Picking path optimization method and system

Abstract

The invention relates to the technical field of optimized path planning, in particular to a picking path optimization method and system. The method comprises the following steps: obtaining storage warehouse data, goods storage data and picker basic information, and extracting working time according to the picker basic information so as to obtain picker working time data; warehouse goods storage generation is performed according to the storage warehouse data and the goods storage data so as to generate warehouse goods storage data, and picker goods storage demand processing is performed according to the goods storage data and picker basic information so as to generate picker goods storage demand data; carrying out order picker demand correction processing on the warehouse goods storage data by utilizing the order picker goods storage demand data so as to generate goods storage space distribution change data; and obtaining future extraction demand data of the goods. By optimizing the order picking path, storage waste and repeated actions are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of optimized path planning, and particularly to a picking path optimization method and system. Background Art

[0002] The picking path refers to the path that a picker passes through when selecting goods from different storage locations or shelves according to an order or demand in a certain route or sequence in a warehouse or logistics center. The main objective of the picking path optimization method is to design an optimal picking path so that the picker picks goods in the warehouse according to the optimal route sequence, thereby improving the picking efficiency and accuracy and reducing the picker's travel time and cost. In the current practice of picking path optimization methods, they often simply consider the warehouse layout and goods storage to provide distribution optimization of goods storage or simple path optimization, without considering the situation of pickers, which leads to overloaded work caused by unreasonable arrangements and low efficiency. Summary of the Invention

[0003] In order to solve the above technical problems, the present invention proposes a picking path optimization method and system to solve at least one of the above technical problems.

[0004] The present application provides a picking path optimization method, including the following steps:

[0005] Step S1: Obtain storage warehouse data, goods storage data, and picker basic information, and extract working hours according to the picker basic information to obtain picker working hour data;

[0006] Step S2: Generate warehouse goods storage data according to the storage warehouse data and goods storage data, and process the picker's goods storage requirements according to the goods storage data and picker basic information to generate picker goods storage requirement data;

[0007] Step S3: Use the picker goods storage requirement data to perform picker requirement correction processing on the warehouse goods storage data to generate goods storage space distribution change data;

[0008] Step S4: Obtain future goods extraction requirement data, and generate picking path data according to the future goods extraction requirement data and goods storage space distribution change data;

[0009] Step S5: Optimize the picking path data using the picker working hour data to generate optimized picking path data.

[0010] The present invention comprehensively considers multiple factors such as storage warehouse data, goods storage data, basic information of pickers, and future demand data for goods retrieval. Through comprehensive analysis and processing of these factors, the picking path can be more comprehensively optimized, and the picking efficiency can be improved. The working time data of pickers is extracted based on the basic information of pickers, and the working time limit of pickers is considered in the process of optimizing the picking path. This can reasonably arrange the workload of pickers, avoid overwork or working time conflicts, and improve work efficiency and the job satisfaction of pickers. The warehouse goods storage data is corrected according to the goods storage demand data of pickers, and the changed data of the goods storage space distribution that better meets the needs of pickers is generated. This helps to optimize the picking path, reduce the moving distance and time of pickers, and improve the picking efficiency. The future demand data for goods retrieval is obtained and incorporated into the process of generating the picking path. By considering future demand situations, the picking path can be planned more flexibly, goods can be prepared in advance, and the waiting time and picking error rate can be reduced.

[0011] Preferably, the picker working time data includes high physical strength working time data and general working time data, and step S1 is specifically as follows:

[0012] Step S11: Obtain the storage warehouse data, goods storage data, and basic information of pickers;

[0013] Step S12: Classify the basic information of pickers by using the picker status working time model to obtain the basic working time data of pickers;

[0014] Step S13: Control the camera to collect the picker image data to obtain the picker image data;

[0015] Step S14: Extract the picker status from the picker image data to obtain the physical strength status data of pickers;

[0016] Step S15: Process the basic working time data of pickers according to the physical strength status data of pickers to obtain the picker working time data.

[0017] In the present invention, through step S12 and step S14, by using the picker status working time model and the picker image data extraction technology, the present invention can objectively and accurately obtain the working time data of pickers. By processing the basic information and image data of pickers, the working time of pickers can be subdivided into high-physical-energy working time and general working time, which can more accurately reflect the working status of pickers. Through step S15, the present invention uses the picker physical status data to perform picker status processing on the basic working time data of pickers, thereby obtaining the picker working time data. In this way, according to the physical status of pickers, the working time of pickers can be reasonably allocated, so that pickers with high physical energy work during peak hours and pickers with general physical energy work during off-peak hours, in order to maximize the working efficiency of pickers. By subdividing the working time of pickers and allocating the working time based on the physical status of pickers, the present invention can more reasonably arrange the work tasks and time of pickers and improve the picking efficiency.

[0018] Preferably, the specific steps for constructing the picker status working time model in step S12 are as follows:

[0019] Step S121: Obtain the basic data of standard pickers, where the basic data of standard pickers includes standard picker identity data, standard picker skill data, and standard picker work record data;

[0020] Step S122: Perform data preprocessing and feature extraction based on the basic data of standard pickers, so as to obtain picker picking intensity duration feature data, picker picking intensity feature data, and picker physiological condition feature data;

[0021] Step S123: Construct a deep model based on the picker picking intensity duration feature data, picker picking intensity feature data, and picker physiological condition feature data, thereby constructing a picker status working time model.

[0022] In the present invention, by obtaining the basic data of standard pickers, including the identity data of standard pickers, the skill data of standard pickers, and the work record data of standard pickers, factors such as the identity characteristics, skill levels, and work experience of pickers can be comprehensively considered, enabling the model to more comprehensively understand the background information of pickers. According to the basic data of standard pickers, data preprocessing and feature extraction are carried out, and the picker's picking intensity duration feature data, the picker's picking intensity feature data, and the picker's physiological condition feature data are extracted. These feature data can reflect the time distribution, work intensity, and physiological condition of pickers during the work process, providing valuable inputs for subsequent model construction. Based on the picker's picking intensity duration feature data, the picker's picking intensity feature data, and the picker's physiological condition feature data, a deep learning model is used to construct a picker status working time model. The deep learning model has powerful learning and expression capabilities and can learn the complex relationship between the picker's status and working time from a large amount of data, thereby improving the prediction accuracy and generalization ability of the model.

