Warehouse management method and device, electronic equipment and storage medium

Abstract

The application provides a warehouse management method and device, electronic equipment and a storage medium. The warehouse management method comprises the following steps: determining different candidate wave division strategies according to current order information, warehouse information of a target warehouse and position information of a transportation device; determining simulation processes corresponding to each candidate wave division strategy, and allocating an independent process isolation domain and an independent simulation time axis to each simulation process corresponding to each candidate wave division strategy, wherein the process isolation domain comprises a virtual memory space; running each simulation process asynchronously based on the process isolation domain and the simulation time axis allocated to each simulation process to obtain simulation results of each simulation process; determining a target wave division strategy from the candidate wave division strategies according to the simulation results, and controlling the transportation device to transport based on the target wave division strategy. The scheme can improve the efficiency of warehouse transportation.

Description

Technical Field

[0001] This invention relates to the field of warehousing and transportation technology, specifically to a warehousing management method, a warehousing management device, an electronic device, a storage medium, and a computer program product. Background Technology

[0002] Engineering simulation fundamentally revolutionizes the traditional linear development model of "design-prototype-test-modification" in engineering. By comprehensively simulating, verifying, and optimizing products, systems, or processes, its performance and reliability can be ensured before deployment. In the field of warehousing and transportation, simulation before controlling transportation equipment for warehousing and transportation allows for the preliminary assessment of actual transportation conditions, facilitating the selection of appropriate warehousing and transportation solutions.

[0003] In existing technologies, during the execution of warehousing and transportation plans, the same system resources are used to simulate different transportation plans sequentially. These simulation processes influence each other, so each process waits for the previous one to complete. Therefore, before actual warehousing and transportation, it is necessary to wait for the simulation results of each process to be determined. Simulating each warehousing and transportation plan takes time; therefore, if there are many plans, the cumulative simulation time will increase exponentially. This consumes excessive time before actual warehousing and transportation, and the mutual influence between simulation processes reduces the accuracy of the simulation results, affecting the final judgment of the transportation plan and thus reducing the efficiency of warehousing and transportation. Summary of the Invention

[0004] The present invention was proposed in view of the above-mentioned problems.

[0005] According to a first aspect of the present invention, a warehouse management method is also provided, comprising: determining different candidate wavelet partitioning strategies based on current order information, warehouse storage information of a target warehouse, and location information of transportation equipment; determining simulation processes corresponding to each candidate wavelet partitioning strategy, and allocating independent process isolation domains and independent simulation time axes to each simulation process corresponding to each candidate wavelet partitioning strategy, wherein the process isolation domain includes virtual memory space; asynchronously running each simulation process based on the process isolation domain and simulation time axis allocated to each simulation process to obtain simulation results of each simulation process; determining a target wavelet partitioning strategy from the candidate wavelet partitioning strategies based on the simulation results, and controlling the transportation equipment to perform transportation based on the target wavelet partitioning strategy.

[0006] For example, determining the simulation process corresponding to each candidate wavelet partitioning strategy includes: generating the simulation process corresponding to each candidate wavelet partitioning strategy based on the simulation parameters corresponding to each candidate wavelet partitioning strategy.

[0007] The method further includes updating the simulation parameters according to the following update operations: determining a prior distribution based on historical data of the same type as the target simulation parameters within a preset time period before running the target simulation process, wherein the target simulation parameters are the simulation parameters corresponding to the target wavelet partitioning strategy, and the target simulation process is the simulation process corresponding to the target wavelet partitioning strategy; processing the prior distribution based on the target simulation parameters and a preset Bayesian update rule to generate a posterior distribution corresponding to the prior distribution; and updating the target simulation parameters using the posterior distribution.

[0008] For example, the preset Bayesian update rule includes an update rule based on a conjugate prior distribution, wherein the type of the prior distribution and the likelihood function type of the target simulation parameter constitute a conjugate distribution pair.

[0009] The step of processing the prior distribution based on the target simulation parameters and a preset Bayesian update rule to generate the posterior distribution corresponding to the prior distribution includes: determining the statistic of the likelihood function based on the target simulation parameters; and determining the posterior distribution through a target analytical function based on the distribution parameters of the prior distribution and the statistic, wherein the target analytical function is a function predetermined by deriving the conjugate distribution pair based on the Bayesian formula.

[0010] For example, the method further includes: acquiring real-time warehouse data of the target warehouse within the target time period during which the transportation equipment is controlled to transport based on the target wave division strategy, and of the same type as the target simulation result, wherein the target simulation result is the simulation result of the simulation process corresponding to the target wave division strategy, and the duration of the target time period is the same as the time consumed by running the target simulation process to generate the duration simulation result; for each type of real-time warehouse data, determining the difference between that type of real-time warehouse data and the target simulation result of the same type, wherein the above-mentioned updating of the simulation parameters according to the following update operation is performed when the difference between any type of real-time warehouse data and the target simulation result of the same type is greater than a preset threshold.

[0011] For example, the simulation parameters include the average moving speed of the transport equipment, the average walking speed of the picker, and the equipment failure rate of the transport equipment, wherein the equipment failure rate is determined based on the number of times the transport equipment interrupts the execution of assigned transport tasks.

[0012] For example, the method further includes: for each simulation process, determining a hash value using a hash algorithm based on the current timestamp and the identifier of the candidate wavelet partitioning strategy corresponding to each simulation process, so as to serve as the random seed for that simulation process;

[0013] The asynchronous execution of each simulation process, based on the process isolation domain and simulation timeline allocated to each simulation process, includes: asynchronously executing each simulation process based on the process isolation domain, simulation timeline, and random seed allocated to each simulation process.

[0014] For example, each simulation result includes at least one type of simulation data, and the simulation data of different simulation results are of the same type;

[0015] The step of determining the target wavelet partitioning strategy from the candidate wavelet partitioning strategies based on the simulation results includes: for each simulation data, determining the normalization result of the simulation data based on the maximum and minimum values ​​of the simulation data of the same data type as the simulation data; for each simulation result, determining the evaluation value of the simulation result based on the normalization result of all simulation data of the simulation result; and determining the target wavelet partitioning strategy from the candidate wavelet partitioning strategies based on the evaluation value of each simulation result.

[0016] For example, the simulation data includes the amount of goods leaving the warehouse within a preset unit of time, the average walking speed of the transportation equipment, and the equipment utilization rate of the transportation equipment.

[0017] For example, the warehousing information includes the distribution of goods and the distribution of workstations.

