Method and apparatus for optimizing designs of product packaging units

Abstract

Disclosed herein are a method and apparatus for optimizing designs of product packaging units. A method based on one aspect of the present disclosure comprises: for each product having a corresponding order quantity among a plurality of products, using an integer programming model to determine the capacity of each level of packaging unit among one or more levels of packaging units above a basic packaging unit of the product, wherein the integer programming model uses minimizing the logistics-related costs of the product having the corresponding order quantity as its objective function; for each clustering scheme among a plurality of clustering schemes applied to the plurality of products, calculating the total logistics-related cost savings compared with a situation where the clustering scheme is not applied, wherein each clustering scheme corresponds to a different number of clusters and is used for dividing the plurality of products into the corresponding number of clusters on the basis of the sizes of the basic packaging units of the plurality of products; and on the basis of the clustering scheme having the maximum total logistics-related cost savings among the plurality of clustering schemes, determining an optimal packaging unit design.

Description

Method and apparatus for optimizing product packaging unit design Technical Field

[0001] The present disclosure relates to methods and apparatus for optimizing product packaging unit design. Background Art

[0002] In fields such as warehousing logistics and supply chain management, in order to facilitate the handling and transportation of products (depending on the application scenario, they can also be called items, goods, etc.), it is usually necessary to design multiple levels of packaging units. Typically, starting from the basic packaging unit (sometimes also called retail or sales packaging unit) that is in direct contact with the product itself and is designed to protect the product itself from damage, destruction, or leakage, several additional levels of packaging units are designed upwards, such as second-level packaging units, third-level packaging units, etc. The capacity of these different levels of packaging units increases successively to accommodate several, dozens, or even hundreds of packaging units of the previous level. Depending on the specific application scenario, the packaging units at each level can cover a variety of different types such as paper boxes, cardboard boxes, wooden boxes, pallets, etc.

[0003] From a variety of perspectives, including production, storage, and transportation, the design of each packaging unit significantly impacts packaging costs, particularly those related to logistics. Currently, the capacity of each packaging unit for each product is typically determined manually, which consumes considerable time and labor. Even so, manual design methods fail to fully account for the various constraints and other factors associated with packaging units, and therefore cannot guarantee an optimal design.

[0004] In view of these and other problems, there exists a pressing need for innovative mechanisms for optimizing product packaging unit design.

[0005] Summary of the Invention

[0006] The Summary section introduces selected concepts in a simplified form that will be further elaborated in the Detailed Description section below. This Summary section is not intended to identify any key features or essential features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter.

[0007] According to one aspect of the present disclosure, a method for optimizing product packaging unit design is provided, the method comprising: for each product with a corresponding order quantity among a plurality of products, determining the capacity of each level of packaging units in one or more levels of packaging units above the basic packaging unit of the product using an integer programming model, wherein the constraints of the integer programming model include that the size of each level of packaging units does not exceed the corresponding size threshold, and the capacity of any level of packaging units is an integer multiple of the previous level of packaging units, and the integer programming model takes minimizing the logistics-related cost of the product with the corresponding order quantity as its objective function; for each of a plurality of clustering schemes applied to the plurality of products, calculating the logistics-related cost compared with a case where the clustering scheme is not applied. The present invention relates to a method for calculating the total amount of logistics-related cost savings, wherein each clustering scheme corresponds to a different number of clusters and is used to divide the multiple products into a corresponding number of clusters based on the size of the basic packaging units of the multiple products, and wherein the calculation includes determining the capacity of each level of common packaging units of the cluster in one or more levels by using the integer programming model in a manner that all products in each cluster under the clustering scheme are regarded as one product; and determining the optimal packaging unit design based on the clustering scheme with the largest total amount of logistics-related cost savings among the multiple clustering schemes, in which the same determined capacity of each level of common packaging units is used for all products of the multiple products divided into each cluster under the clustering scheme.

[0008] According to another aspect of the present disclosure, a computing device is provided, comprising: at least one processor; and a memory coupled to the at least one processor and configured to store instructions, wherein when the instructions are executed by the at least one processor, the at least one processor is configured to perform the method described in the present disclosure.

[0009] According to yet another aspect of the present disclosure, a computer-readable storage medium is provided, on which instructions are stored. When the instructions are executed by at least one processor, the at least one processor is caused to perform the method described in the present disclosure.

[0010] According to yet another aspect of the present disclosure, a computer program product is provided, comprising instructions, which, when executed by at least one processor, cause the at least one processor to perform the method described in the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Implementations of the present disclosure are illustrated by way of example and not limitation in the accompanying drawings in which like reference numerals designate identical or similar parts and in which:

[0012] FIG1 illustrates a flow chart of an exemplary method according to some implementations of the present disclosure;

[0013] FIG2 illustrates a flow chart of exemplary operations according to some implementations of the present disclosure;

[0014] FIG3 illustrates a flow chart of exemplary operations according to some implementations of the present disclosure;

[0015] FIG4 illustrates a block diagram of an exemplary apparatus according to some implementations of the present disclosure; and

[0016] 5 illustrates a block diagram of an example computing device according to some implementations of the present disclosure. DETAILED DESCRIPTION

[0017] In the following description, for purposes of explanation, numerous specific details are set forth to provide a more thorough understanding of the present disclosure. However, these specific details are merely exemplary and non-restrictive, and it will be apparent to those skilled in the art that the present disclosure can be implemented without these specific details. In other instances, well-known circuits, structures, and techniques are not shown in detail to avoid unnecessarily obscuring the understanding of the description.