[0023] Preferably, step S14 is specifically as follows:

[0024] Step S141: Extract the eye image data from the picker image data to obtain the picker's eye image data;

[0025] Step S142: Extract features based on the picker's eye image data to obtain the picker's eyeball movement feature data, the picker's eyelid opening and closing feature data, the picker's blinking frequency feature data, and the picker's pupil feature data;

[0026] Step S143: Perform time series correlation based on the picker's eyelid opening and closing feature data and the picker's blinking frequency feature data to obtain the picker's eyelid movement feature data;

[0027] Step S144: Use the picker's eyeball movement feature data to perform biomimetic correction on the picker's eyelid movement feature data to generate the picker's eyeball state feature data;

[0028] Step S145: Use the picker's pupil feature data to perform state correction on the picker's eyeball state feature data to obtain the picker's precise eyeball state feature data;

[0029] Step S146: Perform picker physical fitness state recognition on the picker's precise eyeball state feature data to obtain the picker's physical fitness state data.

[0030] In the present invention, the eye image data of the order picker is extracted and its features are extracted to obtain the order picker's eyeball movement feature data, eyelid opening and closing feature data, blink frequency feature data, and pupil feature data. These feature data can reflect the movement of the order picker's eyes and provide useful information for subsequent state correction and physical state recognition. By performing temporal correlation on the eyelid opening and closing feature data and blink frequency feature data of the order picker, the eyelid movement feature data of the order picker can be obtained. These feature data can more comprehensively describe the movement state of the order picker's eyes and face, providing a more accurate basis for subsequent state correction. Using the order picker's eyeball movement feature data and pupil feature data, biomimetic correction and state correction are performed on the order picker's eyelid movement feature data, thereby obtaining the order picker's precise eyeball state feature data. This can eliminate the biological individual differences and environmental influences in eye movement and obtain more accurate order picker's eyeball state information. Based on the order picker's precise eyeball state feature data, the physical state of the order picker is recognized. By analyzing the correlation between the eyeball state and the physical state, the current physical state of the order picker, such as fatigue level, attention concentration level, etc., can be judged.

[0031] Preferably, the state correction is calculated through the order picker's precise eyeball state calculation formula, and the order picker's precise eyeball state calculation formula is specifically as follows:

[0032]

[0033] W is the order picker's precise eyeball state feature data, a0 is the first initial adjustment term, a1 is the first weight term, k is the original eyeball state data, a2 is the second weight term, v is the pupil movement data, a3 is the third weight term, m is the order picker's eyeball state feature data, o is the error correction term, b is the constant term, l is the order picker's eyeball state feature data adjustment term, c0 is the second initial adjustment term, c1 is the fourth weight term, d is the pupil shape data, and u is the correction term of the order picker's precise eyeball state feature data.

[0034] The present invention constructs a precise eye state calculation formula for order pickers. This formula fully considers the first initial adjustment term a0, the first weight term a1, the original eye state data k, the second weight term a2, the pupil movement data v, the third weight term a3, the order picker's eye state characteristic data m, the error correction term o, the constant term b, the adjustment term l for the order picker's eye state characteristic data, the second initial adjustment term c0, the fourth weight term c1, the pupil shape data d, and the interaction relationships among them. The original eye state data k represents the initial measurement value of the order picker's eye state. The pupil movement data v represents the movement characteristics of the order picker's pupil. The order picker's eye state characteristic data m includes eyelid opening and closing characteristic data, blinking frequency characteristic data, etc., which are used to describe the details and characteristics of the order picker's eye state. The constant term b is used to adjust the proportional relationship between the adjustment term and the eye state characteristic data. The adjustment term l for the order picker's eye state characteristic data is used to adjust the value range or magnitude of the eye state characteristic data. The pupil shape data d represents the shape characteristics of the order picker's pupil, providing information about the order picker's fatigue level or attention level. By comprehensively considering multiple parameters and characteristic data, as well as their weight relationships and correction terms, the eye state characteristic data of the order picker can be calculated more accurately, providing more accurate and reliable order picker eye state information. This helps to further analyze the physical condition, attention level, and working conditions of the order picker.

[0035] Preferably, step S2 is specifically as follows:

[0036] Step S21: Perform distribution matrix association based on the storage warehouse data and the goods storage data to obtain the warehouse goods storage data;

[0037] Step S22: Optimize the basic warehouse goods storage data using the goods value data and the goods demand data in the goods storage data to generate the warehouse goods storage data;

[0038] Step S23: Process the order picker's goods storage requirements based on the goods storage data and the order picker's basic information to generate the order picker's goods storage requirement data.

[0039] The present invention optimizes the storage layout and location of goods in a warehouse by analyzing the storage warehouse data and the goods storage data, combining the value and demand of the goods, improving the utilization rate of the storage space and the access efficiency of the goods. By considering the working ability and characteristics of the pickers, the goods storage requirements are matched with the pickers, the goods storage tasks are reasonably allocated, the redundancy and waste of the picking paths are reduced, and the working efficiency and accuracy of the pickers are improved. By optimizing the storage and picking processes of the warehouse goods, the supply chain requirements can be responded to more quickly and accurately. High-value or high-demand goods can be given priority treatment, the order processing time is shortened, and the flexibility and response ability of the supply chain are enhanced. By optimizing the picking paths and picking efficiency, the customer's order requirements can be met more quickly and accurately, the delivery speed and order accuracy are improved, and the customer satisfaction and loyalty are enhanced.

[0040] Preferably, step S3 is specifically as follows:

[0041] Step S31: Optimize the picker physical demand of the warehouse goods storage data by using the picker physical goods storage demand data in the picker goods storage demand data, so as to generate the physical goods distribution data;

[0042] Step S32: Optimize the goods value extraction demand of the warehouse goods storage data by using the picker goods value storage demand data in the picker goods storage demand data, so as to generate the goods value distribution data to be extracted;

[0043] Step S33: Generate the goods storage space distribution according to the physical goods distribution data and the goods value distribution data to be extracted, so as to generate the goods storage space distribution change data.

