Warehouse management method and system based on intelligent shelf label

Abstract

The embodiment of the application provides a kind of based on intelligent shelf label warehousing management method and system, it is related to warehousing management technical field, first, the real-time interaction data of intelligent shelf label in warehousing area and the static attribute data of corresponding shelf load warehousing goods are acquired, the dynamic association system of intelligent shelf label and warehousing goods is generated by the dynamic association of the two, then based on the warehousing goods dynamic association system and warehousing area environment monitoring data, intelligent shelf label state deduction model is constructed, for predicting the change of intelligent shelf label sensing state and the attribute retention time of warehousing goods, then according to warehousing goods flow plan data and model prediction result, mark the correlation unit that needs to be adjusted and generate dynamic management instruction, finally, execute dynamic management instruction, update relevant data synchronously, realize the automation and intelligentization of warehousing management, improve management efficiency and warehousing space utilization, guarantee the storage quality of warehousing goods.

Description

Technical Field

[0001] This application relates to the field of warehouse management technology, and more specifically, to a warehouse management method and system based on intelligent shelf labels. Background Technology

[0002] In the field of warehouse management, traditional methods mainly rely on manual record-keeping and simple electronic tag systems. Manual record-keeping is not only inefficient but also prone to human error, leading to inaccurate information about stored goods and consequently affecting the storage, retrieval, and circulation of these goods. While simple electronic tag systems improve the automation level of warehouse management to some extent, their functionality is relatively limited, typically providing only basic location information for stored goods and failing to provide comprehensive perception and dynamic management of detailed attributes of stored goods and the warehouse environment.

[0003] Existing warehouse management technologies struggle to acquire detailed real-time interactive information from smart shelving tags, such as unique tag identifiers, sensor signal waveforms, signal transmission time, and overlapping sensing areas. They also fail to comprehensively grasp the static attribute data of stored goods, including material properties, stacking limitations, and environmental sensitivity indicators. This makes it difficult for warehouse managers to accurately manage and adjust the storage status of goods based on their actual attributes and dynamic changes in the storage environment. Furthermore, existing warehouse management methods lack the ability to predict the duration of smart shelving tag status and stored goods attributes, hindering the ability to plan the flow and storage layout of goods in advance. This results in low warehouse management efficiency and fails to meet the demands of modern warehouse management for high efficiency, accuracy, and intelligence. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a warehouse management method and system based on intelligent shelf labels.

[0005] According to a first aspect of this application, a warehouse management method based on smart shelf labels is provided, the method comprising:

[0006] The system acquires real-time interactive data of smart shelf labels within the storage area and static attribute data of the corresponding shelf-borne stored items. The real-time interactive data includes the unique identifier of the smart shelf label, the waveform of the smart shelf label sensing signal, the signal transmission time between smart shelf labels, and the data of the overlapping area of ​​the smart shelf label sensing range. The static attribute data includes the unique identifier of the stored item, the description of the material properties of the stored item, the stacking restrictions of the stored item, and the environmental sensitivity indicators of the stored item.

[0007] The real-time interactive data and the static attribute data are dynamically associated to generate a dynamic association system between smart shelf labels and stored items. Each association unit in the dynamic association system corresponds to a mapping relationship between a set of real-time interactive data and the static attribute data of a stored item. The association units are connected by association rules based on the similarity of the smart shelf label sensing signal waveform and the matching of the material attribute description of the stored item.

[0008] Based on the dynamic association system between the smart shelf labels and stored goods and the environmental monitoring data of the storage area, a smart shelf label status prediction model is constructed. The smart shelf label status prediction model is used to predict the changes in the sensing status of the smart shelf labels and the duration of attribute retention of the stored goods according to the changing trend of the environmental monitoring data.

[0009] Based on the flow plan data of the stored goods and the prediction results of the intelligent shelf label status inference model, the associated units that need to be adjusted are marked in the intelligent shelf label and stored goods dynamic association system, and a dynamic management instruction is generated that includes the exclusive identifier of the target intelligent shelf label, the exclusive identifier of the target stored goods, the adjustment execution order and the operation time sequence.

[0010] Execute the dynamic management command to synchronously update the intelligent shelf label sensing status, the connection relationship between the intelligent shelf label and the dynamic association system of stored goods, and the input parameters of the intelligent shelf label status inference model in the real-time interactive data.

