Warehousing space optimization method and device fusing in-transit cargo information, and electronic equipment

Abstract

The invention relates to a warehousing space optimization method and device fusing in-transit cargo information and electronic device.The warehousing space optimization method comprises the steps that according to the current-day initial in-warehouse number, the current-day in-transit number and the first initial in-warehouse number of each kind of cargos on the target day, the second initial in-warehouse number of each kind of cargos on the target day is determined, the initial in-library number 1 on the target day is obtained through prediction of a machine learning model; respectively determining respective space occupation change rates according to the initial in-warehouse quantity II of each type of goods on the target day and the initial in-warehouse quantity on the current day; and when there is a cargo whose space occupation change rate is greater than a target threshold value, optimizing warehouse space planning through a machine learning model. According to the method, the change of the cargo quantity can be predicted more accurately, the storage space can be optimized according to the change of the cargo quantity, and the problem that a current warehouse cargo allocation design method cannot cope with a design scene in which the cargo quantity is changed greatly is solved.

Description

Technical Field

[0001] The present application relates to the field of cargo warehousing, and in particular to a method, device and electronic equipment for optimizing storage space by integrating information of cargo in transit. Background Art

[0002] At present, the warehouse storage space design is generally only designed for the goods currently in the warehouse. If the number of different goods entering the warehouse fluctuates greatly recently, such as the sudden high temperature leading to an increase in the demand for refrigerators, freezers and other refrigeration equipment, the storage space in the warehouse needs to be redesigned.

[0003] Currently, the warehouse storage space design method only designs the storage space for the existing goods in the warehouse, which makes it unable to cope with the design scenarios where the quantity of goods changes significantly. No effective solution has been proposed so far. Summary of the invention

[0004] The present invention provides a storage space optimization method, device and electronic device integrating in-transit goods information to solve the problem that the current warehouse cargo space design method cannot cope with the design scenario where the quantity of goods changes significantly.

[0005] In a first aspect, the present invention provides a storage space optimization method integrating in-transit goods information, comprising:

[0006] The initial inventory quantity 2 on the target day is determined according to the initial inventory quantity on the day, the quantity in transit on the day and the initial inventory quantity 1 on the target day of each type of goods. The initial inventory quantity 1 on the target day is predicted by the machine learning model. The calculation formula of the initial inventory quantity 2 is:

[0007] Amount i =α*StoreAmount i +β*WayAmount i +γ*PredictAmount i

[0008] Among them, Amount i Indicates the initial inventory quantity of item i on the target date, StoreAmount i Indicates the current initial inventory quantity of item i, WayAmount i Indicates the quantity of cargo i in transit on the day, PredictAmount i represents the initial inventory quantity of item i on the target date, α, β and γ are the first weight, second weight and third weight respectively and satisfy the following conditions: α>β>γ, α>0, β>0, γ>0 and α+β+γ=1; the closer the target date is to the current date, the larger the first weight and the second weight are;

[0009] Determine the space occupancy change rate of each commodity based on the initial inventory quantity 2 of each commodity on the target date and the initial inventory quantity on the day;

[0010] When there are goods whose space occupancy change rate is greater than the target threshold, the warehouse space planning is optimized through the machine learning model.

[0011] In a second aspect, the present invention provides a warehouse cargo management method, comprising:

[0012] The storage space of the warehouse goods is optimized by the storage space optimization method integrating the in-transit goods information described in the first aspect.

[0013] Compared with the related art, the storage space optimization method provided by the present invention that integrates the in-transit goods information can optimize the storage space and make the storage of goods more reasonable. More importantly, in the optimization process, the quantity of goods in transit on the same day is taken into account to predict the initial quantity of goods in stock on the target day through the machine learning model, so that the change in the quantity of goods can be predicted more accurately, and the storage space can be optimized according to the change in the quantity of goods, which solves the problem that the current warehouse cargo space design method cannot cope with the design scenario where the quantity of goods changes greatly.

[0014] Details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flow chart of a storage space optimization method integrating in-transit goods information provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0016] In order to more clearly understand the purpose, technical solutions and advantages of the present application, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments.