[0044] The present invention reasonably allocates the goods storage tasks by considering the physical condition and ability of the pickers, reduces the burden on the pickers with lower physical strength, improves the working efficiency and accuracy, and reduces the physical consumption and fatigue degree. By considering the value and importance of the goods, the storage locations of high-value or high-demand goods are preferentially arranged to ensure their easy access and extraction, and the order processing speed and customer satisfaction are improved. By reasonably planning the storage locations of the goods and the space utilization of the warehouse, the walking distance and picking time of the pickers are reduced, the picking paths are optimized, the picking efficiency is improved, and the errors and delays are reduced. Through the above optimization measures, the picking efficiency and accuracy are improved, the order processing time is shortened, the flexibility and response ability of the supply chain are enhanced, and the overall supply chain efficiency is promoted.

[0045] Preferably, step S4 is specifically as follows:

[0046] Obtain the historical goods extraction data, and use the historical goods extraction data for prediction calculation to obtain the future goods extraction demand data;

[0047] Generate picking path data based on future cargo extraction demand data and data on changes in the distribution of cargo storage space.

[0048] Through predictive calculations using historical cargo extraction data, the present invention can accurately predict future cargo extraction demands. According to the prediction results and changes in the cargo storage space, the optimal picking path is generated, reducing the walking distance and extraction time of pickers, and improving picking efficiency. Based on future cargo extraction demand data, the storage location and picking order of the cargo are reasonably planned, enabling high-demand or important cargo to be preferentially processed and extracted, improving the utilization efficiency of resources and the accessibility of the cargo. By predicting future cargo extraction demands, the picking path can be planned in advance, reducing ad-hoc adjustments and processing, lowering the error rate and omission rate during the picking process, and improving order accuracy and customer satisfaction. By analyzing historical cargo extraction data and predicting future demands, warehouse managers can obtain important information about cargo flow, thereby optimizing aspects such as the warehouse's storage strategy, personnel scheduling, and equipment configuration. This helps improve the efficiency and flexibility of warehouse operations, reducing costs and time waste.

[0049] Preferably, step S5 is specifically as follows:

[0050] Step S51: Perform first optimized path processing on the picking path data using high physical ability working time data, thereby generating high physical ability picking path data;

[0051] Step S52: Perform second optimized path processing on the picking path data using general working time data, thereby generating general physical ability picking path data;

[0052] Step S53: Perform spatio-temporal coupling screening and optimization on the high physical ability picking path data and the general physical ability picking path data, thereby generating optimized picking path data.

[0053] Through optimized processing for high physical ability working time data, the present invention enables the full utilization of the advantages of pickers in a high physical ability state, improving picking efficiency and work quality. By considering the general working time data of pickers, the task allocation and path planning are reasonably adjusted to avoid excessive fatigue of pickers and an increase in the error rate, improving the accuracy and reliability of the work. Through spatio-temporal coupling screening and optimization, according to the physical ability state and work ability of pickers at different time periods, the picking path and task allocation are dynamically adjusted, enabling pickers to efficiently complete work tasks at different time periods. It can fully consider the physical ability state, fatigue level, and work efficiency of pickers, achieve intelligent optimization of the picking path, thereby improving the work efficiency of the warehouse, reducing waste of human resources, and ensuring the timeliness and accuracy of cargo extraction.

[0054] This application provides a picking path optimization system, and the system includes:

[0055] At least one processor;

[0056] A memory communicatively connected to the at least one processor;

[0057] The memory stores a computer program executable by the at least one processor, and when the computer program is executed by the at least one processor, the at least one processor is enabled to execute a picking path optimization method as described above.

[0058] The beneficial effects of the present invention are as follows: By comprehensively considering the working hours of pickers, the storage requirements of goods, and the picking path data, the planning of the picking path and the task allocation are optimized. This can ensure that pickers carry out picking work during appropriate time periods, reduce task redundancy and unnecessary walking paths, thereby improving picking efficiency and accuracy. Through the optimization of the picking path, the work tasks and working hours of pickers can be reasonably allocated, avoiding waste of resources and excessive fatigue. This can reduce the cost of human resources and improve the operational efficiency of the warehouse. Through the generation of warehouse goods storage data and the processing of picker goods storage requirement data, the optimization and management of the goods storage space are realized. A reasonable goods storage distribution can reduce problems such as goods stacking and misalignment, improving the accessibility and extraction efficiency of goods. By obtaining future extraction requirement data of goods and combining it with data on changes in the distribution of the goods storage space, future demand prediction and planning can be carried out. This can adjust the picking path and task allocation in advance to adapt to future demand changes and ensure timely fulfillment of customer order requirements. By comprehensively considering the working hours of pickers, the storage requirements of goods, and future extraction requirements, the overall optimization of warehouse operations is achieved through the optimization of the picking path. The optimized picking path can make the best use of the working hours and capabilities of pickers, improving the overall operational efficiency and productivity of the warehouse. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Other features, objects, and advantages of the present application will become more apparent by reading the detailed description of the non-limiting embodiments with reference to the following drawings:

[0060] Figure 1 Shows a flowchart of the steps of a picking path optimization method according to an embodiment;

[0061] Figure 2 Shows a flowchart of the steps of step S1 according to an embodiment;

[0062] Figure 3 Shows a flowchart of the steps of a method for constructing a picker status working time model according to an embodiment;

[0063] Figure 4 Shows a flowchart of the steps of step S14 according to an embodiment;

[0064] Figure 5 Shows the step flow chart of step S2 of an embodiment;

[0065] Figure 6 Shows the step flow chart of step S3 of an embodiment;

[0066] Figure 7 Shows the step flow chart of step S5 of an embodiment. Detailed implementation manners

[0067] The technical method of the present invention for patent will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those skilled in the art within the scope of the present invention without creative work based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0068] In addition, the accompanying drawings are only schematic diagrams of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus their repeated description will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0069] It should be understood that although terms such as "first" and "second" may be used here to describe each unit, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used here includes any and all combinations of one or more of the listed related items.