[0011] According to a second aspect of this application, a warehouse management system based on intelligent shelf labels is provided. The warehouse management system based on intelligent shelf labels includes a processor and a readable storage medium. The readable storage medium stores a program that, when executed by the processor, implements the aforementioned warehouse management method based on intelligent shelf labels.

[0012] Based on any of the above aspects, by comprehensively acquiring real-time interactive data of smart shelf tags and static attribute data of the corresponding stored items within the storage area, and dynamically associating them, a dynamic association system between smart shelf tags and stored items is constructed. This system accurately reflects the mapping relationship between each smart shelf tag and the stored item. By establishing connections between associated units through association rules, the storage status information of stored items becomes more complete and accurate. Furthermore, based on the dynamic association system between smart shelf tags and stored items, and the environmental monitoring data of the storage area, a smart shelf tag status prediction model can accurately predict changes in the sensing status of smart shelf tags and the duration of attribute retention of stored items according to the changing trends of environmental monitoring data. This allows warehouse managers to understand changes in the storage status of stored items in advance, rationally arrange the flow plan of stored items, avoid damage or deterioration of stored items due to environmental factors, and improve the storage quality and safety of stored items. Finally, based on the flow plan data of stored goods and the prediction results of the intelligent shelf label status inference model, dynamic management instructions are generated and executed. This enables real-time adjustment of the association between intelligent shelf labels and stored goods, and synchronous updates of relevant data. This achieves automation and intelligence in warehouse management, greatly improving the efficiency of warehouse management, reducing labor costs, and allowing for flexible adjustment of warehouse layout according to actual conditions, thereby improving the utilization rate of warehouse space. Attached Figure Description

[0013] Figure 1 A schematic flowchart of the warehouse management method based on smart shelf labels provided in an embodiment of this application is shown.

[0014] Figure 2 This illustration shows a schematic diagram of the component structure of a warehouse management system based on intelligent shelf tags provided in an embodiment of this application. Detailed Implementation

[0015] Figure 1 The diagram illustrates a flow chart of a warehouse management method based on smart shelf labels provided in an embodiment of this application. It should be understood that in other embodiments, the order of some steps in the warehouse management method based on smart shelf labels can be interchanged according to actual needs, or some steps can be omitted or deleted. The detailed steps of this warehouse management method based on smart shelf labels are described below.

[0016] Step S110: Obtain real-time interactive data of smart shelf labels within the storage area and static attribute data of the corresponding shelf-borne stored items. The real-time interactive data includes the unique identifier of the smart shelf label, the waveform of the smart shelf label sensing signal, the signal transmission time between smart shelf labels, and the data of the overlapping area of ​​the smart shelf label sensing range. The static attribute data includes the unique identifier of the stored item, the description of the material properties of the stored item, the stacking restrictions of the stored item, and the environmental sensitivity indicators of the stored item.

[0017] In this embodiment, an electronics storage center is used as an example. This center is equipped with multiple smart shelves, each labeled with a unique identifier, such as a string of letters and numbers, to distinguish different smart shelf labels. The smart shelf labels continuously sense the surrounding environment and the stored items on the shelves, generating corresponding sensing signal waveforms. These waveforms change with environmental variations and the state of the stored items. The signal transmission time between smart shelf labels refers to the time spent transmitting signals between different labels, reflecting communication status and distance between them. The overlapping area data of the sensing range of smart shelf labels refers to the data related to the overlapping areas of the sensing ranges of adjacent or nearby smart shelf labels, including the size and shape of the overlapping areas.

[0018] For static attribute data, each stored item is assigned a unique identifier upon entry into the warehouse, which can also be a string in a specific format. The material attribute description details the constituent materials of the stored item, such as metal, plastic, glass, or composite materials. Different materials will have different effects on the sensing signals of the smart shelf tags. The stacking restrictions specify the requirements for stacking the stored item, such as maximum stacking height, stacking layer limits, and whether it can be mixed with other items. The environmental sensitivity indicators indicate the degree of sensitivity of the stored item to environmental factors, such as the tolerance range for environmental parameters such as temperature, humidity, light, and airflow.

[0019] In acquiring this data, smart shelf tags collect relevant real-time interactive data through their sensors and transmit it to the warehouse management system's database via wireless network. Static attribute data of stored goods is entered into the warehouse management system by staff using scanning devices when the goods are received, and stored in association with the goods' unique identifier. To protect data privacy and prevent leakage, encryption technology is used during data transmission to encrypt data packets, ensuring that data is not illegally obtained or tampered with during transmission. Simultaneously, during data storage, sensitive information about stored goods and smart shelf tags is anonymized, removing or replacing potentially privacy-sensitive information.

