number of goods in batches, and transfer time intervals;

[0025] 2. It can be extended and applied to the scheduling model of a rotary storage system under multiple modules. The proposed collaborative combination optimization algorithm can evaluate the picking capabilities under different shelf configurations of the system, providing a basis for the operation performance analysis of multiple rotary shelf devices. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is a schematic structural diagram of the multiple rotary shelf devices used in the picking operation optimization method of the multiple rotary shelf devices of the present invention;

[0027] Figure 2 is a schematic diagram of the combination of time series in the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0028] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0029] An optimization method for the picking operation of a multi-rotary shelf device includes the following steps:

[0030] A) Obtain the optimal picking order of a single rotary shelf. From the initial coordinates G of the required goods on each rotary shelf n , search and determine the optimal picking order of each shelf In the formula, n is the shelf number, N is the number of shelves of the overall device, m is the number of goods of the target category on shelf n, and t m is the picking time of the last good on this shelf;

[0031] B) Calculate the picking time set I of each rotary shelf through the optimal picking order obtained in step A) n , I n ={t1,...,t a ,t b ,...,t m}, in the formula, t a ,t b are two adjacent moments;

[0032] C) Perform serialization conversion according to the picking time set to obtain a picking time sequence. If there is an empty set, calculate the picking time sequence E corresponding to the picking time set I n . According to the single optimal picking order n , calculate the picking operation time without time interval constraints, denoted as the shortest picking time. When planning the picking operation, introduce a waiting time between two adjacent goods in sequence, extend the picking time interval between them, and it cannot be less than the minimum time interval. After referring to the transfer time interval t0, given a time interval △t, according to the corresponding goods picking time, calculate an equally spaced sequence E = [e1, e2, e3,..., e . The time interval between the moments corresponding to two adjacent elements in E is △t. The i-th element e m represents the number of goods that can be picked at the moment of (i - 1)△t, that is, at the moment t(e i ). There is e i ≥0, and when e i ≥2, it means that multi-good shipping can be carried out at this moment. Thus, convert I i into a picking time sequence E n with positive integer element values, and the process is as follows: n The process is as follows:

[0033]

[0034] Among them, e1 corresponds to the initial picking time t(e i ), t a , t b Between two adjacent times, a given time interval Δt is satisfied. If the transfer time interval Δt is greater than the required transfer time interval t0, and it meets L = [(t a -t b ) / t0], it means that between t a , t b , L items can be picked at t0. Therefore, L zeros need to be added to E n to indicate that it can be combined with other rotary shelves to meet the collaborative transfer;

[0035] D) Calculate the collaborative picking order E coop according to the transfer time interval and the picking time sequence. Based on the transfer time interval and the picking time sequence, the picking time set I is combined and converted into the picking time sequence E n . After that, according to the given transfer time interval t0 and the number of items in a batch a0, the situations of collaborative combination for E n include: First, the interval Δt in the time sequence satisfies Δt ≤ t0; Second, the element value e in the time sequence satisfies e ≤ a0. Thus, the picking time sequences of multiple rotary shelf devices are traversed and combined to obtain a set of collaborative picking orders E n , denoted as coop . The length of this sequence is the number of picking times, and each element value is greater than or equal to the number of items in a batch. In this way, the collaborative operation between each rotary shelf is represented. Taking the picking operation of items with a transfer time interval of 2Δt and a number of items in a batch of 2 as an example, a combination process of the picking time sequence is as shown. The picking time sequences E1, E2, E3, E4 with an interval of 4Δt can be combined into E Figure 2 . When traversing and combining N rotary shelves, there are coop1 kinds of combination results for multiple rotary shelves, where u is the number of optional picking orders for each shelf. Among them, the element value in the combination sequence may be greater than the number of items in a batch. At this time, reduced picking can be performed. For example, the moment with an element value of 4 in the time sequence can be picked in the way of the number of items in a batch of 2, which can also meet the requirements of the fixed number of items in a batch; i

[0036] E) Search for the rotary shelves and items corresponding to each element in the collaborative picking order E coop and determine the set S to which the item G belongs.​n and S n ←S n ∪G,

[0037] F) Calculate the transfer time corresponding to the set S n and accumulate to obtain the total operation time T turn , and use it as an evaluation index. If T turn (S best ) > T tur ( n S * ), then minT turn = T turn (S * ), thereby determining the optimal collaborative picking order, S best ←S * , where S * is the comparison sequence, S best is the optimal sequence, update the picking task S. If length(S) ≥ U, where U is the total number of goods required for the picking task, end, otherwise return to step B).

[0038] As Figure 1 shown, in this embodiment, it is set that the number of multi-rotary shelves takes values from 6 to 12 respectively, the storage location specifications are fixed, the transfer speed is 0.36 m / s, and 3 kinds of goods are randomly mixed on the shelves. The construction method is as follows: the number of storage locations of each shelf is randomly taken between 50 and 80 respectively, the types of stored goods are random, there are 8 groups of data, the loading rate of the shelf is 0.8, and the goods numbers are randomly generated in the storage location coordinates. To ensure the existence of a feasible solution, usually make the number of target type goods generated sufficient.