[0070] Please refer to Figures 1 to 7 , this application provides a method for optimizing the picking path, including the following steps:

[0071] Step S1: Obtain the storage warehouse data, the goods storage data, and the basic information of the picker, and extract the working hours according to the basic information of the picker, so as to obtain the picker working hour data;

[0072] Specifically, for example, assume that the warehouse data includes information such as warehouse layout, number and location of shelves, the goods storage data includes information such as goods types, quantities, and storage locations, and the basic information of pickers includes information such as names, working hours, and ability levels. According to the basic information of pickers, the working time period of each picker can be extracted. For example, the working time of picker A is from 8 am to 5 pm.

[0073] Step S2: Generate warehouse goods storage based on the storage warehouse data and the goods storage data, thereby generating warehouse goods storage data, and process the goods storage requirements of pickers according to the goods storage data and the basic information of pickers, thereby generating picker goods storage requirement data;

[0074] Specifically, for example, according to the storage warehouse data and the goods storage data, the storage location and layout of goods in the warehouse can be determined. For example, food goods are stored in a specific shelf area, and electronic products are stored in another area. According to the basic information of pickers and the goods storage data, the goods storage requirements of each picker can be determined. For example, picker A is responsible for storing food goods.

[0075] Step S3: Use the picker goods storage requirement data to perform picker requirement correction processing on the warehouse goods storage data, thereby generating goods storage space distribution change data;

[0076] Specifically, for example, according to the picker goods storage requirement data, the warehouse goods storage data is corrected and adjusted to ensure that each picker can conveniently access and store the required goods. For example, according to the requirements of picker A, the food goods under his responsibility are stored on the shelves closer to his working area to reduce the walking path and improve work efficiency.

[0077] Step S4: Obtain the future goods extraction requirement data, and generate pick-up path data according to the future goods extraction requirement data and the goods storage space distribution change data;

[0078] Specifically, for example, predict the future goods extraction requirements based on historical data or market trends. For example, predict that the sales volume of a certain type of commodity will increase during a specific period. Combining the goods storage space distribution change data, determine a reasonable pick-up path to ensure the rapid and efficient extraction of the required goods. For example, in the case of predicting an increase in the sales volume of a certain type of commodity, store this type of commodity in a location close to the picking area to reduce the picking time.

[0079] Step S5: Optimize the pick-up path data using the picker working time data, thereby generating optimized pick-up path data.

[0080] Specifically, for example, based on the working time data of the order picker and combined with the order picking path data, the allocation of order picking tasks and path planning are optimized. For example, high-value or high-demand goods are arranged to be picked during the working time periods when the order picker has high ability and high energy, while low-value or low-demand goods are arranged in other time periods.

[0081] The present invention comprehensively considers multiple factors such as storage warehouse data, goods storage data, basic information of order pickers, and future extraction demand data of goods. Through comprehensive analysis and processing of these factors, the order picking path can be more comprehensively optimized and the order picking efficiency can be improved. The working time data of the order picker is extracted based on the basic information of the order picker, and the working time limit of the order picker is considered during the optimization process of the order picking path. This can reasonably arrange the workload of the order picker, avoid overloading or working time conflicts, and improve work efficiency and the job satisfaction of the order picker. The warehouse goods storage data is corrected according to the goods storage demand data of the order picker to generate changed data of the goods storage space distribution that is more in line with the needs of the order picker. This helps to optimize the order picking path, reduce the moving distance and time of the order picker, and improve the order picking efficiency. The future extraction demand data of goods is obtained and incorporated into the order picking path generation process. By considering future demand situations, the order picking path can be planned more flexibly, goods can be prepared in advance, and the waiting time and order picking error rate can be reduced.

[0082] Preferably, the working time data of the order picker includes high physical fitness working time data and general working time data, and step S1 is specifically as follows:

[0083] Step S11: Obtain the storage warehouse data, goods storage data, and basic information of the order picker;

[0084] Specifically, for example, collect storage warehouse data such as the layout diagram of the warehouse, shelf information, goods classification and quantity, etc.; collect goods storage data, including information such as the type, quantity, and storage location of goods; collect the basic information of the order picker, including name, work experience, physical condition, etc.

[0085] Step S12: Classify the basic information of the order picker by using the order picker status working time model to obtain the basic working time data of the order picker;

[0086] Specifically, for example, according to the basic information of the order picker, such as age, gender, physical condition, etc., combined with the order picker status working time model, classify the order picker. For example, classify the order picker into two categories: high physical fitness and general physical fitness to determine the basic working time of different categories of order pickers.

[0087] Step S13: Control the camera to collect the image data of the order picker to obtain the image data of the order picker;

[0088] Specifically, for example, install a camera in the picking area to collect image data of the picker through a monitoring system. The camera can capture the behavior and state of the picker, such as working posture, walking path, pupil movement, etc.

[0089] Step S14: Extract the picker's status from the picker's image data to obtain the picker's physical condition data;

[0090] Specifically, for example, use image processing and computer vision technologies to analyze and extract the picker's image data. For example, by identifying the picker's posture and actions, their physical condition can be judged, such as whether they are fatigued, work efficiency, etc.

[0091] Step S15: Use the picker's physical condition data to perform picker status processing on the picker's basic working time data, thereby obtaining the picker's working time data.

[0092] Specifically, for example, according to the picker's physical condition data, combined with the picker's basic working time data, perform picker status processing. For example, adjust the working time period of high-physical-condition pickers so that they work in a high-capability and high-energy state; keep the general working time period for pickers with general physical condition and moderately adjust the workload. This can give full play to the working potential and efficiency of the pickers.

[0093] In the present invention, through steps S12 and S14, the present invention uses the picker status working time model and the picker image data extraction technology to objectively and accurately obtain the picker's working time data. By processing the picker's basic information and image data, the picker's working time can be subdivided into high-physical-condition working time and general working time, more accurately reflecting the picker's working state. Through step S15, the present invention uses the picker's physical condition data to perform picker status processing on the picker's basic working time data, thereby obtaining the picker's working time data. In this way, according to the picker's physical condition, the working time of the pickers can be reasonably allocated, so that high-physical-condition pickers work during peak hours and general-physical-condition pickers work during off-peak hours, in order to maximize the working efficiency of the pickers. Through subdividing the picker's working time and allocating the working time based on the picker's physical condition, the present invention can more reasonably arrange the picker's work tasks and time, improving the picking efficiency.

