WO2024067358A1

WO2024067358A1 - Efficiency analysis method and system for warehouse management system, and computer device

Info

Publication number: WO2024067358A1
Application number: PCT/CN2023/120460
Authority: WO
Inventors: 彭仔华
Original assignee: 深圳市库宝软件有限公司
Priority date: 2022-09-30
Filing date: 2023-09-21
Publication date: 2024-04-04
Also published as: CN115689334A

Abstract

Disclosed in the present invention are an efficiency analysis method and system for a warehouse management system, and a storage medium and a computer device. The method comprises: acquiring operation information of a warehouse management system in a time period to be subjected to analysis; acquiring a plurality of pieces of feature information by processing the operation information, and generating a feature set of the warehouse management system; sending the feature set into a pre-stored efficiency analysis model, and acquiring, according to an output of the efficiency analysis model, a target efficiency analysis result of the warehouse management system for said time period, wherein the efficiency analysis result comprises: a storage efficiency problem category. By implementing the technical solution of the present invention, the labor cost is saved on, the results are also more scientific and evidence-based, and the accuracy and convenience of efficiency analysis of a warehouse management system are improved.

Description

Efficiency analysis method, system and computer equipment for warehouse management system

This application claims priority to the Chinese patent application filed with the China Patent Office on September 30, 2022, with application number 202211230481.6 and application name “Efficiency Analysis Method, System and Computer Equipment for Warehouse Management System”, the entire contents of which are incorporated by reference in this application.

Technical Field

The present invention relates to the technical field of computer information processing, and in particular to a warehouse management system efficiency analysis method, system, storage medium and computer equipment.

Background technique

In the warehouse management system, using robots to manage the warehouse and realize the automated processes of picking, inventory, and storage is an innovative approach. Currently, warehouse robot services are still in an emerging track and have great development potential and prospects.

When serving customers, warehouse robot products often face the situation that the statistical efficiency of the warehouse management system cannot reach the promised value. If this problem is not solved, it will lead to a series of serious consequences such as loss of customer trust, breach of contract, and impact on project acceptance and payment. However, the location of such problems is often very complicated, and it is necessary to analyze multiple complex factors such as on-site conditions, software systems, and robot-related algorithms to discover the problem and find the efficiency bottleneck.

At present, when analyzing the statistical efficiency of warehouse management systems, the main approach is to rely on the experience and feedback of on-site personnel. This method requires personnel to be stationed on-site and follow up on the project for a long time. It is not only time-consuming and labor-intensive, but also extremely costly. It also requires a high level of experience and judgment from personnel. The problems identified by different people may be completely inconsistent, with large errors. In the later stage, professional data analysts are needed to collect data based on this. Performing customized analysis to confirm and verify further increases labor costs.

Summary of the invention

The technical problem to be solved by the present invention is to provide an efficiency analysis method, system, storage medium and computer equipment for a warehouse management system in view of the defects of the prior art such as being time-consuming, labor-intensive, costly and having large errors.

The technical solution adopted by the present invention to solve the technical problem is: constructing an efficiency analysis method of a warehouse management system, comprising:

Information acquisition step: acquiring the operation information of the warehouse management system during the period to be analyzed;

Information processing step: obtaining a plurality of feature information by processing the operation information, and generating a feature set of the warehouse management system;

Information analysis step: sending the feature set to a pre-stored efficiency analysis model, and obtaining the target efficiency analysis result of the warehouse management system for the time period to be analyzed based on the output of the efficiency analysis model, wherein the efficiency analysis result includes: warehouse efficiency problem category.

Preferably, the efficiency analysis model is obtained according to the following steps:

Step S01, obtaining the operation information of the warehouse management system in the historical period;

Step S02, obtaining a plurality of feature information by processing the operation information, and generating a training set of the warehouse management system;

Step S03, sending the training set to the determined training model, and training the training model until the training model meets the preset qualification conditions, and taking the training model that meets the qualification conditions as the efficiency analysis model, and storing it.

Preferably, the step S03 includes:

Step S031, receiving a question category label set by a user;

Step S032, sending the training set to the current training model, and obtaining the reference efficiency analysis result corresponding to the training set according to the output of the training model;

Step S033, judging whether the current training model meets the preset qualification conditions according to the problem category label and the reference efficiency analysis result, if yes, executing step S034; if no, executing step S035;

Step S034, using the current training model as the efficiency analysis model and storing it;

Step S035: Receive parameter modification information from the user to modify the parameters of the current training model, and update the current training model according to the parameter modification information, and then execute step S032.