[0018] According to a second aspect of the present invention, a warehouse management device is also provided, comprising:

[0019] The strategy generation module is used to determine different candidate wavelet division strategies based on current order information, target warehouse storage information, and transportation equipment location information;

[0020] The process generation module is used to determine the simulation process corresponding to each candidate wave partitioning strategy, and to allocate an independent process isolation domain and an independent simulation time axis to the simulation process corresponding to each candidate wave partitioning strategy, wherein the process isolation domain includes a virtual memory space.

[0021] The simulation execution module is used to asynchronously run each simulation process based on the process isolation domain and simulation time axis allocated to each simulation process, so as to obtain the simulation results of each simulation process.

[0022] The transportation control module is used to determine the target wave division strategy from the candidate wave division strategies based on the simulation results, and control the transportation equipment to carry out transportation based on the target wave division strategy.

[0023] According to a third aspect of the present invention, an electronic device is also provided, comprising: a processor and a memory, wherein the memory stores computer program instructions, which, when executed by the processor, are used to perform the above-described warehouse management method.

[0024] According to a fourth aspect of the present invention, a storage medium is also provided, on which program instructions are stored, which, when executed, are used to perform the above-described warehouse management method.

[0025] According to a fifth aspect of the present invention, a computer program product is also provided, comprising computer program instructions that, when executed, perform the above-described warehouse management method.

[0026] In the above technical solution, different candidate wavelet partitioning strategies are determined based on current order information, target warehouse storage information, and transportation equipment location information. Then, simulation processes corresponding to each candidate wavelet partitioning strategy are determined, and each simulation process is allocated an independent process isolation domain and an independent simulation time axis. The process isolation domain includes virtual memory space. Subsequently, each simulation process is run asynchronously based on its allocated process isolation domain and simulation time axis to obtain simulation results. Based on the simulation results, a target wavelet partitioning strategy is determined from the candidate strategies, and the transportation equipment is controlled to transport according to the target wavelet partitioning strategy. When simulating different wavelet partitioning strategies, each simulation process is isolated at the process level and run asynchronously. Simulation processes corresponding to different wavelet partitioning strategies do not compete for the same system resources and do not interfere with each other. This improves the independence of the simulation results of each simulation process, making it easier to select the most suitable wavelet partitioning strategy for the actual situation, and reducing the total time required for simulation before actual warehousing and transportation, thus improving the efficiency of warehousing and transportation.

[0027] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0028] The above and other objects, features, and advantages of the present invention will become more apparent from the more detailed description of the embodiments of the invention in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same parts or steps.

[0029] Figure 1A schematic flowchart of a warehouse management method according to an embodiment of the present invention is shown;

[0030] Figure 2 A schematic flowchart of an update operation according to an embodiment of the present invention is shown;

[0031] Figure 3 A schematic flowchart illustrating the generation of the posterior distribution corresponding to the prior distribution according to an embodiment of the present invention is shown.

[0032] Figure 4 A schematic flowchart illustrating the determination of execution conditions for an update operation according to an embodiment of the present invention is shown;

[0033] Figure 5 A schematic flowchart illustrating the simulation process running according to a random seed according to an embodiment of the present invention is shown;

[0034] Figure 6 A schematic flowchart illustrating the determination of a target wave division strategy based on simulation results according to an embodiment of the present invention is shown.

[0035] Figure 7 A schematic block diagram of a warehouse management device according to an embodiment of the present invention is shown;

[0036] Figure 8 A schematic block diagram of an electronic device according to an embodiment of the present invention is shown. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of the present invention more apparent, exemplary embodiments according to the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are merely a part of the embodiments of the present invention, and not all of the embodiments of the present invention. It should be understood that the present invention is not limited to the exemplary embodiments described herein. Based on the embodiments of the present invention described herein, all other embodiments obtained by those skilled in the art without inventive effort should fall within the protection scope of the present invention.

[0038] To at least partially address the aforementioned problems, a warehouse management method is proposed. This method determines different candidate wavelet partitioning strategies based on current order information, target warehouse storage information, and transportation equipment location information. Then, it identifies the simulation process corresponding to each candidate wavelet partitioning strategy and allocates an independent process isolation domain and simulation time axis to each simulation process. The process isolation domain includes virtual memory space. Subsequently, based on the allocated process isolation domain and simulation time axis for each simulation process, the simulation processes are run asynchronously to obtain simulation results. Based on the simulation results, a target wavelet partitioning strategy is determined from the candidate strategies, and the transportation equipment is controlled to transport goods according to the target wavelet partitioning strategy. When simulating different wavelet partitioning strategies, each simulation process is isolated at the process level and run asynchronously. Simulation processes corresponding to different wavelet partitioning strategies do not compete for the same system resources and do not interfere with each other. This improves the independence of the simulation results of each simulation process, facilitates the selection of the most suitable wavelet partitioning strategy for the actual situation, reduces the total time required for simulation before actual warehousing and transportation, and improves the efficiency of warehousing and transportation.

[0039] Figure 1 A schematic flowchart of a warehouse management method according to an embodiment of the present invention is shown. Figure 1 As shown, the warehouse management method may include steps S110 to S140.

[0040] In step S110, different candidate wave division strategies are determined based on the current order information, the storage information of the target warehouse, and the location information of the transportation equipment.

[0041] The current order information includes at least information indicating which goods need to be transported from where and to where. For example, the current order information may also include information about which types of transport equipment will be used to transport which goods.

[0042] Current order information can include unexecuted orders and pending orders.

[0043] Warehouse information can include information related to storage locations, goods, and workstations in the target warehouse. This warehouse information can determine the specific wave division of the identified candidate wave division strategy.

[0044] For example, warehouse information includes the distribution of goods and the distribution of workstations.

[0045] The distribution of goods can include one or more types of goods storage information, such as storage location occupancy (indicating which storage locations have been occupied by goods) and SKU (Stock Keeping Unit) distribution (indicating the type and quantity of SKUs stored in each storage location).

[0046] Workstation distribution information can include the location of each type of workstation and the workstation occupancy status (e.g., which workstations are occupied and the pending tasks at occupied workstations).

[0047] The candidate wave division strategy is determined based on the current order information, the distribution of goods in the target warehouse, the distribution of workstations, and the location information of transportation equipment. By combining the aspects of orders, goods, workstations, and transportation equipment, a more accurate candidate wave division strategy that meets the expectations can be determined.

[0048] Transportation equipment may include Automated Guided Vehicles (AGVs), Autonomous Mobile Robots (AMRs), and modular storage and retrieval robots.