[0094] Preferably, the specific steps for constructing the picker status working time model in step S12 are as follows:

[0095] Step S121: Obtain standard picker basic data, where the standard picker basic data includes standard picker identity data, standard picker skill data, and standard picker work record data;

[0096] Specifically, for example, collect the basic data of a group of standard pickers, including their identity information (such as age, gender, work experience, etc.), skill information (such as proficiency in operating equipment, working speed, etc.), and work record data (such as daily picking duration, number of items picked each time, etc.).

[0097] Step S122: Perform data preprocessing and feature extraction based on the basic data of standard pickers, so as to obtain picker picking intensity duration feature data, picker picking intensity feature data, and picker physiological condition feature data;

[0098] Specifically, for example, preprocess the basic data of standard pickers, including data cleaning, missing value processing, outlier processing, etc. Then, based on the cleaned data, extract the picker picking intensity duration feature data (such as daily picking duration), picker picking intensity feature data (such as number of items picked per hour), and picker physiological condition feature data (such as heart rate, body temperature, etc.).

[0099] Step S123: Build a deep model based on the picker picking intensity duration feature data, picker picking intensity feature data, and picker physiological condition feature data, so as to build a picker status working time model.

[0100] Specifically, for example, use deep learning technology to build a picker status working time model. Take the picker picking intensity duration feature data, picker picking intensity feature data, and picker physiological condition feature data as inputs, and design an appropriate neural network structure, including an input layer, a hidden layer, and an output layer. Through model training and optimization, enable it to predict the working time status of pickers based on their feature data, such as determining high physical fitness working time periods and general working time periods.

[0101] In the present invention, by obtaining the basic data of standard pickers, including the identity data of standard pickers, the skill data of standard pickers, and the work record data of standard pickers, factors such as the identity characteristics, skill levels, and work experience of pickers can be comprehensively considered, enabling the model to more comprehensively understand the background information of pickers. Based on the basic data of standard pickers, data preprocessing and feature extraction are carried out, and the picker picking intensity duration feature data, the picker picking intensity feature data, and the picker physiological condition feature data are extracted. These feature data can reflect the time distribution, work intensity, and physiological condition of pickers during the work process, providing valuable inputs for subsequent model construction. Based on the picker picking intensity duration feature data, the picker picking intensity feature data, and the picker physiological condition feature data, a deep learning model is used to construct a picker status working time model. The deep learning model has powerful learning and expression capabilities and can learn the complex relationship between the picker status and working time from a large amount of data, thereby improving the prediction accuracy and generalization ability of the model.

[0102] Preferably, step S14 is specifically as follows:

[0103] Step S141: Extract the eye image data from the picker image data to obtain the picker eye image data;

[0104] Specifically, for example, devices such as cameras or infrared sensors are used to monitor the picker in real time to obtain the picker eye image data. Or binary calculation is performed on the picker image data, and edge detection and graphics analysis are used to obtain the picker eye image data.

[0105] Step S142: Extract features from the picker eye image data to obtain the picker eyeball movement feature data, the picker eyelid opening and closing feature data, the picker blink frequency feature data, and the picker pupil feature data;

[0106] Specifically, for example, computer vision and image processing technologies are used to analyze and extract features from the picker eye image. For example, the picker eyeball movement feature data, the picker eyelid opening and closing feature data, the picker blink frequency feature data, and the picker pupil feature data are obtained by detecting the pupil size and position, blink frequency, and opening and closing degree of the eyelids.

[0107] Step S143: Perform time series association on the picker eyelid opening and closing feature data and the picker blink frequency feature data to obtain the picker eyelid movement feature data;

[0108] Specifically, for example, the picker eyelid opening and closing feature data and the picker blink frequency feature data are subjected to time series association. For example, the picker eyelid movement feature data is generated by time series analysis methods.

[0109] Step S144: Use the picker's eye movement characteristic data to perform biomimetic correction on the picker's eyelid movement characteristic data, so as to generate the picker's eye state characteristic data;

[0110] Specifically, for example, according to biological laws, use the picker's eye movement characteristic data to correct the picker's eyelid movement characteristic data. For example, locate the focus point of the eyes and match it with the requirements of the picking task, so as to generate the picker's eye state characteristic data.

[0111] Specifically, for example, adapt and correct the picker's eyelid movement characteristic data with the historical picker's eyelid movement characteristic data to generate the picker's eye state characteristic data.

[0112] Step S145: Use the picker's pupil characteristic data to perform state correction on the picker's eye state characteristic data, so as to obtain the picker's accurate eye state characteristic data;

[0113] Specifically, for example, use the picker's pupil characteristic data to perform state correction on the picker's eye state characteristic data. For example, by analyzing the characteristics such as the size, shape and reflection of the pupil, adjust and optimize the picker's eye state characteristic data, so as to obtain more accurate picker's accurate eye state characteristic data.

[0114] Step S146: Perform picker physical state recognition on the picker's accurate eye state characteristic data, so as to obtain the picker's physical state data.

[0115] Specifically, for example, according to the picker's accurate eye state characteristic data, use machine learning and pattern recognition algorithms to recognize and classify the picker's physical state. For example, judge whether the picker is fatigued or overly tense, so as to obtain the picker's physical state data.

[0116] In the present invention, the eye image data of the picker is extracted and the features are extracted to obtain the picker's eyeball movement feature data, the picker's eyelid opening and closing feature data, the picker's blink frequency feature data, and the picker's pupil feature data. These feature data can reflect the movement of the picker's eyes and provide useful information for subsequent state correction and physical state recognition. By performing time series correlation on the picker's eyelid opening and closing feature data and the blink frequency feature data, the picker's eyelid movement feature data can be obtained. These feature data can more comprehensively describe the movement state of the picker's eyes and face, providing a more accurate basis for subsequent state correction. Using the picker's eyeball movement feature data and pupil feature data, biomimetic correction and state correction are performed on the picker's eyelid movement feature data, thereby obtaining the picker's precise eyeball state feature data. This can eliminate the biological individual differences and environmental influences in eye movement and obtain more accurate picker's eyeball state information. Based on the picker's precise eyeball state feature data, the picker's physical state is recognized. By analyzing the correlation between the eyeball state and the physical state, the current physical state of the picker, such as fatigue level, attention concentration level, etc., can be judged.