[0017] Unless otherwise defined, the technical terms or scientific terms involved in this application shall have the general meaning understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "a", "the", "these" and the like in this application do not represent quantitative restrictions, and they can be singular or plural. The terms "include", "comprise", "have" and any variants thereof involved in this application are intended to cover non-exclusive inclusions; for example, a process, method and system, product or device comprising a series of steps or modules (units) is not limited to the listed steps or modules (units), but may include unlisted steps or modules (units), or may include other steps or modules (units) inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether directly or indirectly. The "multiple" involved in this application refers to two or more. "And / or" describes the association relationship of associated objects, indicating that there can be three relationships. For example, "A and / or B" can mean: A exists alone, A and B exist at the same time, and B exists alone. Usually, the character " / " indicates that the objects associated with each other are in an "or" relationship. The terms "first", "second", "third", etc. involved in this application are only used to distinguish similar objects and do not represent a specific ordering of the objects.

[0018] In an embodiment of the present invention, a storage space optimization method integrating in-transit goods information is provided. Figure 1 FIG. 1 is a flow chart of a storage space optimization method integrating in-transit goods information provided in an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0019] Step S110, determining the initial inventory quantity 2 of each cargo on the target date according to the initial inventory quantity on the target date, the in-transit quantity on the target date, and the initial inventory quantity 1 of each cargo on the target date. The initial inventory quantity 1 on the target date is predicted by the machine learning model. The calculation formula of the initial inventory quantity 2 is:

[0020] Amount i =α*StoreAmount i +β*WayAmount i +γ*PredictAmount i

[0021] Among them, Amount i Indicates the initial inventory quantity of item i on the target date, StoreAmount i Indicates the current initial inventory quantity of item i, WayAmounti Indicates the quantity of cargo i in transit on the day, PredctAmount i represents the initial inventory quantity of goods i on the target date, α, β and γ are the first weight, second weight and third weight respectively and satisfy the following conditions: α>β>γ, α>0, β>0, γ>0 and α+β+γ=1; the closer the target date is to the current date, the larger the first weight and the second weight are.

[0022] The initial inventory quantity on the day refers to the inventory quantity at the end of yesterday. For example, for a certain type of goods, its initial inventory quantity on May 4 is the inventory quantity at the end of May 3. The quantity in transit on the day is the quantity being transported to the warehouse on the day. The initial inventory quantity on the target day is predicted by the machine learning model before the day, and the machine learning model can be a large language model.

[0023] In some embodiments, the step of predicting the initial inventory quantity of each product on the target date by a machine learning model includes:

[0024] Step S111, based on the relevant data of the target goods, predict the future initial inventory quantity change of the target goods by training a complete machine learning model, and the relevant data of the target goods includes historical sales volume and current sales location.

[0025] Step S112, determining the initial inventory quantity of the target goods on the target date based on the future change in the initial inventory quantity of the target goods.

[0026] The target goods refer to the goods that need to be predicted at present. For each kind of goods, the above steps can be used to obtain the initial inventory quantity of the goods on the target day. Among them, historical sales volume is an important consideration for future sales volume. At the same time, taking into account regional differences, that is, the popularity of the same goods in different regions is different, and then the sales location also needs to be considered. The current sales location is the location of the optimized storage space. In order to predict more accurately, the relevant data of the target goods also includes the current popularity and current relevant policies. At the same time, when the sales volume of the target goods is related to the weather, the relevant data of the target goods also includes today's weather data, such as umbrellas. As follows, an example is used to illustrate the prediction of the initial inventory quantity of each kind of goods on the target day.

[0027] In one embodiment, the step of predicting the initial inventory quantity of a certain commodity on the target date includes:

[0028] S1: The user inputs the specific sales volume of the goods in previous years to make a preliminary fit to the sales forecast curve.

[0029] S2: The user enters a location to confirm the geographic location information.

[0030] S3: The user inputs the type of goods and previous manual experience in goods forecasting. The big model extracts and analyzes the key words for high-precision sales forecast curve fitting in step S4.

[0031] S4:

[0032] S4.1: The large model uses crawlers to legally crawl the Internet to obtain popular search terms, local weather data in recent days, commodity-related policy releases and other information, which depends on the type of commodity and whether it is related to the weather.

[0033] S4.2: Analyze and compare the overlap between the crawled data and the goods attributes. If the overlap is high, further analyze the number of times the information and its related information appear on the Internet, and finally further fit the sales forecast curve.

[0034] S4.3: Analyze and train the big model on the issuance of policies and the popularity of related goods and products, so that the big model can establish a relationship between the issuance of goods-related policies and the sales volume of goods, thereby calculating a more accurate forecast curve for the quantity of goods.