[0018] References throughout this specification to "an implementation," "implementations," "example implementations," "some implementations," "various implementations," etc., indicate that the implementations of the present disclosure described may include particular features, structures, or characteristics. However, it does not imply that every implementation must include these particular features, structures, or characteristics. Furthermore, some implementations may have some, all, or none of the features described for other implementations.

[0019] In a manner that is most helpful in understanding the claimed subject matter, various operations may be described as multiple discrete actions or operations in a sequential order. However, the order of description should not be interpreted as implying that these operations are necessarily order-dependent. On the contrary, these operations may not be performed in the order presented. In other implementations, various additional operations may be performed, and / or various operations already described may be omitted.

[0020] For a product, it is usually necessary to design multiple levels of packaging units. Above the basic packaging unit (sometimes also called retail or sales packaging unit), there will be several additional levels of packaging units to provide protection for batches of products and facilitate warehousing, transportation, etc.

[0021] The traditional method of manually designing packaging units often consumes considerable time and manpower costs. In addition, considering the complexity, it is difficult to manually determine the optimal capacity of each level of packaging units for, for example, multiple products belonging to a product family, while comprehensively considering various constraints related to each level of packaging units (for example, the size and weight of the packaging units) and even the actual demand for each product.

[0022] Therefore, the present disclosure aims to provide an efficient and feasible automation mechanism for optimizing product packaging unit design.

[0023] Reference is now made to FIG1 , which illustrates a flow chart of an exemplary method 100 according to some implementations of the present disclosure. Exemplary method 100 is used to implement the mechanisms described herein for optimizing product packaging unit design. In some implementations, method 100 can be implemented on a single computer, workstation, server, or server cluster. In some implementations, method 100 can also be implemented using a distributed computing architecture, such as on a cloud computing platform.

[0024] As shown in FIG. 1 , the exemplary method 100 begins at step 110 , in which, for each product among a plurality of products having a corresponding order quantity, an integer programming model is used to determine the capacity of each of one or more levels of packaging units above a basic packaging unit of the product.

[0025] In some implementations of the present disclosure, the multiple products may belong to the same product family. For example, the multiple products may be products of different specifications or models under a category of products of a certain manufacturer. Although the specifications of the corresponding basic packaging units (which are generally designed to accommodate and protect a single product) may vary due to the differences between the products, it is expected that for the products within the product family, a single-level or multi-level packaging unit design that is as common as possible is adopted above the basic packaging unit. In some implementations of the present disclosure, the single-level or multi-level packaging unit includes an additional tertiary packaging unit above the basic packaging unit. However, more or fewer levels of packaging units are also feasible, and the present disclosure is not limited thereto.

[0026] For example, each product has three additional packaging units above the basic packaging unit. These additional packaging units are referred to as second-level packaging units, third-level packaging units, and fourth-level packaging units. For example, a second-level packaging unit designed for a product can accommodate multiple (e.g., 5) basic packaging units, its third-level packaging unit can accommodate multiple (e.g., 10) second-level packaging units, and its fourth-level packaging unit can accommodate multiple (e.g., 8) third-level packaging units. The capacity of each level of packaging unit is an integer multiple of the previous level of packaging unit. For ease of description and calculation, the capacity of these level packaging units can be counted in units of basic packaging units. That is, in the above example, the number of basic packaging units accommodated by the second-level, third-level, and fourth-level packaging units is 5, 50, and 400, respectively, with each number being an integer multiple of the previous one.

[0027] Continuing with the example, let's assume that a product family P contains ten different products P 01 -P 10 , each product has its corresponding order quantity Q 01 -Q 10 ; and as shown above, for each product (for example, the fifth product P 05 , the corresponding order quantity is Q 05 ) In addition to the basic packaging unit, the second-level packaging unit, the third-level packaging unit, and the fourth-level packaging unit must be designed.

[0028] In some implementations of the present disclosure, for each product, the capacity of each level of packaging unit above the basic packaging unit is determined using an integer programming model. Integer programming models involve optimization problems. Optimization involves selecting the optimal solution from a number of feasible options under certain constraints, so that the system's objective function is maximized or minimized within the constraints.

[0029] In some implementations of the present disclosure, as three basic elements of the integer programming model, the decision variables of the model, that is, the quantity to be determined, are the capacity of each level of packaging units in one or more levels (three levels in the above example) of packaging units above the basic packaging unit of a product; the objective function of the model is to minimize the logistics-related costs of the product for the corresponding order quantity; and the constraints of the model include that the size of each level of packaging units does not exceed the corresponding size threshold, and the capacity of any level of packaging units is an integer multiple of the previous level of packaging units.