[0117] Preferably, the state correction is calculated through the picker's precise eyeball state calculation formula, and the picker's precise eyeball state calculation formula is specifically:

[0118]

[0119] W is the picker's precise eyeball state feature data, a0 is the first initial adjustment term, a1 is the first weight term, k is the original eyeball state data, a2 is the second weight term, v is the pupil movement data, a3 is the third weight term, m is the picker's eyeball state feature data, o is the error correction term, b is the constant term, l is the picker's eyeball state feature data adjustment term, c0 is the second initial adjustment term, c1 is the fourth weight term, d is the pupil shape data, and u is the correction term of the picker's precise eyeball state feature data.

[0120] The present invention constructs a precise eye state calculation formula for pickers, which fully considers the first initial adjustment term a0, the first weight term a1, the original eye state data k, the second weight term a2, the pupil movement data v, the third weight term a3, the picker's eye state characteristic data m, the error correction term o, the constant term b, the picker's eye state characteristic data adjustment term l, the second initial adjustment term c0, the fourth weight term c1, the pupil shape data d, and the interaction relationships among them. The original eye state data k represents the initial measurement value of the picker's eye state. The pupil movement data v represents the movement characteristics of the picker's pupil. The picker's eye state characteristic data m includes eyelid opening and closing characteristic data, blink frequency characteristic data, etc., which are used to describe the details and characteristics of the picker's eye state. The constant term b is used to adjust the proportional relationship between the adjustment term and the eye state characteristic data. The picker's eye state characteristic data adjustment term l is used to adjust the value range or magnitude of the eye state characteristic data. The pupil shape data d represents the shape characteristics of the picker's pupil, providing information about the picker's fatigue level or attention level. By comprehensively considering multiple parameters and characteristic data, as well as their weight relationships and correction terms, the picker's eye state characteristic data can be calculated more accurately, providing more accurate and reliable picker's eye state information. This helps to further analyze the physical condition, attention level, and working conditions of pickers.

[0121] Preferably, step S2 is specifically as follows:

[0122] Step S21: Perform distribution matrix association based on the storage warehouse data and the goods storage data to obtain the warehouse goods storage data;

[0123] Specifically, for example, use Internet of Things technology, RFID and other devices to monitor and locate the shelves and goods in the storage warehouse in real time, so as to obtain the warehouse goods storage data, such as information on the shelf location, goods type, quantity, etc. of the goods, and convert it into the form of a distribution matrix.

[0124] Step S22: Optimize the basic warehouse goods storage data by using the goods value data and the goods demand data in the goods storage data to generate the warehouse goods storage data;

[0125] Specifically, for example, according to the goods value data and the goods demand data in the goods storage data, perform optimization calculations through mathematical models or algorithms, such as using greedy strategies or dynamic programming methods.

[0126] Step S23: Process the picker's goods storage requirements based on the goods storage data and the picker's basic information to generate the picker's goods storage requirement data.

[0127] Specifically, for example, based on the goods storage data and the basic information of the order pickers (such as working hours, professional background, etc.), combined with the characteristics and requirements of the order picking tasks, algorithms or mathematical models are used to process and optimize the goods storage requirements of the order pickers. For example, adjacent order picking tasks are assigned to the same order picker to reduce the walking distance of the order picker in the warehouse, thereby generating more effective goods storage requirement data for the order pickers.

[0128] The present invention optimizes the storage layout and location of goods in the warehouse by analyzing the storage warehouse data and the goods storage data, combined with the value and demand of the goods, improving the utilization rate of the storage space and the access efficiency of the goods. By considering the working ability and characteristics of the order pickers, the goods storage requirements are matched with the order pickers, and the goods storage tasks are reasonably allocated, reducing the redundancy and waste of the order picking paths, and improving the working efficiency and accuracy of the order pickers. By optimizing the warehouse goods storage and order picking processes, the supply chain requirements can be responded to more quickly and accurately. High-value or high-demand goods can be given priority treatment, shortening the order processing time, and enhancing the flexibility and response ability of the supply chain. By optimizing the order picking paths and order picking efficiency, the customer order requirements can be met more quickly and accurately, improving the delivery speed and order accuracy, and enhancing customer satisfaction and loyalty.

[0129] Preferably, step S3 is specifically as follows:

[0130] Step S31: Optimize the physical strength requirements of the order pickers for the warehouse goods storage data by using the physical strength goods storage requirement data in the goods storage requirement data of the order pickers, thereby generating physical strength goods distribution data;

[0131] Specifically, for example, based on the physical strength goods storage requirement data in the goods storage requirement data of the order pickers, machine learning or mathematical models are used to analyze and optimize the warehouse goods storage data. For example, goods with a larger weight are more reasonably distributed in positions close to the working area of the order picker, reducing the physical strength consumption of the order picker, thereby generating goods distribution data that more conforms to the physical strength requirements of the order pickers.

[0132] Step S32: Optimize the goods value extraction requirements for the warehouse goods storage data by using the goods value storage requirement data in the goods storage requirement data of the order pickers, thereby generating the goods value distribution data to be extracted;

[0133] Specifically, for example, based on the goods value storage requirement data in the goods storage requirement data of the order pickers, algorithms or models are used to analyze and optimize the warehouse goods storage data. For example, high-value goods are more reasonably distributed in positions with higher security, thereby reducing the theft risk and improving the degree of guarantee of the goods value, thereby generating the goods value distribution data to be extracted that more conforms to the requirements of the order pickers.

[0134] Step S33: Generate the storage space distribution of goods based on the physical energy goods distribution data and the goods value distribution data to be extracted, so as to generate the storage space distribution change data of goods.

[0135] Specifically, for example, based on the physical energy goods distribution data and the goods value distribution data to be extracted, with the help of space planning and optimization algorithms, such as deep learning neural networks, genetic algorithms, etc., analyze and optimize the storage space of warehouse goods. For example, place similar types of goods in one area, place high-value goods in a high-security location, etc., so as to generate storage space distribution change data of goods that is more suitable for the operation of pickers.