[0049] For example, warehouse information can also include location type, type of transport equipment, etc. Understandably, different types of locations may require different types of transport equipment for picking and storing goods. For instance, goods at higher locations need to be transported using rack-mounted transport equipment, and then transferred to lower locations using ground-mounted transport equipment for further transport. Goods at lower locations may only require ground-mounted transport equipment. Therefore, combining information such as location type and transport equipment type can further and more accurately determine a more suitable candidate wavelet segmentation strategy.

[0050] For example, suitable detection components can be used to collect real-time storage information of the target warehouse and location information of transportation equipment. For instance, detection components (such as cameras, pressure sensors, etc.) installed at storage locations can be used to collect cargo information for each location to obtain storage information. Similarly, positioning components installed on transportation equipment can be used to detect the location information of the transportation equipment.

[0051] For example, the storage information of the target warehouse can be determined based on the inbound and outbound records.

[0052] For example, the location information of the transportation equipment can be determined based on the location data fed back by the transportation equipment.

[0053] For example, an inventory snapshot of the target warehouse can be received from the warehouse management system to determine the warehouse's storage information and the location information of the transportation equipment. It can also record each version of the inventory snapshot received from the warehouse management system. If the warehouse management system's interface returns an exception, the received inventory snapshot data fields are missing or incorrectly formatted, validation fails, network transmission is interrupted resulting in an incomplete snapshot, or the snapshot timestamp is inconsistent with the recorded version, it indicates that a reliable inventory snapshot cannot be received normally at this time. In this case, the storage information of the target warehouse and the location information of the transportation equipment can be determined based on the previously recorded inventory snapshot.

[0054] It is understandable that for the same target warehouse, the wave division strategy determined by different division rules will have different impacts on the actual results of subsequent control of transportation equipment for transportation, based on the storage information and location information of transportation equipment.

[0055] The partitioning rules for candidate wavelet partitioning strategies can include order SKU distribution similarity, warehouse location heatmap density, SLA (Service Level Agreement) time window constraints, and work shift partitioning. Regarding order SKU distribution similarity, this partitioning rule aims to group orders containing the same SKU into the same wavelet to reduce overlapping picking paths. For warehouse location heatmap density, this partitioning rule aims to generate warehouse location heatmaps based on historical order information (e.g., frequently accessed warehouse locations are marked as "hot zones"), prioritizing the grouping of orders in hot zones into the same wave for centralized processing. Regarding SLA time window constraints, this partitioning rule aims to group orders with similar SLA time requirements into the same wave to avoid delays for urgent orders. Regarding work shift partitioning, this partitioning rule aims to group orders corresponding to different work shifts (e.g., morning shift, evening shift) into different waves.

[0056] The partitioning rule type used for each wave division strategy can be selected according to actual needs, and will not be detailed here.

[0057] In step S120, the simulation process corresponding to each candidate wave partitioning strategy is determined, and an independent process isolation domain and an independent simulation time axis are allocated to the simulation process corresponding to each candidate wave partitioning strategy. The process isolation domain includes a virtual memory space.

[0058] Candidate wavelet segmentation strategies can be used as input data, and simulation software, large language models, and simulation process templates can be used to generate simulation processes corresponding to each candidate wavelet segmentation strategy. For example, for each candidate wavelet segmentation strategy, the corresponding simulation parameters can be determined based on relevant information of the candidate wavelet segmentation strategy (such as time window, number of orders, cargo attributes, warehouse aisle congestion threshold, etc.). These simulation parameters can then be used to fill in the simulation process template corresponding to the candidate wavelet segmentation strategy or input into the simulation software to obtain the simulation process corresponding to the candidate wavelet segmentation strategy. Alternatively, the relevant information of the candidate wavelet segmentation strategy can be input into a large language model, which can then use the simulation process to create simulation processes. Other methods are similar and will not be detailed here.

[0059] For example, simulation processes with independent process isolation domains and independent simulation timelines can be created based on the operating system type using corresponding system calls or by utilizing independent process creation tools. For instance, for Unix / Linux systems, the fork() system call can be used to create simulation processes corresponding to each candidate wave partitioning strategy. Alternatively, Docker container technology can be used to create separate execution environments for each simulation process corresponding to a candidate wave partitioning strategy, thereby allocating an independent process isolation domain and an independent simulation timeline to each simulation process.

[0060] For example, each simulation process can be assigned an independent process isolation domain and an independent simulation time axis according to the creation order of the simulation processes and a preset allocation rule.

[0061] Understandably, by allocating independent process isolation domains and independent simulation timelines to each simulation process, the simulation processes do not share memory space and simulation timelines, meaning that each simulation process has its own independent data cache and global variables. During subsequent runtime, each simulation process does not need to use a unified simulation timeline or occupy the same system resources, and there is virtually no mutual influence, thus achieving process-level isolation for each simulation process.

[0062] For example, a process isolation domain can also include a virtual kernel.

[0063] In step S130, each simulation process is run asynchronously based on the process isolation domain and simulation time axis allocated to each simulation process to obtain the simulation results of each simulation process.

[0064] Because each simulation process is allocated an independent process isolation domain and an independent simulation time axis, and each simulation process is isolated at the process level, each simulation process can run asynchronously with relatively high reliability, and the operation of each simulation process will not affect other simulation processes.

[0065] For example, simulation processes can be run asynchronously from the same moment, based on the process isolation domain and simulation time axis allocated to each simulation process, thereby minimizing the time required to obtain the simulation results of each simulation process.

[0066] Simulation results can be statistically analyzed during the simulation process according to actual needs. For example, simulation results can include the quantity of goods entering and leaving the warehouse, the number of completed orders, and the distance traveled by transportation equipment.

[0067] Simulation results may include at least one type of simulation data (such as the aforementioned quantity of goods entering and leaving the warehouse, order completion volume, and distance traveled by transportation equipment), or a comprehensive evaluation result calculated by combining simulation data.

[0068] In step S140, based on the simulation results, the target wave division strategy is determined from the candidate wave division strategies, and the transportation equipment is controlled to carry out transportation based on the target wave division strategy.

[0069] Based on simulation results and the preset calculation rules used in practice, the target wavelet partitioning strategy can be determined from the candidate wavelet partitioning strategies. Taking the simulation result as the comprehensive evaluation value as an example, under the preset calculation rules, when a larger simulation result indicates a greater deviation from the expectation, the candidate wavelet partitioning strategy corresponding to the smallest simulation result can be selected as the target wavelet partitioning strategy. Similarly, under the preset calculation rules, when the overall data of the simulation result deviates further from the expectation, the candidate wavelet partitioning strategy corresponding to the simulation result with the smallest overall deviation can be selected as the target wavelet partitioning strategy. Specific preset calculation rules can be set according to actual needs and will not be detailed here.