[0030] Therefore, the integer programming model can be simplified as follows:

[0031] Minimize the logistics-related costs of the product for the corresponding order quantity

[0032] obey:

[0033] The size of the second level packaging unit is ≤ the specified size threshold

[0034] The size of the third-level packaging unit is ≤ the specified size threshold

[0035] The size of the fourth-level packaging unit is ≤ the specified size threshold

[0036] The capacity of any level of packaging unit should be an integer multiple of the previous level

[0037] Decision variables:

[0038] Capacity of the secondary packaging unit

[0039] Capacity of the third-level packaging unit

[0040] Capacity of the fourth level packaging unit

[0041] It is understood that the thresholds specified for packaging units at different levels may not be the same. In some implementations of the present disclosure, the constraint conditions of the integer programming model that the size of each level of packaging units does not exceed the corresponding size thresholds may include: the length, width, and height of each level of packaging units do not exceed the corresponding length threshold, width threshold, and height threshold. In this case, taking the size of the second-level packaging unit as an example, the relevant constraint conditions in the above expression are

[0042] The size of the second level packaging unit ≤ the specified size threshold can be further expressed as:

[0043] The length of the second-level packaging unit is ≤ the specified length threshold

[0044] The width of the second-level packaging unit is ≤ the specified width threshold

[0045] The height of the second level packaging unit is ≤ the specified height threshold

[0046] The same is true for the third-level packaging unit and the fourth-level packaging unit, which will not be repeated here.

[0047] Furthermore, in some implementations of the present disclosure, the constraints of the integer programming model may further include that the weight of each level of packaging unit does not exceed a corresponding weight threshold. The weight here may refer to the gross weight of the packaging unit. In this case, the following constraint item may be further added to the above expression:

[0048] The weight of the second-level packaging unit is ≤ the specified weight threshold

[0049] The weight of the third-level packaging unit is ≤ the specified weight threshold

[0050] The weight of the fourth-level packaging unit is ≤ the specified weight threshold

[0051] Furthermore, in some implementations of the present disclosure, the logistics-related costs may include the pick and pack cost, warehouse handling cost, and transportation cost of the product for the corresponding order quantity. In this case, the objective function of the integer programming model can be further expressed as: min.∑Cost pick&pack +Cost handling +Cost transport

[0052] Here, the picking and packing cost (Cost pick&pack ) is associated with the order quantity of the product. As an example, an order quantity can be considered to be completed by x4 fourth-level packaging units, x3 third-level packaging units, x2 second-level packaging units, and x1 basic packaging units, and the unit price of picking and packing is j. Then the picking and packing cost can be expressed as: Cost pick&pack =j*(x4+x3+x3+x1)

[0053] It is understandable that other calculation methods of picking and packing costs are also feasible, and the present disclosure is not limited thereto.

[0054] Warehouse management cost handling ) is also associated with the order quantity of the product. As an example, the unit price of warehouse management is k, and if the order quantity is calculated in terms of the basic packaging unit that contains a single product, the order quantity can be regarded as including y1 basic packaging units; if it is calculated in terms of the second-level packaging unit, the same order quantity can be regarded as including y2 second-level packaging units (only the integer part is retained, the same below); if it is calculated in terms of the third-level packaging unit, the same order quantity can be regarded as including y3 third-level packaging units; and if it is calculated in terms of the fourth-level packaging unit, the same order quantity can be regarded as including y4 fourth-level packaging units, then the warehouse management cost can be expressed as: Cost handling =k*(y1+y2+y3+y4)

[0055] It is also understandable that other calculation methods of warehouse management costs are also feasible, and the present disclosure is not limited thereto.

[0056] In addition, transportation costs transport) can be associated not only with the order quantity of the product, but also with the size and weight of the packaging units at all levels. In addition, in some implementations according to the present disclosure, other logistics-related cost items can also be considered as a supplement or alternative to the aforementioned.

[0057] According to some implementations of the present disclosure, the integer programming model used here is a mixed integer programming model. Through step 110, the capacity of the primary or multi-level packaging unit optimized for each of the multiple products (ie, optimized by product) can be determined.

[0058] Next, method 100 proceeds to step 120, where, for each of the plurality of clustering schemes applied to the plurality of products, a total logistics-related cost savings is calculated compared to a scenario where the clustering scheme is not applied. According to some implementations of the present disclosure, each clustering scheme corresponds to a different number of clusters and is used to divide the plurality of products into a corresponding number of clusters based on the size of the basic packaging units of the plurality of products.

[0059] In step 120, as will be described in more detail below, by clustering the multiple products according to the size of their basic packaging units and thereby dividing them into clusters, the capacity of cluster-optimized primary or multi-level packaging units (in this document, such packaging units are referred to as common packaging units) can be determined for those products divided into each cluster.

[0060] For clustering algorithms, the number of clusters (also known as the cluster number) is a hyperparameter. According to some implementations of the present disclosure, multiple different numbers of clusters (i.e., multiple different clustering schemes) can be set and the clustering results can be compared in the manner disclosed herein to ultimately determine the optimal design.

[0061] In some implementations of the present disclosure, method 100 may include an optional step of determining possible selections for the number of clusters (not shown in FIG1 ). In this step, a maximum number of clusters is determined by deduplicating the plurality of products by treating products having the same basic packaging unit size as one product, wherein the number of clusters in each of the plurality of clustering schemes does not exceed the maximum number of clusters.

[0062] Combined with the above description, a product family P contains ten different products P 01 -P 10 As an example, suppose that among the ten products, only four products P 03 、P 06 、P 09 、P 10The size of the basic packaging unit is the same, then these four products will be regarded as the same product. Accordingly, in addition to this one product, there are six products P in the plurality of products. 01 、P 02 、P 04 、P 05 、P 07 、P 08 , and therefore, the number of products determined after deduplication is seven. This number is taken as the maximum number of clusters here.