[0136] The present invention, by considering the physical condition and ability of pickers, reasonably distributes the goods storage tasks, reduces the burden on pickers with lower physical energy, improves work efficiency and accuracy, and reduces physical energy consumption and fatigue. By considering the value and importance of goods, preferentially arrange the storage locations of high-value or high-demand goods to ensure their easy access and extraction, improve the order processing speed and customer satisfaction. By reasonably planning the storage locations of goods and the space utilization of the warehouse, reduce the walking distance and picking time of pickers, optimize the picking path, improve the picking efficiency, and reduce errors and delays. Through the above optimization measures, improve the picking efficiency and accuracy, shorten the order processing time, enhance the flexibility and response ability of the supply chain, and promote the improvement of the overall supply chain efficiency.

[0137] Preferably, step S4 is specifically as follows:

[0138] Obtain the historical goods extraction data, and use the historical goods extraction data for prediction calculation to obtain the future goods extraction demand data;

[0139] Specifically, for example, obtain the historical goods extraction data: With the help of Internet of Things devices, RFID and other technologies, record and monitor the inbound and outbound of goods in real time to obtain the historical goods extraction data. Use the historical goods extraction data for prediction calculation: Use machine learning or time series analysis and other methods to analyze and model the historical goods extraction data, and use the obtained model to predict and calculate the future goods extraction demand. For example, take the goods extraction data of the past several months as input, use algorithms such as neural networks or decision trees to predict the future goods extraction volume, and generate the future goods extraction demand data.

[0140] Generate the picking path data based on the future goods extraction demand data and the storage space distribution change data of goods.

[0141] Specifically, for example, picking path data is generated based on future cargo extraction demand data and changes in the distribution of cargo storage space: by combining future cargo extraction demand data and changes in the distribution of cargo storage space, optimization algorithms such as graph theory algorithms and genetic algorithms are used to generate optimal picking path data. For example, pickers are preferentially arranged to pick goods in areas with shorter walking distances within the warehouse, and at the same time, considering future cargo extraction demand data, goods of the same type are concentrated in one area, etc., so as to generate more efficient picking path data.

[0142] Through predictive calculations using historical cargo extraction data, the present invention can accurately predict future cargo extraction demands. According to the prediction results and changes in the cargo storage space, optimal picking paths are generated, reducing the walking distance and extraction time of pickers, and improving picking efficiency. Based on future cargo extraction demand data, the storage locations and picking sequences of goods are reasonably planned, enabling high-demand or important goods to be preferentially processed and extracted, improving the utilization efficiency of resources and the accessibility of goods. By predicting future cargo extraction demands, picking paths can be planned in advance, reducing ad-hoc adjustments and processing, lowering the error rate and omission rate during the picking process, and improving order accuracy and customer satisfaction. By analyzing historical cargo extraction data and predicting future demands, warehouse managers can obtain important information about cargo flow, thereby optimizing aspects such as the warehouse's storage strategy, personnel scheduling, and equipment configuration. This helps improve the efficiency and flexibility of warehouse operations, reducing costs and time waste.

[0143] Preferably, step S5 is specifically as follows:

[0144] Step S51: Perform first optimized path processing on the picking path data using high physical fitness working time data, thereby generating high physical fitness picking path data;

[0145] Specifically, for example, according to high physical fitness working time data, in combination with the picking path data, algorithms or models are used to perform first optimization processing on the picking path. For example, high physical fitness pickers are assigned to areas with short distances and high cargo demand density, thereby generating picking path data that is more suitable for high physical fitness pickers.

[0146] Step S52: Perform second optimized path processing on the picking path data using general working time data, thereby generating general physical fitness picking path data;

[0147] Specifically, for example, according to general working time data, in combination with the picking path data, algorithms or models are used to perform second optimization processing on the picking path. For example, general physical fitness pickers are assigned to areas with a reasonable distribution of cargo storage space and lower picking difficulty, thereby generating picking path data that is more suitable for general physical fitness pickers.

[0148] Step S53: Perform spatio-temporal coupling screening and optimization on the high-physical-fitness picking path data and the general-physical-fitness picking path data, so as to generate optimized picking path data.

[0149] Specifically, for example, based on the high-physical-fitness picking path data and the general-physical-fitness picking path data, with the help of spatio-temporal coupling analysis and optimization algorithms, such as dynamic programming, greedy algorithm, etc., screen and optimize the picking path. For example, provide multiple optional picking path plans for the pickers, and by comprehensively considering factors such as time, distance, cargo demand, and picker's physical fitness, select the optimal picking path method, so as to generate more efficient and optimized picking path data.

[0150] The present invention optimizes the high-physical-fitness working time data, so that the advantages of pickers can be fully utilized in the high-physical-fitness state, improving the picking efficiency and work quality. By considering the general working time data of pickers, reasonably adjust the task allocation and path planning, avoid the excessive fatigue of pickers and the increase of error rate, and improve the accuracy and reliability of work. Through spatio-temporal coupling screening and optimization, dynamically adjust the picking path and task allocation according to the physical fitness state and working ability of pickers at different time periods, so that pickers can efficiently complete work tasks at different time periods. It can fully consider the physical fitness state, fatigue degree and work efficiency of pickers, realize the intelligent optimization of the picking path, thereby improving the work efficiency of the warehouse, reducing the waste of human resources, and ensuring the timeliness and accuracy of cargo extraction.

[0151] This application provides a picking path optimization system, and the system includes:

[0152] At least one processor;

[0153] A memory communicatively connected to the at least one processor;

[0154] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor, so that the at least one processor can execute a picking path optimization method as described above.