[0070] Understandably, based on the simulation results, the determined target wave division strategy should be the one among the candidate wave division strategies that makes the actual result of controlling the transportation equipment to carry out transportation closest to the expectation. Therefore, controlling the transportation equipment to carry out transportation based on the target wave division strategy can basically ensure that the subsequent transportation situation meets the expectations.

[0071] For example, the transportation task of each transportation device can be determined first according to the target wave division strategy, and then the task can be sent to the corresponding device to control the transportation device to carry out transportation based on the target wave division strategy.

[0072] For example, the target wave division strategy can be distributed to the transportation equipment, allowing each transportation equipment to determine its own transportation task.

[0073] In the above technical solution, different candidate wavelet partitioning strategies are determined based on current order information, target warehouse storage information, and transportation equipment location information. Then, simulation processes corresponding to each candidate wavelet partitioning strategy are determined, and each simulation process is allocated an independent process isolation domain and an independent simulation time axis. The process isolation domain includes virtual memory space. Subsequently, each simulation process is run asynchronously based on its allocated process isolation domain and simulation time axis to obtain simulation results. Based on the simulation results, a target wavelet partitioning strategy is determined from the candidate strategies, and the transportation equipment is controlled to transport according to the target wavelet partitioning strategy. When simulating different wavelet partitioning strategies, each simulation process is isolated at the process level and run asynchronously. Simulation processes corresponding to different wavelet partitioning strategies do not compete for the same system resources and do not interfere with each other, thus improving the independence of the simulation results of each simulation process. This facilitates the selection of the most suitable wavelet partitioning strategy for the actual situation and reduces the total time required for simulation before actual warehousing and transportation, thereby improving the efficiency of warehousing and transportation.

[0074] For example, determining the simulation process corresponding to each candidate wavelet partitioning strategy includes: generating the simulation process corresponding to each candidate wavelet partitioning strategy based on the simulation parameters corresponding to each candidate wavelet partitioning strategy.

[0075] Simulation parameters corresponding to each candidate wavelet partitioning strategy can be extracted according to preset rules, preset algorithms, large language models, etc., without limitation here. After obtaining the simulation parameters corresponding to each candidate wavelet partitioning strategy, simulation processes corresponding to each candidate wavelet partitioning strategy can be generated using simulation process templates, large language models, etc., without limitation here. Alternatively, simulation processes corresponding to each candidate wavelet partitioning strategy can be generated according to the steps in the above embodiments based on the simulation parameters corresponding to each candidate wavelet partitioning strategy.

[0076] For example, simulation parameters of the simulation process can be recorded after each simulation, and the recorded simulation parameters of each simulation process can be used as the simulation parameters corresponding to each candidate wavelet partitioning strategy.

[0077] For example, simulation parameters include the average moving speed of the transport equipment, the average walking speed of the pickers, and the equipment failure rate of the transport equipment, wherein the equipment failure rate is determined based on the number of times the transport equipment interrupts the execution of assigned transport tasks. The average moving speed of the transport equipment and the average walking speed of the pickers can be used to calculate the handling time of orders between different storage locations, and the equipment failure rate can be used to trigger random equipment interruption events during the simulation process.

[0078] For example, simulation parameters may also include parameters that need to be considered in warehousing and transportation, such as order execution order and the order of storing and retrieving goods for transportation equipment for each storage location.

[0079] The aforementioned warehouse management method also includes updating simulation parameters according to the following update operations. Figure 2 A schematic flowchart of an update operation according to an embodiment of the present invention is shown. Figure 2 As shown, the above update operation may include steps S210 to S230.

[0080] In step S210, a priori distribution is determined based on historical data within a preset time period prior to running the target simulation process, which is of the same type as the target simulation parameters. Here, the target simulation parameters are the simulation parameters corresponding to the target wave division strategy, and the target simulation process is the simulation process corresponding to the target wave division strategy.

[0081] Historical data can represent the actual transportation situation of the target warehouse. Determining the prior distribution based on historical data can provide accurate initial hyperparameters (such as α and β for a Beta distribution, or μ and σ for a normal distribution), which can reduce the amount of target simulation parameters required as observation data.

[0082] The preset duration can be a fixed value or can be set according to needs.

[0083] In step S220, the prior distribution is processed according to the target simulation parameters and based on the preset Bayesian update rule to generate the posterior distribution corresponding to the prior distribution.

[0084] The target simulation parameters can be used as observation data, and the posterior distribution corresponding to the prior distribution can be calculated using a preset Bayesian update rule and the prior distribution. The preset Bayesian update rule can include variational inference, Kalman filtering, particle filtering, update rules based on conjugate priors and Bayes' theorem, etc.

[0085] In step S230, the target simulation parameters are updated using the posterior distribution.

[0086] This posterior distribution mathematically "weights and combines" historical data with the target simulation parameters, generating a complete probability distribution that includes both old and new information. The target simulation parameters can be replaced with the simulation parameters corresponding to this posterior distribution.

[0087] In the above technical solution, a prior distribution is determined based on historical data within a preset time period prior to the execution of the target simulation process, which is of the same type as the target simulation parameters. Here, the target simulation parameters are the simulation parameters corresponding to the target wavelet partitioning strategy, and the target simulation process is the simulation process corresponding to the target wavelet partitioning strategy. Then, based on the target simulation parameters, the prior distribution is processed according to a preset Bayesian update rule to generate a posterior distribution corresponding to the prior distribution. The posterior distribution is then used to update the target simulation parameters. By deriving the target simulation parameters based on the preset Bayesian update rule and historical data, the determined new target simulation parameters better reflect the actual transportation situation of the target warehouse. After updating the target simulation parameters, the accuracy of subsequent simulation results of the target simulation process can be improved.

[0088] For example, the preset Bayesian update rules include update rules based on conjugate prior distributions, where the type of the prior distribution and the likelihood function type of the target simulation parameter constitute a conjugate distribution pair. Figure 3 A schematic flowchart illustrating the generation of the posterior distribution corresponding to the prior distribution according to an embodiment of the present invention is shown. Figure 3 As shown, step S220 may include steps S310 to S320.

[0089] In step S310, the statistics of the likelihood function are determined based on the target simulation parameters.

[0090] In step S320, the posterior distribution is determined by the objective analytical function based on the distribution parameters and statistics of the prior distribution. The objective analytical function is a function that is predetermined by deriving the conjugate distribution pair based on Bayes' theorem.

[0091] The likelihood function can be calculated based on the distribution of the target simulation parameters.