[0035] Among them, the cargo quantity prediction curve is a curve of cargo quantity with respect to future time. In the cargo quantity prediction curve, the initial inventory quantity of the cargo on the target day can be found.

[0036] After determining the initial inventory quantity of the goods on the day, the quantity in transit on the day, and the initial inventory quantity 1 on the target day, the initial inventory quantity 2 of the goods on the target day can be calculated by the calculation formula of the initial inventory quantity 2. The calculation formula can be understood as correcting the initial inventory quantity 1 of the goods on the target day predicted by the machine learning model by the initial inventory quantity of the goods on the day and the quantity in transit on the day.

[0037] In order to more accurately calculate the initial inventory quantity of the goods on the target date, the weight determination in the formula is one of the key factors. For the weight distribution requirement, since the initial inventory quantity of the goods on the day and the quantity in transit on the day are the actual quantities, and the initial inventory quantity of the goods on the target date is the predicted quantity, the former two have higher weights. At the same time, the closer the target date is to the day, the more the initial inventory quantity of the goods on the day and the quantity in transit on the day can determine the initial inventory quantity of the goods on the target date, so the weight of the former two is greater.

[0038] In some embodiments, the first weight is determined by:

[0039]

[0040] The formula for determining the second weight is:

[0041]

[0042] The formula for determining the third weight is:

[0043]

[0044] Where d represents the number of days between the target day and the current day.

[0045] It should be noted that the above weight determination formula is not unique, it is only a preferred embodiment. In other embodiments, other weight determination formulas can also be used under the premise of meeting the weight distribution requirements.

[0046] Step S120, determining the space occupancy change rate of each commodity according to the initial inventory quantity 2 of each commodity on the target date and the initial inventory quantity on the current day.

[0047] Specifically, the step includes:

[0048] Step S121, calculate the initial space volume of each type of cargo on the target day:

[0049] Vpredict i =Width i *Length i *Height i *Amount i

[0050] Among them, Vpredict i represents the initial space volume of cargo i on the target day, Width i , Length i and Height i Respectively represent the length, width and height of cargo i, Amount i represents the initial inventory quantity of goods i on the target date;

[0051] Step S122, respectively calculate the space occupancy change rate of each type of goods:

[0052]

[0053] Among them, Trans i represents the change rate of space occupancy of cargo i, Vpredict i represents the initial space volume of cargo i on the target day, Vpresent i Represents the initial spatial volume of cargo i on that day.

[0054] When the initial space volume of a certain cargo on the day and the initial space volume on the target day are calculated, the change rate of the cargo's space occupancy on the target day relative to the current day can be further calculated. The initial space volume of the cargo on the day and the initial space volume on the target day can be normalized, and the target threshold can be set to 0.2.

[0055] Step S130: When there are goods whose space occupancy change rate is greater than the target threshold, the warehouse space planning is optimized through the machine learning model.

[0056] When the space occupancy change rate of a certain type of goods is greater than the target threshold, it means that the current warehouse space planning may not be able to meet the space occupancy of the goods on the target day. Then the warehouse space planning needs to be optimized. The optimization constraints of warehouse space planning include: the change of the current warehouse space planning is as small as possible; the goods with larger initial inventory on the target day are closer to the outbound port, because the goods with higher outbound frequency are outbound more frequently, and at this time, the goods should be moved closer to the outbound port to reduce the moving distance when outbound. The outbound frequency has not been taken into account in future planning, so it is assumed that the larger the amount of goods, the greater the outbound frequency, and the closer they need to be to the outbound port to reduce the outbound distance and workload; height constraint, the stacking height of goods on the same cargo space; storage space spacing constraint; aisle width constraint.

[0057] Through the above method, the storage space can be optimized to make the storage of goods more reasonable. More importantly, in the optimization process, the quantity of goods in transit on the same day is taken into account to predict the initial quantity of goods in stock on the target day through the machine learning model, so that the change in the quantity of goods can be predicted more accurately, and the storage space can be optimized according to the change in the quantity of goods, solving the problem that the current warehouse cargo space design method cannot cope with the design scenario where the quantity of goods changes greatly.

[0058] In an embodiment of the present invention, a warehouse cargo management method is also provided, comprising:

[0059] The storage space of warehouse goods is optimized by the storage space optimization method integrating the in-transit goods information provided by the present invention.