[0063] In other words, in this case, there can be at most seven clustering schemes for these ten products. The first clustering scheme divides these products into one cluster, the second clustering scheme divides them into two clusters, and so on, up to the seventh clustering scheme, which divides these products into seven clusters. Here, the number of clusters in each clustering scheme is represented by n, and the value of n ranges from n∈[1,N]. In this example, N=7.

[0064] Return to step 120. In some implementations of the present disclosure, step 120 may include, for each of the plurality of clustering schemes, i.e., for each n (where n∈[1,N]): using a Gaussian mixture model to divide the plurality of products into a corresponding number (i.e., n) of clusters in the clustering scheme based on the length, width, and height of the basic packaging unit of each product in the plurality of products. For example, for the case where n=4, this operation will divide the ten products P in a product family P in the above example into n clusters. 01 -P 10 They are divided into 4 clusters according to the length, width and height of their respective basic packaging units.

[0065] In some implementations of the present disclosure, the clustering algorithm uses a Gaussian mixture model. The Gaussian mixture model consists of several Gaussian distributions (normal distributions), each of which corresponds to a cluster. The Gaussian mixture model will calculate the probability that the data points to be clustered obey these distributions and divide them into corresponding clusters based on this. Since soft clustering is achieved using a probability-based approach, the Gaussian mixture model provides a more flexible mechanism. In addition, in some implementations of the present disclosure, the K-means clustering algorithm can also be used as an alternative.

[0066] In some implementations of the present disclosure, step 120 may include, for each of the multiple clustering schemes: using the integer programming model to determine the capacity of each level of common packaging units in the cluster's one or more levels of common packaging units in a manner that treats all products in each cluster under the cluster as one product.

[0067] Continuing with the previous example, for the case of n=4, the ten products P are clustered by the clustering algorithm. 01 -P 10 After being divided into four clusters, the products in each cluster can be considered as a single product, meaning that each cluster represents a single product. The previously described integer programming model is then used to determine the capacity of each level of packaging within that cluster, including the primary or multi-level packaging units. As mentioned previously, the packaging units determined per cluster are called shared packaging units; that is, they are shared by all products within that cluster.

[0068] In some implementations of the present disclosure, step 120 may include, for each clustering scheme in the plurality of clustering schemes, that is, for each n (where n∈[1,N]), performing the following operations:

[0069] For each cluster in this clustering scheme, first, the length, width, height, and weight of the basic packaging unit of each product in the cluster are replaced by the maximum length, maximum width, maximum height, and maximum weight of the basic packaging units of all products in the cluster to determine the maximum basic packaging unit of the cluster. It is understood that this maximum basic packaging unit may not be an actual basic packaging unit (for example, designed for a specific product in the cluster), but is instead a summary based on the specific data of the basic packaging units of all products in the cluster for the purpose of performing the following calculations.

[0070] Secondly, by treating all products in the cluster as one product and based on the determined size and weight of the largest basic packaging unit of the cluster, the integer programming model is used to determine the capacity of each level of the one or more levels of common packaging units above the largest basic packaging unit of the cluster.

[0071] The above operation is repeated for each cluster under a clustering scheme, i.e., the first cluster, the second cluster, ... until the nth cluster, so that the capacity of the first-level or multi-level common packaging unit optimized by cluster under the clustering scheme can be determined for multiple products in a product family.

[0072] When the capacity of the first-level or multi-level packaging units optimized for each product individually and the capacity of the first-level or multi-level common packaging units optimized by cluster according to the clustering scheme are obtained respectively in the aforementioned manner, the impact of applying or not the clustering scheme on the logistics-related costs can be calculated, thereby making it possible to make subsequent comparisons between the various clustering schemes.

[0073] 2, which illustrates a flow chart of exemplary operations 200 according to some implementations of the present disclosure. Exemplary operations 200 may be part of step 120 of method 100. More specifically, for each of the plurality of clustering schemes, i.e., for each n (where n∈[1,N]), exemplary operations 200 may include:

[0074] For each cluster under this clustering scheme, in step 210, the logistics-related costs for all products in the cluster are calculated based on the determined capacity of the cluster's primary or multi-level shared packaging units, as first logistics-related costs. The calculation of these logistics-related costs has been described and exemplified in the previous discussion related to step 110 of method 100 and will not be repeated here. The calculation result in step 210 indicates the logistics-related costs required for the products in this cluster if this clustering scheme is applied and the packaging unit design is optimized by cluster.

[0075] Next, in step 220, the logistics-related costs for all products in the cluster are calculated based on the determined capacity of the primary or multi-level packaging units for each product in the cluster, as second logistics-related costs. The calculation result of step 220 indicates the logistics-related costs that would be required for these products in the cluster if the clustering solution and, therefore, cluster-based optimization of packaging unit design were not applied.

[0076] Then, in step 230, the logistics-related cost saving amount of the cluster is obtained by calculating the difference between the second logistics-related cost and the first logistics-related cost.

[0077] After completing the above processing for a cluster under the clustering scheme, example operation 200 proceeds to step 240, where it is determined whether all clusters under the clustering scheme have been processed. If the determination result of step 240 is "no," example operation 200 jumps back to step 210 and begins processing the next cluster. Conversely, if the determination result of step 240 is "yes," indicating that all clusters under the clustering scheme have been processed, example operation 200 proceeds to step 250.

[0078] In step 250 , the total logistics-related cost savings of the clustering scheme is obtained by summing up the logistics-related cost savings of each cluster under the clustering scheme.