[0039] For a certain typical collaborative picking task, the number of batches of a certain kind of goods is 2, the transfer time interval is 4 s, and the picking goods batches are 6. In eight kinds of shelf configurations, each group of data is tested 100 times to obtain the best collaborative picking order. Combining with the method provided by the present invention, the transfer time under the path corresponding to the optimal solution is obtained. Compared with the other two typical heuristic algorithms (time-first algorithm, distance-first algorithm), on the one hand, the picking operation time of the collaborative combination optimization algorithm is reduced by about 35%, and the task execution efficiency is higher. This is mainly because the combination strategy introduced by the algorithm can reduce the waiting time between the transfers of goods in the same batch, making the planning result better; on the other hand, compared with the design method, the average calculation time overhead is saved by 0.211 s and 0.258 s respectively, which is increased by about 48%. This is because the time serialization conversion performed by the algorithm can simplify the process of selecting goods in the time set, making the timeliness of the collaborative combination optimization algorithm more advantageous.

[0040] The optimization method for the picking operation of the multi-rotary shelf device of the present invention improves the collaborative operation efficiency of the multi-rotary shelf device, enabling it to obtain the picking path with the optimal execution efficiency and timeliness under complex constraint conditions such as picking batches, the quantity of goods in batches, and transfer time intervals; it can be extended and applied to the scheduling model of the rotary storage system under multiple modules, and the proposed collaborative combination optimization algorithm can evaluate the picking capabilities under different shelf configurations of the system, providing a basis for the operation performance analysis of the multi-rotary shelf device.

Claims

1. An optimization method for picking operations of a multi-rotary shelf device, characterized in that: Including the following steps: A) Obtain the optimal picking order of a single carousel rack, starting from the initial coordinates G of the required goods on each carousel rack n , and search to determine the optimal picking order of each rack In the formula, n is the rack number, N is the number of racks of the overall device, m is the number of goods of the target category on rack n, and t m is the picking time of the last good on this rack; B) Calculate the picking time sets of each carousel based on the optimal picking order obtained in step A), and calculate the picking time set I of each carousel n , I n ={t1,..., t a , t b ,..., t m}, where t a , t b are two adjacent moments; C) Serialize and convert according to the picking time set to obtain a picking time series. If is an empty set, calculate the picking time set I n corresponding picking time series E n ; D) Calculate the collaborative picking order based on the transfer time interval and the picking time series, and calculate the collaborative picking order E coop , combine according to the transfer time interval and the picking time series to obtain E) Search for the rotary shelves and goods corresponding to each element in the collaborative picking order, determine the set to which the goods belong, and search for the collaborative picking order E coop for the rotary shelves and goods corresponding to each element in it, and determine the set S to which the goods G belong n and S n ←S n ∪G, F) Calculate the transfer time corresponding to the set to which the goods belong, accumulate to obtain the total operation time, and use it as an evaluation index to determine the optimal collaborative picking order, and calculate the set S n The corresponding transfer time, accumulate to obtain the total operation time T turn , and use it as an evaluation index. If T turn (S best ) > T turn (S * ), then minT turn = T turn (S * ), thereby determining the optimal collaborative picking order, S best ← S * , where S * is the comparison sequence, and S best is the optimal sequence.

2. The optimization method for picking operations of the multi-rotary shelf device according to claim 1, characterized in that: In step F), determine the optimal collaborative picking order, update the picking task S. If length(S)≥U, where U is the total number of goods required for the picking task, end. Otherwise, return to step B).

3. The optimization method for picking operations of the multi-rotary shelf device according to claim 1, wherein: In step C), according to the single optimal picking order calculate the picking operation time without time interval constraint, denoted as the shortest picking time. When planning the picking operation, introduce waiting time for two adjacent goods in sequence, extend the picking time interval between them, and it shall not be less than the minimum time interval. After referring to the transfer time interval t0, given a time interval Δt, according to the corresponding goods picking time, calculate an equidistant sequence E = [e1, e2, e3,..., e m , and the time interval between two adjacent elements in E is Δt. The i-th element e i represents the number of goods that can be picked at the moment of (i - 1)Δt, that is, at the moment of t(e i ), and e i ≥0. And when e i ≥2, it means that multi-good shipping can be carried out at this moment, so as to convert I n into a picking time sequence E n with positive integer element values, where e1 corresponds to the initial picking moment t(e i ), and the time interval between t a and t b satisfies the given time interval Δt. If the transfer time interval Δt is greater than the required transfer time interval t0 and meets L = [(t a -t b ) / t0], it means that L goods can be picked between t a and t b at t0. Therefore, L zeros need to be added to E n to indicate that it can be combined with other rotary shelves to meet collaborative transfer.

4. The optimization method for picking operations of the multi-rotary shelf device according to claim 1, characterized in that: In the said step D), the picking time set I n is converted into a picking time series E n After that, according to the given transfer time interval t0 and the quantity a0 of goods in each batch, E n The situations for collaborative combination include: First, the interval Δt in the time series satisfies Δt ≤ t0; Second, the element value e in the time series satisfies e ≤ a0. Thus, the picking time series of multiple carousel devices are traversed and combined to obtain a set of collaborative picking orders E coop , denoted as The length of this sequence is the number of picking times, and each element value is greater than or equal to the quantity of goods in each batch.

Images