[0060] In the process of warehouse cargo management, in addition to optimizing the storage space, it is also necessary to record the goods in and out of the warehouse in real time.

[0061] Specifically, a two-way guide rail is installed on the top of the warehouse to form a grid, and a four-way trolley is installed on the grid. The trolley is equipped with an infrared laser module (hereinafter referred to as infrared laser). When goods are put into the warehouse, infrared lasers are allocated to the four corners of each storage location (the storage location is a rectangle), and an infrared laser is arranged at the midpoint of each of the four sides of the storage location, for a total of eight infrared lasers (provided by the infrared laser module moving to the top of the corresponding storage location), which are used to detect the entry of goods into the storage location and the placement standards in real time. The specific performance is:

[0062] The worker holds a terminal device and scans the code of the incoming goods. The terminal screen automatically displays the location of the goods and guides the workers to place the goods into the warehouse.

[0063] The infrared laser module detects whether the length, width and height of the object entering the storage location matches the goods in the storage location based on the changes in the received infrared rays, thereby determining the quantity of goods entering the storage. If the goods are not entering the storage location in a standardized manner and exceed the storage location line, the length and width of the goods sensed by the infrared laser module will not match, thereby reminding the workers to place the goods correctly.

[0064] After the goods are put into storage, the infrared laser module of the storage location can be turned off and removed, and the next storage location can be moved to work, and the corresponding in-transit quantity is converted to the in-stock quantity. After all the goods are put into storage, go to the next step.

[0065] After the goods enter the warehouse and before the goods leave the warehouse, the system will calculate the quantity of each type of goods every half a day. If an order for in-transit goods information is received, the corresponding order number will be converted into the in-transit quantity on that day. If the goods structure in the warehouse changes significantly in a short period of time and the warehouse location needs to be adjusted, the storage space optimization method provided by the present invention will be used to optimize the storage space, that is, the large model will calculate the warehouse location change based on the existing model and the target warehouse location structure, and obtain a new warehouse location that can contain the current warehouse location and try to meet the target warehouse location structure. The optimal solution for the path and the optimal solution for the order of goods changes are given within the constraint time for the workers to transport. If the goods reach the warehouse exit link, go to the next step.

[0066] When the goods are shipped out of the warehouse, the infrared laser module is transferred to the corresponding storage location, and the terminal in the worker's hand will also prompt the location where the goods should be shipped out. After the worker has finished shipping out the goods, the laser is turned off, and the number of goods in stock is also reduced by the corresponding number of goods shipped out.

[0067] When goods are moved in, the infrared sensor detects an additional object in the storage location (the laser distance becomes shorter), and its length, width and height meet the length, width and height of the SKU that needs to be moved into the storage location, then the quantity of goods in the storage location information increases by 1. At this time, after multiple goods are moved into the storage location, the first goods put in may be covered, but the quantity of goods in the goods information still exists. After all goods are moved in, the infrared laser can be removed. After the removal, it is assumed that all goods in the storage location will not move until the quantity of goods in the storage location needs to be changed (this assumption is because, except for outbound delivery and storage location adjustment, the quantity of goods should not change).

[0068] Similarly, when goods are moved out, the infrared sensor detects that an object whose length, width and height match those of the goods in this storage location is missing from the area (the laser distance becomes longer), so the number of goods in the cargo information of this area is reduced by 1.

[0069] It can be seen that the covering of the goods does not affect the infrared laser. The infrared laser is only suitable for sensing the volume, length, width and height of objects in the area and transmitting them to the system for careful calculation.

[0070] The storage space optimization method and warehouse cargo management method provided by the present invention are further explained through an embodiment as follows.

[0071] In one embodiment, the storage space optimization and cargo management method includes:

[0072] Step S1, the user inputs the specific sales volume of the goods in previous years to make a preliminary fit to the sales forecast curve.

[0073] Step S2: The user inputs a location to confirm the geographic location information.

[0074] In step S3, the user inputs the type of goods and previous manual experience in goods forecasting, and the large model extracts and analyzes the key words for high-precision sales forecast curve fitting in step S4.

[0075] Step S4

[0076] S4.1, the big model uses crawlers to legally crawl the Internet to obtain popular search terms, local recent weather data, commodity-related policy releases and other information, which depends on the type of commodity and whether it is related to the weather.