[0079] 3 , which illustrates a flow chart of exemplary operations 300 according to some implementations of the present disclosure. Exemplary operations 300 may serve as an alternative to the aforementioned exemplary operations 200. More specifically, for each of the plurality of clustering schemes, i.e., for each n (where n∈[1,N]), exemplary operations 300 may include:

[0080] For each cluster under the clustering scheme, the logistics costs of all products in the cluster are calculated using the determined capacity of the first or multi-level common packaging units of the cluster in step 310. Step 310 is the same as the process performed in step 210 described above.

[0081] Next, in step 320, a determination is made as to whether all clusters under the clustering scheme have been processed. If the determination result of step 320 is "no," exemplary operation 300 jumps back to step 310 to begin processing the next cluster. Conversely, if the determination result of step 320 is "yes," indicating that all clusters under the clustering scheme have been processed, exemplary operation 300 proceeds to step 330.

[0082] In step 330, the third logistics-related cost for the multiple products under the clustering scheme is obtained by summing the logistics-related costs for each cluster under the clustering scheme. Unlike the processing logic of exemplary operation 200, exemplary operation 300 first calculates the logistics-related costs required for all of the multiple products when the clustering scheme is applied and the packaging unit design is optimized by cluster.

[0083] Then, in step 340, the determined capacity of the primary or multi-level packaging units for each of the plurality of products is used to calculate the logistics-related costs for the plurality of products as the fourth logistics-related costs for the plurality of products. The calculation result of step 240 indicates the logistics-related costs that would be required for all of the plurality of products if the clustering solution and therefore cluster-based optimization of the packaging unit design were not applied.

[0084] Finally, in step 350 , the total logistics-related cost savings of the clustering solution is obtained by calculating the difference between the fourth logistics-related cost and the third logistics-related cost.

[0085] Returning to the exemplary method 100 of FIG. 1 , after step 120 is completed, the total logistics-related cost savings for each of the multiple clustering schemes are obtained. This allows the pros and cons of each clustering scheme to be determined. Therefore, method 100 proceeds to step 130 , in which an optimal packaging unit design is determined based on the clustering scheme with the greatest total logistics-related cost savings among the multiple clustering schemes. In this optimal packaging unit design, the same capacity of the shared packaging units at each level is used for all products in each cluster assigned to that clustering scheme.

[0086] According to some implementations of the present disclosure, for multiple products belonging to the same product family, the clustering scheme that maximizes the total cost savings represents the optimal grouping. Therefore, from a logistics-related cost perspective, the optimal packaging unit design is achieved by grouping these products into a corresponding number of clusters in this clustering scheme and applying the capacity of the primary or multi-level shared packaging unit for each cluster, as calculated in step 120, to all products in each cluster.

[0087] 4, which illustrates a block diagram of an exemplary apparatus 400 according to some implementations of the present disclosure. The exemplary apparatus 400 is used to implement the mechanism described herein for optimizing product packaging unit design. Those skilled in the art will appreciate that the apparatus 400 can be implemented using software, hardware, firmware, or any combination thereof.

[0088] As shown in FIG4 , apparatus 400 may include a module 410 for determining, for each product with a corresponding order quantity among a plurality of products, the capacity of each level of packaging units in one or more levels of packaging units above the basic packaging unit of the product using an integer programming model, wherein constraints of the integer programming model include that the size of each level of packaging units does not exceed a corresponding size threshold and that the capacity of any level of packaging units is an integer multiple of the previous level of packaging units, and wherein the integer programming model minimizes the logistics-related cost of the product with the corresponding order quantity as its objective function. Apparatus 400 may also include a module 420 for calculating, for each of a plurality of clustering schemes applied to the plurality of products, the total logistics-related cost savings compared to a scenario where the clustering scheme is not applied, wherein each clustering scheme corresponds to a different number of clusters and is used to divide the plurality of products into a corresponding number of clusters based on the size of the basic packaging unit of the plurality of products, and wherein the calculation includes determining, using the integer programming model, the capacity of each level of common packaging units in the one or more levels of common packaging units of each cluster, treating all products in each cluster under the clustering scheme as a single product. In addition, the device 400 may also include a module 430, which is used to determine the optimal packaging unit design based on the clustering scheme with the largest total logistics-related cost savings among the multiple clustering schemes, in which the same determined capacity of the shared packaging unit at each level is used for all products in each cluster among the multiple products that are divided into the clustering scheme.

[0089] In some implementations of the present disclosure, the apparatus 400 may further include additional modules for performing other operations described in the specification, such as those described in conjunction with the flowchart of the exemplary method 100 in FIG1 , the flowchart of the exemplary operation 200 in FIG2 , the flowchart of the exemplary operation 300 in FIG3 , and their respective variations. Furthermore, in some implementations, the various modules of the apparatus 400 may be combined or separated depending on actual needs, without departing from the scope of the present disclosure.

[0090] 5 shows a block diagram of an exemplary computing device 500 according to some implementations of the present disclosure. The exemplary computing device 500 is used to implement the mechanisms described herein for optimizing product packaging unit designs.

[0091] As shown in Figure 5, the computing device 500 may include at least one processor 510. The processor 510 may include any type of general-purpose processing unit (such as a CPU, GPU), a dedicated processing unit, a core, a circuit, a controller, and the like. In addition, the computing device 500 may also include a memory 520. The memory 520 may include any type of medium that can be used to store data. In some implementations, the memory 520 is configured to store instructions that, when executed by the at least one processor 510, cause the processor 510 to perform the operations described herein, for example, the various operations described in conjunction with the flowchart of the exemplary method 100 of Figure 1, the flowchart of the exemplary operation 200 of Figure 2, the flowchart of the exemplary operation 300 of Figure 3, and their respective variations.