Preferably, the efficiency analysis result also includes the probability of storage efficiency problems;

The step S031 includes: receiving a question category label and a question probability label set by a user;

The step S033 includes: judging whether the current training model meets the preset qualification conditions according to the problem category label, the problem probability label and the reference efficiency analysis result.

Preferably, the information acquisition step includes:

Acquire first operation information of the warehouse management system in the time period to be analyzed, and perform data cleaning processing on the first operation information to obtain second operation information;

The information processing step comprises:

A plurality of feature information is obtained by processing the second operation information, and a feature set of the warehouse management system is generated.

Preferably, the obtaining of a plurality of characteristic information by processing the operation information includes:

Filtering specific data objects from the operation information, and calculating a plurality of different efficiency indicators of the warehouse management system based on the filtered data objects;

Aggregating and concatenating the plurality of different efficiency indicators to generate a summary concatenation table;

The data in the summary connection table is formatted and a plurality of feature information is obtained.

Preferably, the format conversion of the data in the summary connection table and obtaining multiple feature information includes:

Standardize and normalize the data in the summary connection table;

One-hot encoding is used to extract feature information from standardized and normalized data.

The present invention also constructs a storage medium storing a computer program, which implements the steps of the above-mentioned warehouse management system efficiency analysis method when executed by a processor.

The present invention also constructs a computer device, including a processor and a memory storing a computer program, wherein the processor implements the steps of the above-mentioned warehouse management system efficiency analysis method when executing the computer program.

The present invention also constructs an efficiency analysis system for a warehouse management system, comprising:

An information acquisition module, used to acquire the operation information of the warehouse management system during the period to be analyzed;

An information processing module, used for acquiring a plurality of feature information by processing the operation information, and generating a feature set of the warehouse management system;

The information analysis module is used to send the feature set to a pre-stored efficiency analysis model, and obtain the target efficiency analysis result of the warehouse management system for the time period to be analyzed based on the output of the efficiency analysis model, wherein the efficiency analysis result includes: warehouse efficiency problem category.

By implementing the technical solution of the present invention, the operation information of the warehouse management system in the period to be analyzed is first obtained, and then the operation information is processed to obtain multiple feature information. Finally, combined with the pre-established efficiency analysis model, the storage efficiency problem category of the warehouse management system in the period to be analyzed can be automatically identified. Compared with the existing method that relies on manual judgment, this technical solution not only saves labor costs, but also has more scientific and well-documented results, thereby improving the accuracy and convenience of warehouse management system efficiency analysis.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:

FIG1 is a flow chart of a first embodiment of an efficiency analysis method for a warehouse management system of the present invention;

FIG2 is a flow chart of a method for obtaining an efficiency analysis model according to a first embodiment of the present invention;

FIG3 is a flow chart of an embodiment 1 of step S03 in FIG2 ;

FIG4 is a block diagram of a computer device according to a first embodiment of the present invention;

FIG. 5 is a logical structure diagram of the first embodiment of the efficiency analysis system of the warehouse management system of the present invention.

Detailed ways

The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

FIG1 is a flow chart of a first embodiment of an efficiency analysis method for a warehouse management system of the present invention. The efficiency analysis method of this embodiment includes the following steps:

Information acquisition step S10: acquiring the operation information of the warehouse management system during the period to be analyzed;

In this step, the time period to be analyzed may be the current time period, such as the current day, the current week, the current month, or a previous time period, such as the previous day, the previous week, etc. The warehouse management system may be an intelligent warehouse management system (IWMS), an equipment scheduling system (ESS), etc., which are not limited here. The operation information of the time period to be analyzed is obtained from the warehouse management system. The operation information includes: execution instructions, status information, box moving information, order information, picking completion information, etc. This information may be stored in a table, such as: outbound order information table, outbound picking task table, box status record table, robot execution task table, etc. Service table, box arrival and departure console information table, etc., are not limited here.

The types of these operation information can be various, such as structured data, log files, etc. Among them, structured data can be stored in the database through IWMS and ESS, and can be directly read from the database when used; the data in the log file needs to be parsed before it can be stored in the database for use. The parsing method can use keywords and log format templates to extract data.

Information processing step S20: acquiring a plurality of feature information by processing the operation information, and generating a feature set of the warehouse management system;

In this step, the obtained operation information cannot be directly fed into the efficiency analysis model, and not all data are meaningful. Therefore, it is necessary to process the operation information first to obtain feature information, which can be an indicator with obvious value for judging the category of warehouse efficiency problems. These indicators are some high-dimensional aggregated data and can measure the overall operation status of the warehouse management system, such as but not limited to the average number of boxes moved per hour by the warehouse robot, the average walking mileage of the warehouse robot per task, etc. In one implementation, the feature information can be obtained by grouping and aggregating multiple tables in the operation information according to different dimensions (for example, robot ID, task ID, box ID, time).