[0077] S4.2, analyze and compare the overlap between the crawled data and the goods attributes. If the overlap is high, further analyze the number of times the information and its related information appear on the Internet, and finally further fit the sales forecast curve.

[0078] S4.3 analyzes and trains the big model on policy issuance and the popularity of related goods and products, so that the big model can establish a relationship between the issuance of goods-related policies and the sales volume of goods, thereby calculating a more accurate goods quantity prediction curve.

[0079] Step S5: The forecast curve calculated by the model can be used to predict the quantity of each cargo on each day in the future. The storage space is optimized by combining the cargo information currently in the warehouse, the cargo information of the cargo in transit, and the forecasted cargo information.

[0080] S5.1 assigns different weights to the three types of cargo information (specifically, warehouse cargo information is the highest, in-transit cargo information is the second highest, and predicted cargo information is the lowest) for warehouse location planning. The specific performance is as follows:

[0081] Assume that the quantity of goods of a SKU on a certain day in the future is, where i is the SKU of the goods, then:

[0082] Amount i =α*StoreAmount i +β*WayAmount i +γ*PredictAmount i

[0083] in:

[0084] StoreAmount i It is the quantity of a single SKU in the warehouse on that day; WayAmount i is the quantity of a single SKU in transit on that day; PredictAmount i It is the predicted quantity of a single SKU in the warehouse on a certain day.

[0085] α, β, and γ are the weights of the three, satisfying α>β>γ, α>0, β>0, γ>0, α+β+γ=1.

[0086] The formula is as follows:

[0087]

[0088] d represents the number of days between the target day and the current day.

[0089] According to the above formula, the planned inventory quantity for each SKU on that day can be calculated.

[0090] Step S6:

[0091] S6.1, based on the calculated number of warehouse locations for all SKUs and their length, width and height, calculate the estimated warehouse volume Vpredict for each SKU in the future i .

[0092] Vpredict i =Width i *Length i *Height i *Amount i

[0093] in:

[0094] i is SKU, Length i , Width i , Height i SKU i Length, width and height, Amount i SKU i The quantity of goods on a particular day.

[0095] S6.2 Calculate the change rate Trans for each SKU i .

[0096]

[0097] If Trans i >0.2, it means that the storage location of a certain SKU will change significantly in the future and needs to be replanned.

[0098] S6.3 gives each SKU to the big model i Occupancy volume Vpredict i The large model provides the optimal solution for warehouse location planning based on a series of constraints, such as the specific length, width, and height of the warehouse, the entrance and exit, the aisle width requirements, and the spacing requirements of the warehouse locations. This enables dynamic planning of warehouse locations.

[0099] S6.4, follow the following principles when dynamically planning storage locations:

[0100] 1. Make changes to the current storage location as small as possible.

[0101] 2.Amount i The larger it is, the closer the corresponding goods should be to the exit.

[0102] 3. Satisfy height constraints, distance constraints between storage locations, and aisle width constraints.

[0103] Step S7, install a two-way guide rail on the top of the warehouse to form a grid, install a four-way trolley on the grid, and equip the trolley with an infrared laser module (hereinafter referred to as infrared laser). When the goods are put into the warehouse, infrared lasers are allocated to the four corners of each storage location (the storage location is a rectangle), and an infrared laser is arranged at the midpoint of each of the four sides of the storage location, a total of eight infrared lasers, which are used to detect the entry of goods into the storage location and the placement standard in real time. The specific performance is:

[0104] S7.1, workers hold the terminal device and scan the code of the incoming goods. The terminal screen automatically displays the location of the goods and guides the workers to place the goods into the warehouse.

[0105] S7.2, infrared laser detects whether the length, width and height of the object entering the warehouse matches the SKU of the warehouse based on the changes in the received infrared light, thereby determining the quantity of goods entering the warehouse. If the goods are not entering the warehouse in a standardized manner and exceed the warehouse line, the length and width of the goods sensed by the infrared laser will not match, thereby reminding the workers to place the goods correctly.

[0106] Step S8: After the goods are put into storage, the infrared laser of the storage location can be turned off and removed, and the next storage location can be moved to work, and the corresponding WayAmount i Convert to StoreAmount i After all goods have been put into storage, go to step S9.