[0020] Step S120: Dynamically associate the real-time interactive data with the static attribute data to generate a dynamic association system between smart shelf labels and stored items. Each association unit in the dynamic association system corresponds to a mapping relationship between a set of real-time interactive data and the static attribute data of a stored item. The association units are connected through association rules based on the similarity of the smart shelf label sensing signal waveform and the matching of the material attribute description of the stored item.

[0021] In this embodiment, the above-mentioned dynamic association processing is a key link in realizing intelligent warehouse management. By effectively associating real-time changing interactive data with relatively stable static attribute data, the correspondence between intelligent shelf labels and stored items can be accurately reflected.

[0022] Step S121: Extract the unique identifier of each smart shelf label and the corresponding smart shelf label sensing signal waveform from the real-time interactive data, and organize them into a smart shelf label waveform sequence according to the acquisition time of the smart shelf label sensing signal. Each smart shelf label waveform sequence contains the smart shelf label sensing signal waveform characteristics of the same smart shelf label at different acquisition time points.

[0023] Specifically, from the real-time interactive data database, data is filtered based on the unique identifier of each smart shelf tag, extracting all the sensing signal waveform data corresponding to each smart shelf tag. Then, these sensing signal waveforms are arranged in chronological order of their acquisition time to form a smart shelf tag waveform sequence. Each element in each smart shelf tag waveform sequence represents the sensing signal waveform of that smart shelf tag at a specific acquisition time point, and waveform features can include information reflecting waveform characteristics such as peak value, valley value, frequency, and amplitude. For example, for smart shelf tag A, the sensing signal waveform acquired at different time points will be different. Arranging these waveforms in chronological order constitutes the waveform sequence of smart shelf tag A. Each waveform in this sequence contains characteristics such as the peak shape, rising edge, and falling edge variation trends at that time point.

[0024] Step S122: Extract the unique identifier of each stored item and the corresponding material attribute description of the stored item from the static attribute data, and organize them into a storage item attribute sequence according to the storage item entry time. Each storage item attribute sequence contains the storage item material attribute description of the same storage item and the record of the status change of the storage item after entry into the warehouse.

[0025] Similarly, in the static attribute data, the material attribute description of each stored item is extracted based on its unique identifier. Then, according to the order of the stored items' arrival time, the relevant information of the same stored item is organized into a stored item attribute sequence. In addition to the material attribute description, this stored item attribute sequence also includes records of the stored item's state changes after arrival, such as whether it has been moved or experienced abnormal environmental conditions. For example, when stored item B arrives at the warehouse, its material attribute description is plastic. After arrival, it may develop minor scratches on its surface due to environmental changes; these state changes will be recorded in the attribute sequence of stored item B.

[0026] Step S123: Define a continuous time window. Within each time window, extract the waveform features of the intelligent shelf label sensing signal in the intelligent shelf label waveform sequence and match them with the material attribute descriptions of the stored items in the stored item attribute sequence. The matching basis is a preset association rule between the waveform features of the intelligent shelf label sensing signal and the material attribute descriptions of the stored items. The association rule is determined based on the reflection and absorption characteristics of different materials to the intelligent shelf label sensing signal.

[0027] First, based on the actual needs of warehouse management and the frequency of data collection, continuous time windows are defined. The length of the time window can be set according to the actual situation, such as 1 hour, 2 hours, etc. Within each time window, the waveform characteristics of the smart shelf tag sensing signal are extracted from the smart shelf tag waveform sequence, and the corresponding material attribute descriptions of the stored items are extracted from the stored item attribute sequence.

[0028] The pre-defined association rules are derived from extensive experiments and data analysis. Different materials used in storage exhibit varying reflection and absorption characteristics to the sensing signals emitted by smart shelf tags. For example, metal materials reflect certain frequencies of sensing signals more strongly and absorb them less strongly, resulting in sensing signal waveforms received by smart shelf tags with higher peak values ​​and specific frequency distributions. Conversely, plastic materials may reflect sensing signals less strongly and absorb them more strongly, leading to sensing signal waveforms with lower peak values ​​and different frequency distributions. Based on these characteristics, a correspondence between different materials and sensing signal waveform features is established as the basis for matching.