[0092] In addition, in some implementations, the computing device 500 is also equipped with a communication interface that can support various types of wired / wireless communication protocols to communicate with a communication network (e.g., a local area network, a metropolitan area network, a wide area network, the Internet).

[0093] Those skilled in the art will appreciate that the above description of the structure of the computing device 500 is merely exemplary and non-limiting, and devices with other structures are also feasible as long as they can be used to implement the functions described herein.

[0094] Various implementations of the present disclosure may include or operate multiple components, parts, units, modules, instances, or mechanisms, which may be implemented in hardware, software, firmware, or any combination thereof. Examples of hardware may include, but are not limited to, devices, processors, microprocessors, circuits, circuit elements (e.g., transistors, resistors, capacitors, inductors, etc.), integrated circuits, application specific integrated circuits (ASICs), programmable logic devices (PLDs), digital signal processors (DSPs), field programmable gate arrays (FPGAs), memory cells, logic gates, registers, semiconductor devices, chips, microchips, chipsets, and the like. Examples of software may include, but are not limited to, software components, programs, applications, computer programs, application programs, system programs, machine programs, operating system software, middleware, software modules, routines, subroutines, functions, methods, procedures, software interfaces, application programming interfaces (APIs), instruction sets, computer code, computer code segments, words, values, symbols, or any combination thereof. Determining whether an implementation is performed using hardware, software, and / or firmware can vary depending on a variety of factors, such as desired computational rate, power level, thermal tolerance, processing cycle budget, input data rate, output data rate, memory resources, data bus speed, and other design or performance constraints as desired for a given implementation.

[0095] Some implementations described herein may include an article of manufacture. The article of manufacture may include a storage medium. Examples of storage media may include volatile and non-volatile, removable and non-removable media implemented using any method or technology for storing information (e.g., computer-readable instructions, data structures, program modules, or other data). Storage media may include, but are not limited to, random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other semiconductor memory, compact disc (CD), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium capable of storing information. In some implementations, the article of manufacture may store executable computer program instructions that, when executed by one or more processing units, cause the processing units to perform the operations described herein. Executable computer program instructions may include any suitable type of code, such as source code, compiled code, interpreted code, executable code, static code, dynamic code, and the like. Executable computer program instructions may be implemented using any suitable high-level, low-level, object-oriented, visual, compiled and / or interpreted programming language.

[0096] Some exemplary implementations of the present disclosure are described below.

[0097] Example 1 may include a method for optimizing product packaging unit design, the method comprising: for each product with a corresponding order quantity among a plurality of products, using an integer programming model to determine the capacity of each level of packaging units in one or more levels of packaging units above a basic packaging unit of the product, wherein the constraints of the integer programming model include that the size of each level of packaging units does not exceed a corresponding size threshold, and the capacity of any level of packaging units is an integer multiple of the previous level of packaging units, and the integer programming model takes minimizing the logistics-related cost of the product with the corresponding order quantity as its objective function; for each of a plurality of clustering schemes applied to the plurality of products, calculating the total logistics-related cost savings compared to a case where the clustering scheme is not applied amount, wherein each clustering scheme corresponds to a different number of clusters and is used to divide the plurality of products into a corresponding number of clusters based on the size of the basic packaging units of the plurality of products, and wherein the calculation includes determining the capacity of each level of common packaging units of the cluster in one or more levels by using the integer programming model in a manner that all products in each cluster under the clustering scheme are regarded as one product; and determining the optimal packaging unit design based on the clustering scheme with the largest total amount of logistics-related cost savings among the plurality of clustering schemes, in which the same determined capacity of each level of common packaging units is used for all products of the plurality of products divided into each cluster under the clustering scheme.

[0098] Example 2 may include the subject matter described in the aforementioned Example 1, wherein the number of deduplicated products of the plurality of products is determined as the maximum number of clusters by deduplicating the plurality of products in such a manner that products having basic packaging units of the same size among the plurality of products are treated as one product, wherein the number of clusters of each of the plurality of clustering schemes does not exceed the maximum number of clusters.

[0099] Example 3 may include the subject matter of any of the preceding Examples 1-2, wherein the plurality of products belong to a product family, and wherein the primary or multi-level packaging units include additional tertiary packaging units above the basic packaging unit.

[0100] Example 4 may include the subject matter of any one of the aforementioned Examples 1-3, wherein the size of each level packaging unit does not exceed the corresponding size threshold, including the length, width, and height of each level packaging unit not exceeding the corresponding length threshold, width threshold, and height threshold.

[0101] Example 5 may include the subject matter of any one of the preceding Examples 1-4, wherein the constraints of the integer programming model further include that the weight of each level of packaging unit does not exceed a corresponding weight threshold.

[0102] Example 6 may include the subject matter of any of the preceding Examples 1-5, wherein the logistics-related costs include picking and packing costs, warehouse management costs, and transportation costs for the product for the corresponding order quantity.

[0103] Example 7 may include the subject matter of any one of the aforementioned Examples 1-6, wherein, for each clustering scheme, calculating the total logistics-related cost savings compared to a situation where the clustering scheme is not applied includes: based on the length, width, and height of the basic packaging unit of each of the multiple products, using a Gaussian mixture model to divide the multiple products into a corresponding number of clusters of the clustering scheme.