[0107] Step S9: From the time the goods enter the warehouse to the time they leave the warehouse, the system will calculate the amount of each SKU every half a day. If an order for in-transit goods information is received, the corresponding order number will be converted into WayAmount. i If the SKU structure in the warehouse changes significantly in the short term and the warehouse location needs to be adjusted, the large model will calculate the warehouse location change based on the existing model and the target warehouse location structure, and obtain a new warehouse location that can contain the current warehouse location and meet the target warehouse location structure as much as possible, and give the optimal path solution and the optimal solution for the order of goods change within the constraint time for workers to transport. If the goods are in the outbound link, go to step 10.

[0108] Step S10: When the goods are shipped out, the infrared laser is transferred to the corresponding storage location, and the terminal in the worker's hand will also prompt the location where the goods should be shipped out. After the worker has finished shipping out, the laser is turned off, and the StoreAmount i The number of goods leaving the warehouse will also be reduced accordingly.

[0109] In an embodiment of the present invention, a storage space optimization device integrating in-transit goods information is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and will not be repeated hereafter. The terms "module", "unit", "subunit", etc. used below may be a combination of software and / or hardware that implements predetermined functions. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0110] The storage space optimization device integrating the information of goods in transit includes:

[0111] The cargo quantity determination module is used to determine the initial inventory quantity 2 of each cargo on the target day according to the initial inventory quantity on the day, the quantity in transit on the day, and the initial inventory quantity 1 on the target day. The initial inventory quantity 1 is predicted by the machine learning model, and the calculation formula of the initial inventory quantity 2 is:

[0112] Amount i =α*StoreAmount i +β*WayAmount i +γ*PredictAmount i

[0113] Among them, Amount i Indicates the initial inventory quantity of item i on the target date, StoreAmount i Indicates the current initial inventory quantity of item i, WayAmount i Indicates the quantity of cargo i in transit on the day, PredictAmount i represents the initial inventory quantity of goods i on the target date, α, β and γ are the first weight, second weight and third weight respectively and satisfy the following conditions: α>β>γ, α>0, β>0, γ>0 and α+β+γ=1; the closer the target date is to the current date, the larger the first weight and the second weight are.

[0114] The space occupancy change determination module is used to determine the respective space occupancy change rates according to the initial inventory quantity of each type of goods on the target date and the initial inventory quantity on the day.

[0115] The warehouse space planning module is used to optimize the warehouse space planning through a machine learning model when there are goods whose space occupancy change rate is greater than the target threshold.

[0116] Through the above device, the storage space can be optimized to make the storage of goods more reasonable. More importantly, in the optimization process, the quantity of goods in transit on the same day is taken into account to predict the initial quantity of goods in stock on the target day through the machine learning model, so that the change in the quantity of goods can be predicted more accurately, and the storage space can be optimized according to the change in the quantity of goods, solving the problem that the current warehouse cargo space design method cannot cope with the design scenario where the quantity of goods changes greatly.

[0117] It should be noted that the above modules can be functional modules or program modules, and can be implemented by software or hardware. For modules implemented by hardware, the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.

[0118] In an embodiment of the present invention, an electronic device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the storage space optimization method for integrating in-transit goods information provided by the present invention.

[0119] It should be understood that the specific embodiments described herein are only used to explain the application, rather than to limit it. Based on the embodiments provided in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the protection scope of this application.

[0120] Obviously, the drawings are only some examples or embodiments of the present application. For ordinary technicians in the field, the present application can also be applied to other similar situations based on these drawings without creative work. In addition, it is understandable that although the work done in this development process may be complicated and lengthy, for ordinary technicians in the field, certain changes in design, manufacturing or production based on the technical content disclosed in this application are only conventional technical means and should not be regarded as insufficient content disclosed in this application.

Claims

1. A storage space optimization method integrating in-transit goods information, characterized in that: include: The initial inventory quantity 2 on the target day is determined according to the initial inventory quantity on the day, the quantity in transit on the day and the initial inventory quantity 1 on the target day of each type of goods. The initial inventory quantity 1 on the target day is predicted by the machine learning model. The calculation formula of the initial inventory quantity 2 is: Amount i =α*StoreAmount i +β*WayAmount i +γ*PredictAmount i Among them, Amount i Indicates the initial inventory quantity of item i on the target date, StoreAmount i Indicates the current initial inventory quantity of item i, WayAmount i Indicates the quantity of cargo i in transit on the day, PredictAmount i represents the initial inventory quantity of item i on the target date, α, β and γ are the first weight, second weight and third weight respectively and satisfy the following conditions: α>β>γ, α>0, β>0, γ>0 and α+β+γ=1; the closer the target date is to the current date, the larger the first weight and the second weight are; Determine the space occupancy change rate of each commodity based on the initial inventory quantity 2 of each commodity on the target date and the initial inventory quantity on the day; When there are goods whose space occupancy change rate is greater than the target threshold, the warehouse space planning is optimized through the machine learning model.