[0092] Taking the moving speed of the transportation equipment as an example, as the target simulation parameter, if it follows a normal distribution N(μ=1.2m / s, σ 2 Taking (e.g., 0.01), the likelihood function statistic can be calculated based on the following formula 1:

[0093] Formula 1

[0094] Where μ = 1.2 m / s and σ 2 =0.01 is only an example value, D i Let represent the i-th sampled value of the moving speed of the transportation equipment. P(D|θ) represents the probability of observing historical data D under the target simulation parameter θ, i.e., the statistic of the likelihood function.

[0095] Substituting the likelihood function into Bayes' theorem, we obtain the following formula 2 as the objective analytic function:

[0096] Formula 2

[0097] Wherein, P(θ|D) is the posterior distribution, representing the corrected distribution after observing historical data and benchmarking the simulation parameter θ.

[0098] Based on this objective analytic function, when the type of the prior distribution and the type of the likelihood function of the historical data form a conjugate distribution pair, a posterior distribution of the same type as the prior distribution can be calculated. Taking a normal prior distribution as an example, the likelihood function used can satisfy a normal distribution, and the prior distribution also satisfies a normal distribution.

[0099] In the above technical solution, the likelihood function statistic is determined based on historical data. Then, based on the distribution parameters and statistic of the prior distribution, the posterior distribution is determined through an objective analytical function. This objective analytical function is a pre-determined function derived from the conjugate distribution pairs using Bayes' theorem. Using conjugate prior pairs and Bayes' theorem to determine the posterior distribution reduces the computational complexity of the posterior distribution calculation.

[0100] Figure 4 A schematic flowchart illustrating the determination of execution conditions for an update operation according to an embodiment of the present invention is shown. Figure 4 As shown, the above warehouse management method may further include steps S410 to S420.

[0101] In step S410, real-time warehouse data of the target warehouse, which is of the same type as the target simulation result, is obtained within the target time period during which the control of the transportation equipment is carried out based on the target wave division strategy. The target simulation result is the simulation result of the simulation process corresponding to the target wave division strategy, and the duration of the target time period is the same as the time consumed by running the target simulation process to generate the duration simulation result.

[0102] The target time period can be the period from the start of the transportation equipment being controlled based on the target wave division strategy to any time thereafter.

[0103] For example, the target time period could be the time period starting from the moment when the transportation equipment begins to operate based on the target wave division strategy. This ensures that the final real-time warehouse data most closely approximates the corresponding simulation results.

[0104] In step S420, for each type of real-time warehouse data, the difference between that type of real-time warehouse data and the target simulation result of the same type is determined, wherein the above-mentioned update of simulation parameters according to the following update operation is performed when the difference between any type of real-time warehouse data and the target simulation result of the same type is greater than a preset threshold.

[0105] When the difference between any real-time warehouse data and the target simulation result of the same type is greater than a preset threshold, it indicates that the target simulation process does not reflect the real transportation situation well in some aspects. Therefore, the simulation parameters need to be updated so that the simulation results are closer to the actual transportation situation when the updated simulation parameters are used for the next simulation.

[0106] For example, the preset threshold can be a fixed value set in advance, or it can be a calculated value determined based on real-time warehouse data or target simulation results and a preset ratio (e.g., 20% of real-time warehouse data, 25% of target simulation results, etc.).

[0107] In the above technical solution, real-time warehouse data of the target warehouse, within the target time period during which the transportation equipment is controlled based on the target wave division strategy, and of the same type as the target simulation result, is obtained. The target simulation result is the simulation result of the simulation process corresponding to the target wave division strategy, and the duration of the target time period is the same as the time consumed by the target simulation process to generate the simulation result. Then, for each type of real-time warehouse data, the difference between that type of real-time warehouse data and the target simulation result of the same type is determined. The simulation parameters are updated according to the following update operation only when the difference between any type of real-time warehouse data and the target simulation result of the same type is greater than a preset threshold. Thus, the simulation parameters of the simulation process are only updated when the simulation result differs significantly from the actual transportation situation, thereby reducing the update frequency of the simulation process and ensuring the stability of the simulation results.

[0108] Figure 5 A schematic flowchart illustrating the simulation process running according to a random seed is shown, based on an embodiment of the present invention. Figure 5 As shown, the above warehouse management method may further include step S510, and step S130 may include step S520.

[0109] In step S510, for each simulation process, a hash value is determined by a hash algorithm based on the current timestamp and the identifier of the candidate wavelet partitioning strategy corresponding to each simulation process, and is used as the random seed for that simulation process.

[0110] The identifier for a candidate wavelet partitioning strategy can be a unique identifier representing information such as the name and number of the candidate wavelet partitioning strategy. The hash algorithm can include encryption hash algorithms such as MD5, SM3, and SHA-256.

[0111] The current timestamp can be determined based on the time when the candidate wavelet partitioning strategy was generated, or it can be determined based on the time when each simulation process was generated. If the random seed for the simulation process is determined sequentially, the current time can also be used directly as the current timestamp.

[0112] For candidate wavelet partitioning strategies, the identifier of the candidate wavelet partitioning strategy and the current timestamp can be concatenated to obtain the concatenated result. Then, a hash algorithm is used to calculate the hash value of each concatenated result to serve as the random seed for each simulation process.

[0113] The current timestamp and the identifier of the candidate wavelet partitioning strategy corresponding to each simulation process will not be completely the same. Even if they are partially the same, the random seeds of each simulation process determined by the hash algorithm will be different.

[0114] In step S520, each simulation process is run asynchronously based on the process isolation domain, simulation time axis and random seed allocated to each simulation process.

[0115] The random seed is an initial value used to initialize the random number generator. Its core function is to ensure that all "random" events during the simulation (such as order arrival, picking time, and equipment failure) can be accurately reproduced. Each simulation process can be run asynchronously based on a process isolation domain, simulation timeline, and random seed allocated separately for each simulation process, based on any feasible implementation method; no restrictions are imposed here.

[0116] In the above technical solution, for each simulation process, a hash value is determined using a hash algorithm based on the current timestamp and the identifier of the candidate wavelet partitioning strategy corresponding to each simulation process. This hash value serves as the random seed for that simulation process. Then, each simulation process is run asynchronously based on its allocated process isolation domain, simulation timeline, and random seed. Combining the current timestamp, the identifier of the candidate wavelet partitioning strategy corresponding to the simulation process, and the random seed determined by the hash algorithm ensures the independence of the random seed for each simulation process, guarantees the reliability of the simulation results, and makes the final selected target wavelet strategy more in line with expectations.