[0104] Example 8 may include the subject matter of any one of the aforementioned Examples 1-7, wherein, for each clustering scheme, calculating the total logistics-related cost savings compared to a situation where the clustering scheme is not applied also includes, for each cluster under the clustering scheme: replacing the length, width, height, and weight of the basic packaging unit of each product in the cluster with the maximum length, maximum width, maximum height, and maximum weight of the basic packaging units of all products in the cluster, respectively, to determine the maximum basic packaging unit of the cluster; and determining the capacity of each level of the one or more levels of common packaging units above the maximum basic packaging unit of the cluster by treating all products in the cluster as one product and based on the determined size and weight of the maximum basic packaging unit of the cluster using the integer programming model.

[0105] Example 9 may include the subject matter of any one of the aforementioned Examples 1-8, wherein, for each clustering scheme, calculating the total logistics-related cost savings compared to a situation where the clustering scheme is not applied further comprises: for each cluster under the clustering scheme: calculating the logistics-related costs of all products in the cluster based on the determined capacity of the first-level or multi-level common packaging units of the cluster as a first logistics-related cost; calculating the logistics-related costs of all products in the cluster based on the determined capacity of the first-level or multi-level packaging units of each product in the cluster as a second logistics-related cost; and obtaining the logistics-related cost savings of the cluster by calculating the difference between the second logistics-related cost and the first logistics-related cost; and obtaining the total logistics-related cost savings of the clustering scheme by summing the logistics-related cost savings of each cluster under the clustering scheme.

[0106] Example 10 may include the subject matter of any one of the aforementioned Examples 1-8, wherein, for each clustering scheme, calculating the total logistics-related cost savings compared to a situation where the clustering scheme is not applied further includes: for each cluster under the clustering scheme, calculating the logistics-related costs of all products in the cluster based on the determined capacity of the first-level or multi-level common packaging units of the cluster; obtaining a third logistics-related cost of the multiple products under the clustering scheme by summing the logistics-related costs of the clusters under the clustering scheme; calculating the logistics-related costs of the multiple products based on the determined capacity of the first-level or multi-level packaging units of each of the multiple products as a fourth logistics-related cost of the multiple products; and obtaining the total logistics-related cost savings of the clustering scheme by calculating the difference between the fourth logistics-related cost and the third logistics-related cost.

[0107] Example 11 may include the subject matter of any of the preceding Examples 1-10, wherein the integer programming model is a mixed integer programming model.

[0108] Example 12 may include the subject matter of any one of the aforementioned Examples 1-6, wherein, for each clustering scheme, calculating the total logistics-related cost savings compared to a situation where the clustering scheme is not applied includes: based on the length, width, and height of the basic packaging unit of each of the multiple products, using a K-means clustering algorithm to divide the multiple products into a corresponding number of clusters of the clustering scheme.

[0109] Example 13 may include an apparatus for optimizing product packaging unit design, the apparatus comprising: a module for determining, for each product with a corresponding order quantity among a plurality of products, the capacity of each level of packaging units in one or more levels of packaging units above the basic packaging unit of the product using an integer programming model, wherein the constraints of the integer programming model include that the size of each level of packaging units does not exceed the corresponding size threshold, and the capacity of any level of packaging units is an integer multiple of the previous level of packaging units, and the integer programming model takes minimizing the logistics-related cost of the product with the corresponding order quantity as its objective function; and for calculating, for each of a plurality of clustering schemes applied to the plurality of products, a total logistics-related cost saving compared to a situation where the clustering scheme is not applied. A module for calculating the amount of packaging units, wherein each clustering scheme corresponds to a different number of clusters and is used to divide the multiple products into a corresponding number of clusters based on the size of the basic packaging units of the multiple products, and wherein the calculation includes using the integer programming model to determine the capacity of each level of common packaging units of the cluster in a manner that all products in each cluster under the clustering scheme are regarded as one product; and a module for determining an optimal packaging unit design based on the clustering scheme with the largest total amount of logistics-related cost savings among the multiple clustering schemes, in which the same determined capacity of each level of common packaging units is used for all products of the multiple products divided into each cluster under the clustering scheme.

[0110] Example 14 may include a computing device comprising: at least one processor; and a memory coupled to the at least one processor and used to store instructions, wherein the instructions, when executed by the at least one processor, cause the at least one processor to perform a method according to any one of the foregoing Examples 1-12.

[0111] Example 15 may include a computer-readable storage medium having instructions stored thereon, which, when executed by at least one processor, cause the at least one processor to perform the method according to any one of the foregoing examples 1-12.

[0112] Example 16 may include a computer program product comprising instructions that, when executed by at least one processor, cause the at least one processor to perform the method according to any of the preceding Examples 1-12.

[0113] What has been described above includes examples of the disclosed architecture. It is, of course, not possible to describe every conceivable combination of components and / or methodologies, but those skilled in the art will appreciate that many other combinations and permutations are possible. Therefore, the novel architecture is intended to embrace all such alternatives, modifications, and variations that fall within the spirit and scope of the appended claims.