[0029] Step S1231: Within the current time window, extract the waveform features of the smart shelf tag sensing signal from the smart shelf tag waveform sequence. The waveform features of the smart shelf tag sensing signal include the waveform peak shape, rising edge variation trend, falling edge variation trend, and oscillation frequency distribution.

[0030] Within the currently defined time window, each waveform in the smart shelf tag waveform sequence is analyzed to extract the waveform peak morphology, i.e., the height and shape of the peak; the rising edge trend refers to the changes in the waveform as it rises from the starting point to the peak, such as the speed and steepness of the rise; the falling edge trend refers to the changes in the waveform as it falls from the peak to the end point; and the oscillation frequency distribution reflects the energy distribution of the waveform in different frequency bands. By extracting the above features, the characteristics of the smart shelf tag sensing signal waveform can be comprehensively described.

[0031] Step S1232: Extract the material attribute description of the stored items from the stored item attribute sequence. The material attribute description of the stored items includes the material density parameter, surface smoothness parameter, electrical conductivity parameter, and dielectric constant category parameter of the stored items.

[0032] From the attribute sequence of stored goods, detailed parameters describing the material properties of the stored goods are extracted. The material density parameter represents the mass of the material per unit volume; the surface smoothness parameter describes the smoothness of the surface; the conductivity parameter reflects the conductivity of the material; and the dielectric constant category parameter categorizes the dielectric constant of the material into different types. These parameters collectively constitute the core content of the material property description and are also the key basis for matching with the waveform characteristics of the intelligent shelf tag sensing signals.

[0033] Step S1233: Invoke the preset association rules between the waveform characteristics of the intelligent shelf label sensing signal and the material properties of the stored items. In the association rules, different material density parameters of stored items correspond to different waveform peak shapes, different surface smoothness parameters of stored items correspond to different rising edge trends, different conductivity parameters of stored items correspond to different falling edge trends, and different dielectric constant category parameters of stored items correspond to different oscillation frequency distributions.

[0034] During matching, a pre-defined association rule base is invoked. This rule base explicitly defines the correspondence between various material attribute parameters and waveform characteristics. For example, a higher material density parameter of stored goods corresponds to a higher and sharper peak shape in the induced signal waveform; a higher surface smoothness parameter of stored goods will make the rising edge of the induced signal waveform more steep; a better conductivity parameter of stored goods corresponds to a smoother falling edge trend; different dielectric constant categories of stored goods correspond to different oscillation frequency distributions, such as a higher dielectric constant category may correspond to a more concentrated energy distribution in the lower frequency range.

[0035] Step S1234: Compare the waveform peak shape in the extracted intelligent shelf tag sensing signal waveform features with the storage item material density parameter in the storage item material attribute description to determine whether the two conform to the correspondence in the association rules.

[0036] According to the association rules, the peak shape of the extracted smart shelf tag sensing signal is first compared one by one with the material density parameters of the stored goods in the material attribute description. It is then checked whether the peak shape of the current waveform matches the peak shape specified in the association rules based on the material density parameters. For example, if the association rules specify a certain peak height and shape for a material with a certain density parameter range, then it is determined whether the current waveform peak shape falls within that specified range.

[0037] Step S1235: Compare the rising edge variation trend of the extracted intelligent shelf label sensing signal waveform features with the surface smoothness parameter of the stored goods in the description of the material properties of the stored goods, and determine whether the two conform to the correspondence in the association rules.

[0038] Next, the rising edge trend of the waveform is compared with the surface smoothness parameters of the stored goods. According to the association rules, different surface smoothness parameters correspond to different rising edge steepness or change rates. The matching status is determined by comparing whether the rising edge trend of the current waveform is consistent with the trend specified in the rules.

[0039] Step S1236: Compare the falling edge variation trend of the extracted intelligent shelf tag sensing signal waveform features with the conductive characteristic parameters of the stored goods in the material attribute description of the stored goods, and determine whether the two conform to the correspondence in the association rules.

[0040] Then, the falling edge trend of the waveform is compared with the conductivity parameters of the stored goods. The association rules will specify the falling edge characteristics corresponding to different conductivity parameters, such as the speed of falling edge change and whether there are fluctuations. By checking whether the falling edge of the current waveform meets these characteristics, the degree of matching between the two is determined.