[0117] For example, each simulation result includes at least one type of simulation data, and the simulation data of different simulation results are of the same type. Figure 6 A schematic flowchart illustrating a target wavelet division strategy based on simulation results according to an embodiment of the present invention is shown. Figure 6 As shown, step S140 may include steps S610 to S630.

[0118] In step S610, for each simulation data, the normalization result of the simulation data is determined based on the maximum and minimum values ​​of the simulation data of the same data type as the simulation data in each simulation result.

[0119] Simulation data can be data that reflects the overall simulation situation.

[0120] For example, the simulation data includes the amount of goods leaving the warehouse within a preset unit of time, the average walking speed of the transportation equipment, and the equipment utilization rate of the transportation equipment.

[0121] The outbound volume of goods within a preset unit of time reflects the outbound efficiency during the simulation process. The average travel speed of the transport equipment reflects its utilization level and also indirectly reflects its operating energy consumption. The equipment utilization rate of the transport equipment reflects the efficiency of its use and also reflects the efficiency of the simulation process.

[0122] The equipment utilization rate of transportation equipment can be determined based on the number of used and unused equipment. Alternatively, it can be determined based on the time taken for the transportation equipment to perform its tasks and the duration of the simulation process.

[0123] For example, simulation data may also include order completion volume,

[0124] For example, for each simulation data, the first difference between the maximum and minimum values ​​of the simulation data can be calculated first, then the difference between the maximum value and the simulation data can be calculated as the second difference, and then the ratio of the second difference to the first difference can be calculated as the normalization result of the simulation data.

[0125] For example, for each simulation data, the first difference between the maximum and minimum values ​​of the simulation data can be calculated first, then the difference between the simulation data and the minimum value can be calculated as the second difference, and then the ratio of the second difference to the first difference can be calculated as the normalization result of the simulation data.

[0126] For example, the calculation method for each type of simulation data can be determined based on the expected outcome of each type of simulation data. Then, based on the determined calculation method, the aforementioned maximum and minimum values, the normalized result for each simulation data can be determined separately. For instance, regarding the average utilization rate of equipment, from a cost-benefit perspective, a higher average utilization rate is desirable. In this case, for each equipment's average utilization rate, the calculation method can first calculate the first difference between the maximum and minimum values ​​of the average utilization rate, then calculate the difference between the maximum value and the average utilization rate as the second difference, and finally calculate the ratio of the second difference to the first difference as the normalized result of the average utilization rate. Similarly, regarding the travel distance of transportation equipment, from a cost-benefit perspective, a lower travel distance is desirable. In this case, for each transportation equipment's travel distance, the calculation method can first calculate the first difference between the maximum and minimum values ​​of the travel distance, then calculate the difference between the travel distance and the minimum value as the second difference, and finally calculate the ratio of the second difference to the first difference as the normalized result of the average utilization rate.

[0127] For example, the normalized result of each smaller simulation data in the k-th simulation result can be calculated according to the following formula 3:

[0128] norm j =(KPJ k,j -KPJ min,j ) / (KPJ max,j -KPJ min,j ) Formula 3

[0129] Among them, norm j KPJ represents the normalized result of the j-th simulation data in the k-th simulation result. k,j KPJ represents the value of the j-th simulation data in the k-th simulation result. min,j KPJ represents the minimum value of simulation data with the same data type as the j-th simulation data. max,j This represents the maximum value of the simulation data that is of the same data type as the j-th simulation data.

[0130] For example, the normalized result of each larger (best) simulation data in the k-th simulation result can be calculated according to the following formula 4:

[0131] norm j =(KPI max,j -KPI k,j ) / (KPI max,j -KPI min,j ) Formula 4

[0132] Among them, norm j KPI represents the normalized result of the j-th simulation data in the k-th simulation result. k,j KPI represents the value of the j-th simulation data in the k-th simulation result. min,j KPI represents the minimum value of simulation data with the same data type as the j-th simulation data. max,j This represents the maximum value of the simulation data that is of the same data type as the j-th simulation data.

[0133] Different simulation data have different units (for example, the unit of the number of outbound items is pieces / second, and the unit of the distance traveled by the transportation equipment is meters), so they cannot be directly added together. After normalizing to the [0,1] interval, they can be directly weighted and summed to calculate the evaluation value.

[0134] In step S620, for each simulation result, the evaluation value of the simulation result is determined based on the normalized result of all simulation data of the simulation result.

[0135] The evaluation value of the simulation result can be calculated based on the normalized results of all simulation data and their corresponding weight coefficients. The weight coefficients corresponding to the normalized results can be fixed values ​​or determined in real time as needed.

[0136] For example, the weighting coefficients for each normalization result can be determined based on the storage type of the target warehouse. Different storage types have different focuses for different types of simulation data. For example, in e-commerce, high-frequency small-item storage prioritizes outbound efficiency (the weighting coefficient for the number of outbound items can be taken as 0.6-0.7), while in manufacturing, low-frequency large-item storage prioritizes equipment utilization (the weighting coefficient for equipment utilization can be taken as 0.5-0.6).

[0137] For example, the evaluation value of the simulation result corresponding to the k-th candidate wavelet partitioning strategy can be calculated according to the following formula 5:

[0138] Score k =∑ω j ·norm j Formula 5

[0139] Among them, Score k ω represents the evaluation value of the simulation result corresponding to the simulation process of the k-th candidate wavelet partitioning strategy. j The weight coefficient represents the normalization result of the j-th simulation data.

[0140] In step S630, the target wave partitioning strategy is determined from the candidate wave partitioning strategies based on the evaluation value of each simulation result.

[0141] Based on simulation results and the preset calculation rules used in practice, the target wavelet partitioning strategy can be determined from the candidate wavelet partitioning strategies. For example, under the preset calculation rules, when a larger simulation result indicates a greater deviation from the expected value, the candidate wavelet partitioning strategy corresponding to the smallest simulation result can be selected as the target wavelet partitioning strategy. Alternatively, under the preset calculation rules, the candidate wavelet partitioning strategy corresponding to the simulation result with the smallest deviation from the expected value can be selected as the target wavelet partitioning strategy. Specific preset calculation rules can be set according to actual needs and will not be detailed here.

[0142] In the above technical solution, for each simulation data, the normalization result of the simulation data is determined based on the maximum and minimum values ​​of the simulation data of the same data type as each simulation result. Then, for each simulation result, the evaluation value of the simulation result is determined based on the normalization result of all simulation data for that simulation result. Finally, based on the evaluation value of each simulation result, the target wavelet partitioning strategy is determined from the candidate wavelet partitioning strategies. This allows for the unification of the dimensions of different types of simulation data, and the evaluation value of each simulation result alone can be used to assess whether each candidate wavelet partitioning strategy meets the requirements of the target candidate wavelet partitioning strategy, thereby reducing computational complexity.