Information analysis step S30: sending the feature set to a pre-stored efficiency analysis model, and obtaining the target efficiency analysis result of the warehouse management system for the time period to be analyzed according to the output of the efficiency analysis model, wherein the efficiency analysis result includes: warehouse efficiency problem category.

In this step, the efficiency analysis model is pre-established and stored. When it is officially used, the obtained feature information is sent to the efficiency analysis model, and the efficiency analysis model can output the target efficiency analysis results of the warehouse management system. The efficiency analysis results include warehouse efficiency problem categories, and the warehouse efficiency problem categories may be, for example: order structure problems, robot task allocation problems, robot path planning problems, robot congestion problems, operating table task allocation problems, manual picking problems, etc.

In the technical solution of this embodiment, firstly, the operation status of the warehouse management system in the period to be analyzed is obtained. The information is then processed to obtain multiple feature information. Finally, combined with the pre-established efficiency analysis model, the category of storage efficiency problems of the warehouse management system in the period to be analyzed can be automatically identified. Compared with the existing method that relies on manual judgment, this technical solution not only saves labor costs, but also has more scientific and well-documented results, improving the accuracy and convenience of efficiency analysis of the warehouse management system.

Furthermore, in an optional embodiment, as shown in FIG2 , in the model training stage, the efficiency analysis model may be obtained according to the following steps:

In this step, the historical period can be the past day, week, etc. The operation information of the historical period also comes from warehouse management systems such as IWMS and ESS, and the operation information includes: execution instructions, status information, box moving information, order information, picking completion information, etc. This information may also be stored in tables, such as: outbound order information table, outbound picking task table, box status record table, robot execution task table, and box arrival and departure operation table information table.

In this step, the determined training model can be a pre-set one or one selected by the user from multiple models. For example, the user can select the training model through the machine learning library Scikit-Learn. Specifically, the clustering model kmeans or DBSCAN can be freely selected by rewriting the configuration. Among them, kmeans has the advantages of being widely used, easy to implement, and highly efficient; DBSCAN has the advantages of being able to discover clusters of any shape and identify noise points.

Further, in an optional embodiment, as shown in FIG3 , step S03 specifically includes:

Step S031, receiving a question category label set by a user;

In this step, the user can determine the indicators that do not meet the specifications by checking the training set obtained in step S02, and then determine the corresponding problem category. For example, on a certain day in a project, the average distance traveled by the warehouse robot per trip to move boxes is longer (greater than the specification value), and then it can be determined that there is a problem with the path planning of the warehouse robot (no suitable path is selected for the warehouse robot). At this time, the corresponding problem category can be labeled.

In this step, regarding the current training model, it should be noted that the initial training model is a pre-set or user-selected training model. As the training progresses, the parameters of the initial training model will be modified to obtain the latest training model.

In this step, whether the current training model meets the preset qualification conditions can be determined by judging whether the matching degree between the storage efficiency problem category in the reference efficiency analysis result and the problem category label is greater than a set value (for example, 80%).

In this step, if the matching degree between the reference efficiency analysis result output by the training model and the problem category label judged by humans in advance is not greater than the set value, the parameters of the training model can be adjusted and the model can be re-trained. Type training, repeat this process until the matching degree between the two is greater than the set value.

Furthermore, in an optional embodiment, the efficiency analysis result also includes the probability of warehousing efficiency problems, and step S031 includes: receiving problem category labels and problem probability labels set by the user; step S033 includes: judging whether the current training model meets the preset qualification conditions based on the problem category labels, the problem probability labels and the reference efficiency analysis results.

In this embodiment, the efficiency analysis model outputs not only the storage efficiency problem category but also the storage efficiency problem probability (accuracy). Accordingly, when the efficiency analysis model is trained, the user sets not only the problem category label but also the problem probability label. Moreover, when evaluating the model, it is necessary to determine whether the storage efficiency problem category and the storage efficiency problem probability in the reference efficiency analysis result match the problem category label and the problem probability label. If the matching degree between the storage efficiency problem category output by the training model and the problem category label judged by humans in advance is greater than the first set value, but the matching degree between the storage efficiency problem probability output by the training model and the problem probability label judged by humans in advance is not greater than the second set value, then it is still considered that the current training model does not meet the preset qualification conditions, and the parameters of the training model should continue to be adjusted and the model training should be re-performed. The above process is repeated until the matching degrees of the two items are both greater than the corresponding set values.