Claims

1. A method for optimizing product packaging unit design, the method comprising: For each product with a corresponding order quantity among the plurality of products, determine the capacity of each level of packaging units in one or more levels of packaging units above the basic packaging unit of the product using an integer programming model, wherein constraints of the integer programming model include that the size of each level of packaging units does not exceed a corresponding size threshold and the capacity of any level of packaging units is an integer multiple of the previous level of packaging units, and the integer programming model takes minimizing the logistics-related cost of the product with the corresponding order quantity as its objective function; calculating, for each of a plurality of clustering schemes applied to the plurality of products, a total amount of logistics-related cost savings compared to a situation in which the clustering scheme is not applied, wherein each clustering scheme corresponds to a different number of clusters and is used to divide the plurality of products into a corresponding number of clusters based on sizes of basic packaging units of the plurality of products, and wherein the calculating includes determining, using the integer programming model, a capacity of each of one or more levels of common packaging units of the cluster in such a manner that all products in the cluster under the clustering scheme are treated as one product; and An optimal packaging unit design is determined based on a clustering scheme with the largest total logistics-related cost savings among the multiple clustering schemes, in which the same determined capacity of the shared packaging unit at each level is used for all products in each cluster among the multiple products that are divided into the clustering scheme.

2. The method according to claim 1, further comprising: The number of products after deduplication of the multiple products is determined as the maximum number of clusters by deduplicating the multiple products in a manner that products with basic packaging units of the same size among the multiple products are treated as one product, wherein the number of clusters of each clustering scheme in the multiple clustering schemes does not exceed the maximum number of clusters.

3. The method according to claim 1, wherein The plurality of products belong to a product family, and wherein the primary or multi-level packaging unit comprises an additional tertiary packaging unit above the basic packaging unit.

4. The method according to claim 1, wherein The size of each level packaging unit does not exceed the corresponding size threshold, including that the length, width, and height of each level packaging unit do not exceed the corresponding length threshold, width threshold, and height threshold.

5. The method according to claim 4, wherein The constraint condition of the integer programming model also includes that the weight of each level of packaging unit does not exceed the corresponding weight threshold.

6. The method according to claim 5, wherein: The logistics-related costs include picking and packing costs, warehouse management costs, and transportation costs for the product for the corresponding order quantity.

7. The method according to claim 5, wherein: For each clustering solution, the total logistics-related cost savings compared to the situation where the clustering solution is not applied are calculated, including: Based on the length, width, and height of a basic packaging unit of each of the plurality of products, a Gaussian mixture model is utilized to divide the plurality of products into a corresponding number of clusters of the clustering scheme.

8. The method according to claim 7, wherein: For each clustering solution, the total logistics-related cost savings compared to the situation without the clustering solution are calculated, including: For each cluster under this clustering scheme: Replacing the length, width, height, and weight of the basic packaging unit of each product in the cluster with the maximum length, maximum width, maximum height, and maximum weight of the basic packaging units of all products in the cluster, respectively, to determine the largest basic packaging unit of the cluster; and By treating all products in the cluster as one product and based on the determined size and weight of the largest basic packaging unit of the cluster, the integer programming model is used to determine the capacity of each level of the one or more levels of common packaging units above the largest basic packaging unit of the cluster.

9. The method according to claim 8, wherein For each clustering solution, the total logistics-related cost savings compared to the situation without the clustering solution are calculated, including: For each cluster under this clustering scheme: Calculating the logistics-related costs of all products in the cluster based on the determined capacity of the primary or multi-level common packaging units of the cluster as first logistics-related costs; Calculating the logistics-related costs of all products in the cluster based on the determined capacity of the primary or multi-level packaging unit for each product in the cluster as the second logistics-related costs; and Obtaining a logistics-related cost saving amount for the cluster by calculating a difference between the second logistics-related cost and the first logistics-related cost; and The total logistics-related cost savings of the clustering scheme is obtained by summing up the logistics-related cost savings of each cluster under the clustering scheme.

10. The method according to claim 8, wherein For each clustering solution, the total logistics-related cost savings compared to the situation without the clustering solution are calculated, including: For each cluster under this clustering scheme: calculating the logistics-related costs for all products in the cluster based on the determined capacity of the primary or multi-level common packaging units of the cluster; Obtaining third logistics-related costs of the plurality of products under the clustering scheme by summing the logistics-related costs of each cluster under the clustering scheme; calculating the logistics-related cost of the plurality of products based on the determined capacity of the primary or multi-level packaging unit of each product in the plurality of products as a fourth logistics-related cost of the plurality of products; and The total logistics-related cost savings of the clustering solution is obtained by calculating the difference between the fourth logistics-related cost and the third logistics-related cost.

11. The method according to claim 1, wherein The integer programming model is a mixed integer programming model.

12. The method according to claim 6, wherein: For each clustering solution, the total logistics-related cost savings compared to the situation where the clustering solution is not applied are calculated, including: Based on the length, width and height of the basic packaging unit of each of the multiple products, the multiple products are divided into a corresponding number of clusters of the clustering scheme using a K-means clustering algorithm.

13. A computing device, comprising: at least one processor; as well as A memory coupled to the at least one processor and configured to store instructions, wherein when the instructions are executed by the at least one processor, the at least one processor is caused to perform the method according to any one of claims 1 to 12.

14. A computer-readable storage medium having instructions stored thereon, wherein when the instructions are executed by at least one processor, the at least one processor is caused to perform the method according to any one of claims 1 to 12.

15. A computer program product comprising instructions which, when executed by at least one processor, cause the at least one processor to perform the method according to any one of claims 1 to 12.

Images