[0143] Figure 7 A schematic block diagram of a warehouse management device according to an embodiment of the present invention is shown. Figure 7 As shown, the warehouse management device includes a strategy generation module 710, a process generation module 720, a simulation execution module 730, and a transportation control module 740.

[0144] The strategy generation module 710 is used to determine different candidate wave division strategies based on the current order information, the storage information of the target warehouse, and the location information of the transportation equipment.

[0145] The process generation module 720 is used to determine the simulation process corresponding to each candidate wave partitioning strategy, and to allocate an independent process isolation domain and an independent simulation time axis to the simulation process corresponding to each candidate wave partitioning strategy. The process isolation domain includes a virtual memory space.

[0146] The simulation execution module 730 is used to asynchronously run each simulation process based on the process isolation domain and simulation time axis allocated to each simulation process, so as to obtain the simulation results of each simulation process.

[0147] The transportation control module 740 is used to determine the target wave division strategy from the candidate wave division strategies based on the simulation results, and control the transportation equipment to carry out transportation based on the target wave division strategy.

[0148] For example, the process generation module 720 may include a first generation submodule, and the warehouse management device may further include an update module. The first generation submodule is used to generate simulation processes corresponding to each candidate wavelet partitioning strategy based on the simulation parameters corresponding to each candidate wavelet partitioning strategy. The update module is used to update the simulation parameters according to the following update operations: determining a prior distribution based on historical data of the same type as the target simulation parameters within a preset time period before running the target simulation process, wherein the target simulation parameters are the simulation parameters corresponding to the target wavelet partitioning strategy, and the target simulation process is the simulation process corresponding to the target wavelet partitioning strategy; processing the prior distribution based on a preset Bayesian update rule according to the target simulation parameters to generate a posterior distribution corresponding to the prior distribution; and updating the target simulation parameters using the posterior distribution.

[0149] For example, the preset Bayesian update rules include update rules based on conjugate prior distributions, where the type of the prior distribution and the likelihood function type of the target simulation parameters constitute a conjugate distribution pair. The update module may include a statistics calculation submodule and an analysis submodule. The statistics calculation submodule is used to determine the statistics of the likelihood function based on the target simulation parameters. The analysis submodule is used to determine the posterior distribution based on the distribution parameters and statistics of the prior distribution, using a target analytical function, where the target analytical function is a function pre-determined by deriving the conjugate distribution pair using Bayes' theorem.

[0150] For example, the warehouse management device further includes a real-time data acquisition module and a difference calculation module. The real-time data acquisition module is used to acquire real-time warehouse data of the target warehouse within the target time period during which the control of the transportation equipment is carried out based on the target wave division strategy, and which is of the same type as the target simulation result. The target simulation result is the simulation result of the simulation process corresponding to the target wave division strategy, and the duration of the target time period is the same as the time consumed by running the target simulation process to generate the duration simulation result. The difference calculation module is used to determine the difference between each type of real-time warehouse data and the target simulation result of the same type. The above-mentioned update of simulation parameters according to the following update operation is performed when the difference between any type of real-time warehouse data and the target simulation result of the same type is greater than a preset threshold.

[0151] For example, simulation parameters include the average moving speed of the transport equipment, the average walking speed of the pickers, and the equipment failure rate of the transport equipment, wherein the equipment failure rate is determined based on the number of times the transport equipment interrupts the execution of assigned transport tasks.

[0152] For example, the warehouse management device further includes a random seed generation module, and the simulation execution module 730 includes a first execution submodule. The random seed generation module is used to determine a hash value for each simulation process using a hash algorithm, based on the current timestamp and the identifier of the candidate wavelet partitioning strategy corresponding to each simulation process, to serve as the random seed for that simulation process. The first execution submodule is used to asynchronously run each simulation process based on the process isolation domain, simulation timeline, and random seed allocated to each simulation process.

[0153] For example, each simulation result includes at least one type of simulation data, and the simulation data of different simulation results are of the same type. The transportation control module 740 includes a normalization submodule, an evaluation submodule, and a strategy selection submodule. The normalization submodule is used to determine the normalization result of each simulation data based on the maximum and minimum values ​​of the simulation data of the same type as that simulation data. The evaluation submodule is used to determine the evaluation value of each simulation result based on the normalization result of all simulation data for that simulation result. The strategy selection submodule is used to determine the target wave division strategy from candidate wave division strategies based on the evaluation value of each simulation result.

[0154] For example, the simulation data includes the amount of goods leaving the warehouse within a preset unit of time, the average walking speed of the transportation equipment, and the equipment utilization rate of the transportation equipment.

[0155] For example, warehouse information includes the distribution of goods and the distribution of workstations.

[0156] According to another aspect of the present invention, an electronic device is also provided. Figure 8A schematic block diagram of an electronic device according to an embodiment of the present invention is shown. Figure 8 As shown, the electronic device includes a processor and a memory, wherein the memory stores computer program instructions, which are executed by the processor to perform the warehouse management method described above.

[0157] Furthermore, according to another aspect of the present invention, a storage medium is provided, on which program instructions are stored. When the program instructions are executed by a computer or processor, the computer or processor performs corresponding steps of the warehouse management method described in the embodiments of the present invention, and is used to implement corresponding modules in the warehouse management apparatus described in the embodiments of the present invention. The storage medium may, for example, include a storage component of a tablet computer, a hard disk of a personal computer, a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a portable compact disc read-only memory (CD-ROM), a USB memory, or any combination of the above storage media. The computer-readable storage medium may be any combination of one or more computer-readable storage media.

[0158] According to another aspect of the present invention, a computer program product is also provided, including computer program instructions, which, when executed, are used to perform the warehouse management method described above.

[0159] Those skilled in the art can understand the specific implementation and beneficial effects of the above-described warehouse management device, electronic equipment, storage medium, and computer program product by reading the detailed description of the warehouse management method above, and for the sake of brevity, they will not be described in detail here.

[0160] Although exemplary embodiments have been described herein with reference to the accompanying drawings, it should be understood that the above exemplary embodiments are merely illustrative and are not intended to limit the scope of the invention thereto. Various changes and modifications can be made therein by those skilled in the art without departing from the scope and spirit of the invention. All such changes and modifications are intended to be included within the scope of the invention as claimed in the appended claims.

[0161] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0162] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed.