In a specific embodiment, if the efficiency analysis of the warehouse management system is performed once a day, then, for two consecutive days, based on the acquired operation data of the day, the automatically identified target efficiency analysis results are shown in Table 1 below:

Table 1

Further, in an optional embodiment, the information acquisition step S10 includes: acquiring first operating information of the warehouse management system in the time period to be analyzed, and performing data cleaning processing on the first operating information to obtain second operating information. Correspondingly, the information processing step includes: acquiring multiple feature information by processing the second operating information, and generating a feature set of the warehouse management system. It should be understood that in step S01, after acquiring the operating information of the warehouse management system in the historical period, the operating information is also subjected to data cleaning processing to obtain the operating information after data cleaning processing, and, in step S02, the operating information after data cleaning processing is processed to obtain multiple feature information. In this embodiment, whether it is when performing efficiency analysis or model training, after obtaining the operating information, it is necessary to first perform conventional data cleaning work on these data, for example, including: missing value processing, outlier processing, etc.

Furthermore, in an optional embodiment, in the information processing step S20 and step S02, a plurality of feature information is obtained by processing the operation information, specifically including:

Specific data objects are screened out from the operation information, and multiple different efficiency indicators of the warehouse management system are calculated based on the screened data objects. In this step, since not all data in the operation information is meaningful, it is necessary to first screen out specific data objects from the operation information, and the specific data objects are data with obvious value for classifying warehouse efficiency issues. Then, multiple different efficiency indicators of the warehouse management system are obtained by calculating the screened data objects, for example, including: the average number of boxes moved per hour by the robot, the average walking mileage per task by the robot, etc.

The multiple different efficiency indicators are aggregated and joined to generate an aggregated joined table. For example, a wide table may be generated by combining different data sources.

The data in the summary connection table is formatted and multiple feature information is obtained. In this step, due to the inconsistency of the units and types of some data in the summary connection table, the efficiency analysis model/training The model cannot recognize it, so the data needs to be converted into a format so that the efficiency analysis model/training model can recognize it.

Furthermore, in an optional embodiment, the format conversion may be performed in the following manner:

Standardize and normalize the data in the summary connection table;

In this embodiment, the data may be standardized first to unify the unit and type of the data, and then normalized to unify the unit and value range. Finally, one-hot encoding is used to extract feature information from the normalized data, for example, to convert some text data into machine-recognizable values.

The present invention also constructs a storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the labor efficiency analysis method of the warehouse management system described above are implemented.

The storage medium of the present invention can be a U disk, a mobile hard disk, a read-only memory (ROM), a magnetic disk or an optical disk, and other computer-readable storage media that can store program codes.

FIG4 is a structural block diagram of a computer device embodiment 1 of the present invention. The efficiency analysis system 400 of this embodiment may be a computer or a server, and the server may be an independent server or a server cluster composed of multiple servers.

4 , the computer device 400 includes a processor 402 , a memory, and a network interface 405 connected via a system bus 401 , wherein the memory may include a non-volatile storage medium 403 and an internal memory 404 .

The non-volatile storage medium 403 can store an operating system 4031 and a computer program 4032. The computer program 4032 includes program instructions, and when the program instructions are executed by the processor 402, the processor 402 can execute the steps of the labor efficiency analysis method of the warehouse management system.

The processor 402 is used to provide computing and control capabilities to support the operation of the entire computer device 400. It should be understood that in the embodiment of the present application, the processor 402 can be a central processing unit (CPU), and the processor 402 can also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

The internal memory 404 provides an environment for the operation of the computer program 4032 in the non-volatile storage medium 403. When the computer program 4032 is executed by the processor 402, the processor 402 can execute the steps of the efficiency analysis method of the warehouse management system.

The network interface 405 is used to perform network communication with other devices.

Those skilled in the art will appreciate that the structure shown in FIG. 4 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device 400 to which the solution of the present application is applied. The specific computer device 400 may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

FIG5 is a logical structure diagram of the first embodiment of the efficiency analysis system of the warehouse management system of the present invention. The system of this embodiment includes: an information acquisition module 10, an information processing module 20, and an information analysis module 30, wherein the information acquisition module 10 is used to obtain the operation information of the warehouse management system in the time period to be analyzed; the information processing module 20 is used to obtain multiple feature information by processing the operation information, and generate a feature set of the warehouse management system; the information analysis module 30 is used to send the feature set to a pre-stored efficiency analysis model, and obtain the target efficiency analysis result of the warehouse management system for the time period to be analyzed according to the output of the efficiency analysis model, wherein the efficiency analysis result includes: Warehousing efficiency problem category.