2. The storage space optimization method integrating in-transit goods information according to claim 1 is characterized in that: The formula for determining the first weight is: The formula for determining the second weight is: The third weight is determined by: Where d represents the number of days between the target day and the current day.

3. The storage space optimization method integrating in-transit goods information according to claim 1 is characterized in that: The steps of predicting the initial inventory quantity of each product on the target date through the machine learning model include: Based on the relevant data of the target goods, predicting the future initial inventory quantity change of the target goods through the trained machine learning model, wherein the relevant data of the target goods includes historical sales volume and current sales location; The initial inventory quantity of the target goods on the target date is determined according to the future change in the initial inventory quantity of the target goods.

4. The storage space optimization method integrating in-transit goods information according to claim 3 is characterized in that: The relevant data of the target goods also includes current popularity and current relevant policies; When the sales volume of the target goods is related to the weather, the relevant data of the target goods also includes today's weather data.

5. The method for optimizing storage space by integrating in-transit goods information according to claim 1, characterized in that: Determining the respective space occupancy change rates according to the initial inventory quantity 2 of each type of goods on the target date and the initial inventory quantity on the day includes: Calculate the initial space volume of each cargo on the target day: Vpredict i =Width i *Length i *Height i *Amount i Among them, Vpredict i represents the initial space volume of cargo i on the target day, Width i , Length i and Height i Respectively represent the length, width and height of cargo i, Amount i represents the initial inventory quantity of goods i on the target date; Calculate the change rate of space occupancy for each type of cargo separately: Among them, Trans i represents the change rate of space occupancy of cargo i, Vpredict i represents the initial space volume of cargo i on the target day, Vpresent i Represents the initial spatial volume of cargo i on that day.

6. The method for optimizing storage space by integrating in-transit goods information according to claim 5 is characterized in that: The initial space volume of the goods on the day and the initial space volume on the target day are normalized, and the target threshold is 0.

2.

7. The method for optimizing storage space by integrating in-transit goods information according to claim 5, characterized in that: The optimization constraints of the warehouse space planning include: Changes to the current warehouse space plan are as minimal as possible; Goods with larger initial inventory quantities on the target day are closer to the shipping port; Highly constrained; Storage location spacing constraints; Aisle width constraints.

8. A warehouse cargo management method, characterized in that: include: The storage space of the warehouse goods is optimized by the storage space optimization method integrating the in-transit goods information described in any one of claims 1-7.

9. A storage space optimization device integrating in-transit goods information, characterized in that: include: The goods quantity determination module is used to determine the initial inventory quantity 2 of each goods on the target day according to the initial inventory quantity on the day, the quantity in transit on the day and the initial inventory quantity 1 on the target day. The initial inventory quantity 1 is predicted by the machine learning model, and the calculation formula of the initial inventory quantity 2 is: Amount i =α*StoreAmount i +β*WayAmount i +γ*PredictAmount i Among them, Amount i Indicates the initial inventory quantity of item i on the target date, StoreAmount i Indicates the current initial inventory quantity of item i, WayAmount i Indicates the quantity of cargo i in transit on the day, PredictAmount i represents the initial inventory quantity of item i on the target date, α, β and γ are the first weight, second weight and third weight respectively and satisfy the following conditions: α>β>γ, α>0, β>0, γ>0 and α+β+γ=1; the closer the target date is to the current date, the larger the first weight and the second weight are; A space occupancy change determination module is used to determine the space occupancy change rate of each commodity according to the initial inventory quantity 2 of each commodity on the target date and the initial inventory quantity on the day; The warehouse space planning module is used to optimize the warehouse space planning through a machine learning model when there are goods whose space occupancy change rate is greater than the target threshold.

10. An electronic device comprising a memory and a processor, characterized in that: The memory stores a computer program, and the processor is configured to run the computer program to execute the storage space optimization method integrating in-transit goods information according to any one of claims 1 to 7.

Images