[0163] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0164] Similarly, it should be understood that, in order to streamline the invention and aid in understanding one or more of the various aspects of the invention, features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof in the description of exemplary embodiments of the invention. However, this approach should not be construed as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as reflected in the corresponding claims, its inventive point lies in solving the corresponding technical problem with fewer features than all of those in a single disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of the invention.

[0165] Those skilled in the art will understand that, apart from the mutual exclusion of features, all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or elements of any method or apparatus so disclosed may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.

[0166] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features but not others included in other embodiments, combinations of features from different embodiments are intended to be within the scope of the invention and form different embodiments. For example, in the claims, any of the claimed embodiments can be used in any combination.

[0167] The various component embodiments of the present invention can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some modules in a warehouse management apparatus according to embodiments of the present invention. The present invention can also be implemented as an apparatus program (e.g., a computer program and computer program product) for performing some or all of the methods described herein. Such programs implementing the present invention can be stored on a computer-readable medium or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

[0168] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names.

[0169] The above description is merely a specific embodiment of the present invention or an explanation of that embodiment. The scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. The scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A warehouse management method, characterized in that, The method includes: Based on current order information, target warehouse storage information, and transportation equipment location information, different candidate wavelet division strategies are determined; The simulation process corresponding to each candidate wave partitioning strategy is determined, and an independent process isolation domain and an independent simulation time axis are allocated to the simulation process corresponding to each candidate wave partitioning strategy. The process isolation domain includes a virtual memory space. Based on the process isolation domain and simulation time axis allocated to each simulation process, each simulation process is run asynchronously to obtain the simulation results of each simulation process; Based on the simulation results, a target wave division strategy is determined from the candidate wave division strategies, and the transportation equipment is controlled to perform transportation based on the target wave division strategy.

2. The method according to claim 1, characterized in that, The simulation process for determining the corresponding strategies for each candidate wavelet division includes: Based on the simulation parameters corresponding to each candidate wavelet partitioning strategy, generate the simulation process corresponding to each candidate wavelet partitioning strategy. The method further includes updating the simulation parameters according to the following update operations: Based on historical data of the same type as the target simulation parameters within a preset time period before running the target simulation process, a prior distribution is determined, wherein the target simulation parameters are the simulation parameters corresponding to the target wave division strategy, and the target simulation process is the simulation process corresponding to the target wave division strategy; Based on the target simulation parameters, the prior distribution is processed according to a preset Bayesian update rule to generate the posterior distribution corresponding to the prior distribution; The target simulation parameters are updated using the posterior distribution.

3. The method according to claim 2, characterized in that, The preset Bayesian update rule includes an update rule based on a conjugate prior distribution, wherein the type of the prior distribution and the likelihood function type of the target simulation parameter constitute a conjugate distribution pair. The step of processing the prior distribution based on the target simulation parameters and a preset Bayesian update rule to generate the posterior distribution corresponding to the prior distribution includes: The statistics of the likelihood function are determined based on the target simulation parameters; The posterior distribution is determined by an objective analytical function based on the distribution parameters of the prior distribution and the statistic, wherein the objective analytical function is a function predetermined by deriving the conjugate distribution pair based on Bayes' theorem.

4. The method according to claim 2, characterized in that, The method further includes: Acquire real-time warehouse data of the target warehouse within the target time period during which the transportation equipment is controlled to transport based on the target wave division strategy, and the data is of the same type as the target simulation result. The target simulation result is the simulation result of the simulation process corresponding to the target wave division strategy, and the duration of the target time period is the same as the time consumed by running the target simulation process to generate the duration simulation result. For each type of real-time warehouse data, the difference between that type of real-time warehouse data and the target simulation result of the same type is determined, wherein the above-mentioned update of the simulation parameters according to the following update operation is performed when the difference between any type of real-time warehouse data and the target simulation result of the same type is greater than a preset threshold.

5. The method according to claim 2, characterized in that, The simulation parameters include the average moving speed of the transport equipment, the average walking speed of the pickers, and the equipment failure rate of the transport equipment, wherein the equipment failure rate is determined based on the number of times the transport equipment interrupts the execution of assigned transport tasks.

6. The method according to claim 1, characterized in that, The method further includes: For each simulation process, a hash value is determined by a hash algorithm based on the current timestamp and the identifier of the candidate wave partitioning strategy corresponding to each simulation process, and is used as the random seed for that simulation process. The asynchronous execution of each simulation process, based on a process isolation domain and simulation time axis allocated to each simulation process, includes: Each simulation process is run asynchronously, based on a process isolation domain, simulation timeline, and random seed allocated to each simulation process.

7. The method according to claim 1, characterized in that, Each simulation result includes at least one type of simulation data, and the simulation data types of different simulation results are the same; The step of determining the target wavelet partitioning strategy from the candidate wavelet partitioning strategies based on the simulation results includes: For each simulation data, the normalization result of the simulation data is determined based on the maximum and minimum values ​​of the simulation data of the same data type as the simulation data in each simulation result; For each simulation result, the evaluation value of the simulation result is determined based on the normalized result of all simulation data for that simulation result. Based on the evaluation value of each simulation result, the target wavelet partitioning strategy is determined from the candidate wavelet partitioning strategies.

8. The method according to claim 7, characterized in that, The simulation data includes the outbound volume of goods within a preset unit of time, the average walking speed of the transportation equipment, and the equipment utilization rate of the transportation equipment.

9. The method according to claim 1, characterized in that, The warehousing information includes the distribution of goods and the distribution of workstations.

10. A warehouse management device, characterized in that, include: The strategy generation module is used to determine different candidate wavelet division strategies based on current order information, target warehouse storage information, and transportation equipment location information; The process generation module is used to determine the simulation process corresponding to each candidate wave partitioning strategy, and to allocate an independent process isolation domain and an independent simulation time axis to the simulation process corresponding to each candidate wave partitioning strategy, wherein the process isolation domain includes a virtual memory space. The simulation execution module is used to asynchronously run each simulation process based on the process isolation domain and simulation time axis allocated to each simulation process, so as to obtain the simulation results of each simulation process. The transportation control module is used to determine the target wave division strategy from the candidate wave division strategies based on the simulation results, and control the transportation equipment to carry out transportation based on the target wave division strategy.

11. An electronic device comprising a processor and a memory, characterized in that, The memory stores computer program instructions, which, when executed by the processor, are used to perform the warehouse management method as described in any one of claims 1 to 9.

12. A storage medium on which program instructions are stored, characterized in that, The program instructions are used to execute the warehouse management method as described in any one of claims 1 to 9 when the program is run.

13. A computer program product comprising computer program instructions, characterized in that, The computer program instructions, when executed, are used to perform the warehouse management method as described in any one of claims 1 to 9.

Images