Furthermore, in an optional embodiment, the efficiency analysis system of the warehouse management system of the present invention also includes a model training module, and when performing model training, the information acquisition module 10 is also used to obtain the operation information of the warehouse management system in a historical period; the information processing module 20 is also used to obtain multiple feature information by processing the operation information, and generate a training set for the warehouse management system; the model training module is used to send the training set to the determined training model, and train the training model until the training model meets the preset qualification conditions, and the training model that meets the qualification conditions is used as the efficiency analysis model, and stored.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention shall be included in the scope of the claims of the present invention.

Claims

An efficiency analysis method for a warehouse management system, characterized by comprising:

Information acquisition step: acquiring the operation information of the warehouse management system during the period to be analyzed;

Information processing step: obtaining a plurality of feature information by processing the operation information, and generating a feature set of the warehouse management system;

Information analysis step: sending the feature set to a pre-stored efficiency analysis model, and obtaining the target efficiency analysis result of the warehouse management system for the time period to be analyzed based on the output of the efficiency analysis model, wherein the efficiency analysis result includes: warehouse efficiency problem category.
The efficiency analysis method of a warehouse management system according to claim 1 is characterized in that the efficiency analysis model is obtained according to the following steps:

Step S01, obtaining the operation information of the warehouse management system in the historical period;

Step S02, obtaining a plurality of feature information by processing the operation information, and generating a training set of the warehouse management system;

Step S03, sending the training set to the determined training model, and training the training model until the training model meets the preset qualification conditions, and taking the training model that meets the qualification conditions as the efficiency analysis model, and storing it.
The efficiency analysis method of a warehouse management system according to claim 2, characterized in that step S03 comprises:

Step S031, receiving a question category label set by a user;

Step S032, sending the training set to the current training model, and obtaining the reference efficiency analysis result corresponding to the training set according to the output of the training model;

Step S033, judging the current Whether the training model meets the preset qualification conditions, if yes, execute step S034; if no, execute step S035;

Step S034, using the current training model as the efficiency analysis model and storing it;

Step S035: Receive parameter modification information from the user to modify the parameters of the current training model, and update the current training model according to the parameter modification information, and then execute step S032.
The efficiency analysis method of a warehouse management system according to claim 3, characterized in that the efficiency analysis result also includes the probability of storage efficiency problems;

The step S031 includes: receiving a question category label and a question probability label set by a user;

The step S033 includes: judging whether the current training model meets the preset qualification conditions according to the problem category label, the problem probability label and the reference efficiency analysis result.
The efficiency analysis method of a warehouse management system according to any one of claims 1 to 4, characterized in that the information acquisition step comprises:

Acquire first operation information of the warehouse management system in the time period to be analyzed, and perform data cleaning processing on the first operation information to obtain second operation information;

The information processing step comprises:

A plurality of feature information is obtained by processing the second operation information, and a feature set of the warehouse management system is generated.
The efficiency analysis method of a warehouse management system according to any one of claims 1 to 4, characterized in that the processing of the operation information to obtain a plurality of feature information comprises:

Filtering specific data objects from the operation information, and calculating a plurality of different efficiency indicators of the warehouse management system based on the filtered data objects;

Aggregating and concatenating the plurality of different efficiency indicators to generate a summary concatenation table;

The data in the summary connection table is formatted and a plurality of feature information is obtained.
The efficiency analysis method of a warehouse management system according to claim 6 is characterized in that the format conversion of the data in the summary connection table and the acquisition of multiple feature information include:

Standardize and normalize the data in the summary connection table;

One-hot encoding is used to extract feature information from standardized and normalized data.
A storage medium storing a computer program, characterized in that when the computer program is executed by a processor, the steps of the efficiency analysis method of a warehouse management system described in any one of claims 1 to 7 are implemented.
A computer device comprises a processor and a memory storing a computer program, wherein the processor implements the steps of the efficiency analysis method of a warehouse management system as described in any one of claims 1 to 7 when executing the computer program.
An efficiency analysis system for a warehouse management system, characterized by comprising:

An information acquisition module, used to acquire the operation information of the warehouse management system during the period to be analyzed;

An information processing module, used for acquiring a plurality of feature information by processing the operation information, and generating a feature set of the warehouse management system;

The information analysis module is used to send the feature set to a pre-stored efficiency analysis model, and obtain the target efficiency analysis result of the warehouse management system for the time period to be analyzed based on the output of the efficiency analysis model, wherein the efficiency analysis result includes: warehouse efficiency problem category.