[0155] The beneficial effects of the present invention are as follows: By comprehensively considering the working hours of order pickers, the storage requirements of goods, and the order picking path data, the planning of the order picking path and the task allocation are optimized. This can ensure that order pickers perform order picking work during appropriate time periods, reduce task redundancy and unnecessary walking paths, thereby improving the order picking efficiency and accuracy. Through the optimization of the order picking path, the work tasks and working hours of order pickers can be reasonably allocated, avoiding waste of resources and over-fatigue. This can reduce the cost of human resources and improve the operational efficiency of the warehouse. By generating warehouse goods storage data and processing the goods storage requirement data of order pickers, the optimization and management of the goods storage space are realized. A reasonable goods storage distribution can reduce problems such as goods stacking and misalignment, improving the accessibility and extraction efficiency of goods. By obtaining future extraction requirement data of goods and combining it with the data on changes in the distribution of the goods storage space, the prediction and planning of future requirements can be carried out. This can adjust the order picking path and task allocation in advance to adapt to future demand changes and ensure timely fulfillment of customer order requirements. By comprehensively considering the working hours of order pickers, the goods storage requirements, and the future extraction requirements, and through the optimization of the order picking path, the overall optimization of warehouse operations is achieved. The optimized order picking path can make the most of the working hours and capabilities of order pickers, improving the overall operational efficiency and productivity of the warehouse.

[0156] Therefore, from any perspective, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended application documents rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the application documents are intended to be encompassed within the present invention.

[0157] The above description is only a specific implementation manner of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather will conform to the broadest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for optimizing the picking path, characterized in that It includes the following steps: Step S1: Obtain the storage warehouse data, goods storage data, and picker's basic information, and extract the working hours according to the picker's basic information to obtain the picker's working hour data; Step S2: Generate the warehouse goods storage data according to the storage warehouse data and the goods storage data, and process the picker's goods storage requirements according to the goods storage data and the picker's basic information to generate the picker's goods storage requirement data; Step S3: Use the picker's goods storage requirement data to correct the picker's requirements for the warehouse goods storage data to generate the goods storage space distribution change data; Step S4: Obtain the future goods extraction requirement data, and generate the picking path data according to the future goods extraction requirement data and the goods storage space distribution change data; Step S5: Optimize the picking path data using the picker's working hour data to generate the optimized picking path data.

2. The method according to claim 1, wherein The picker's working hour data includes high physical strength working hour data and general working hour data. Step S1 is specifically as follows: Step S11: Obtain the storage warehouse data, goods storage data, and picker's basic information; Step S12: Classify the picker's basic information using the picker's status working hour model to obtain the picker's basic working hour data; Step S13: Control the camera to collect the picker's image data to obtain the picker's image data; Step S14: Extract the picker's status from the picker's image data to obtain the picker's physical status data; Step S15: Use the picker's physical status data to process the picker's basic working hour data to obtain the picker's working hour data.

3. The method according to claim 2, wherein Among them, the specific steps for constructing the picker's status working hour model in Step S12 are as follows: Obtain the standard picker's basic data, where the standard picker's basic data includes standard picker identity data, standard picker skill data, and standard picker work record data; Perform data preprocessing and feature extraction according to the standard picker's basic data to obtain the picker's picking intensity duration feature data, picker's picking intensity feature data, and picker's physiological condition feature data; Construct a deep model according to the picker's picking intensity duration feature data, picker's picking intensity feature data, and picker's physiological condition feature data to construct the picker's status working hour model.

4. The method according to claim 2, wherein Step S14 is specifically as follows: Extract the eye image data from the picker's image data to obtain the picker's eye image data; Perform feature extraction according to the picker's eye image data to obtain the picker's eye movement feature data, picker's eyelid opening and closing feature data, picker's blink frequency feature data, and picker's pupil feature data; Perform time series correlation according to the picker's eyelid opening and closing feature data and the picker's blink frequency feature data to obtain the picker's eyelid movement feature data; Use the picker's eye movement feature data to perform biomimetic correction on the picker's eyelid movement feature data to generate the picker's eye state feature data; The state of the picker's eye state characteristic data is corrected using the picker's pupil characteristic data, so as to obtain the accurate picker's eye state characteristic data; The picker's physical state is recognized based on the accurate picker's eye state characteristic data, so as to obtain the picker's physical state data.

5. The method according to claim 4, characterized in that, Among them, the state correction is calculated through the picker's accurate eye state calculation formula, and the picker's accurate eye state calculation formula is specifically: W is the picker's accurate eye state characteristic data, a0 is the first initial adjustment term, a1 is the first weight term, k is the original eye state data, a2 is the second weight term, v is the pupil movement data, a3 is the third weight term, m is the picker's eye state characteristic data, o is the error correction term, b is the constant term, l is the picker's eye state characteristic data adjustment term, c0 is the second initial adjustment term, c1 is the fourth weight term, d is the pupil shape data, and u is the correction term of the picker's accurate eye state characteristic data.

6. The method according to claim 1, wherein Step S2 is specifically: The distribution matrix is associated according to the storage warehouse data and the goods storage data, so as to obtain the warehouse goods storage data; The warehouse goods storage basic data is optimized using the goods value data and goods demand data in the goods storage data, so as to generate the warehouse goods storage data; The picker's goods storage demand data is processed according to the goods storage data and the picker's basic information, so as to generate the picker's goods storage demand data.

7. The method according to claim 1, wherein Step S3 is specifically: The picker's physical energy goods storage demand data in the picker's goods storage demand data is used to optimize the picker's physical energy demand for the warehouse goods storage data, so as to generate the physical energy goods distribution data; The picker's goods value storage demand data in the picker's goods storage demand data is used to optimize the goods value extraction demand for the warehouse goods storage data, so as to generate the goods value distribution data to be extracted; According to the physical energy goods distribution data and the goods value distribution data to be extracted, the goods storage space distribution is generated, so as to generate the goods storage space distribution change data.

8. The method according to claim 1, characterized in that, Step S4 is specifically: Historical goods extraction data is obtained, and prediction calculations are performed using the historical goods extraction data, so as to obtain the future goods extraction demand data; The picking path data is generated according to the future goods extraction demand data and the goods storage space distribution change data.

9. The method according to claim 2, characterized in that Step S5 is specifically: The first optimized path processing is performed on the picking path data using the high physical energy working time data, so as to generate the high physical energy picking path data; The second optimized path processing is performed on the picking path data using the general working time data, so as to generate the general physical energy picking path data; The high physical energy picking path data and the general physical energy picking path data are subjected to spatio-temporal coupling screening and optimization, so as to generate the optimized picking path data.

10. A picking path optimization system, characterized in that, The system includes: At least one processor; A memory communicatively connected to the at least one processor; The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor, so that the at least one processor can execute a picking path optimization method according to any one of claims 1 to 9.

Images