[0041] Step S1237: Compare the oscillation frequency distribution in the extracted intelligent shelf tag sensing signal waveform features with the dielectric constant category parameter of the stored goods in the material attribute description of the stored goods, and determine whether the two conform to the correspondence in the association rules.

[0042] Finally, the oscillation frequency distribution of the waveform is compared with the dielectric constant category parameters of the stored goods. According to the association rules, different dielectric constant category parameters correspond to different frequency distribution patterns. This comparison is completed by analyzing whether the distribution of the current waveform's oscillation frequency in each frequency band matches the pattern specified in the rules.

[0043] Step S1238: If all four comparisons above match the corresponding relationship in the association rules, then it is determined that the exclusive identifier of the smart shelf label and the exclusive identifier of the stored item are successfully matched within the current time window; if one or more comparisons do not match, then the division range of the time window is readjusted and the comparison operation is performed again until a matching combination of the exclusive identifier of the smart shelf label and the exclusive identifier of the stored item is found or it is confirmed that there is no matching combination within the time window.

[0044] If all four comparisons above satisfy the correspondence in the association rules, then it can be determined that within the current time window, the unique identifier of the smart shelf label and the unique identifier of the stored item have successfully matched, and a corresponding mapping relationship exists between them. If any one or more comparisons do not meet the rules, it indicates that the current time window division may be inappropriate, or that the smart shelf label and the stored item do not actually have a corresponding relationship within that time window. In this case, it is necessary to readjust the start and end times of the time window, expand or shrink the scope of the time window, and then repeat the comparison operation according to the above steps. After multiple adjustments and comparisons, if a matching combination that meets the conditions can be found, the matching relationship is recorded; if no matching combination can be found, it is confirmed that there is no matching smart shelf label and stored item combination within that time window.

[0045] Step S124: For each time window, generate an association record for the smart shelf label exclusive identifier and the warehouse item exclusive identifier that are successfully matched. The association record includes the smart shelf label exclusive identifier, the warehouse item exclusive identifier, the time window identifier, the smart shelf label waveform feature matching item, and the warehouse item material attribute matching item.

[0046] After matching is completed within each time window, a correlation record is automatically generated for each successfully matched smart shelf label identifier and stored item identifier. This correlation record details the unique identifiers of both parties for subsequent querying and identification. The time window identifier marks the time window corresponding to this correlation record, facilitating the tracing of the time range in which the correlation occurred. The smart shelf label waveform feature matching item records which characteristics of the smart shelf label's sensing signal waveform match the stored item's material attribute description during the matching process, such as peak shape and rising edge variation trend. The stored item material attribute matching item records which parameters in the stored item's material attribute description participated in the matching and the matching result.

[0047] Step S125: Arrange all associated records of all time windows in chronological order, filter out the combination of smart shelf label exclusive identifier and warehouse item exclusive identifier that maintains a matching relationship within a continuous time window, and use the combination as the core association unit.

[0048] All generated associated records are arranged chronologically according to their time windows, forming a complete time series. This time series is then analyzed to identify combinations that maintain the same matching relationship between the smart shelf label's unique identifier and the stored item's unique identifier across multiple consecutive time windows. These combinations, due to their high stability and persistence, are determined as core associated units, representing a relatively fixed correspondence between smart shelf labels and stored items. For example, if smart shelf label C and stored item D match successfully across multiple consecutive time windows, then they constitute a core associated unit.

[0049] Step S126: Based on the similarity of the sensing signal waveforms of the smart shelf labels with unique identifiers corresponding to the core associated units, a connection is established between adjacent core associated units. Core associated units with similarity reaching a preset association standard form a continuous associated unit chain.

[0050] For each identified core associated unit, the smart shelf tag sensing signal waveform, uniquely identified by the smart shelf tag, is extracted. The degree of association between different core associated units is determined by calculating the similarity between these waveforms. Similarity calculation can comprehensively consider multiple aspects, such as the peak shape of the waveform, the trend of rising and falling edges, and the distribution of oscillation frequency. When the waveform similarity of two adjacent core associated units reaches a preset association standard, a connection is established between them. In this way, multiple core associated units with similar waveform characteristics are connected to form a continuous associated unit chain, which reflects the group relationship between related smart shelf tags and stored items.

[0051] Step S127: Supplement the associated records within the non-continuous time window as auxiliary associated units, and establish an indirect connection between the auxiliary associated units and the core associated units through the waveform characteristics of the intelligent shelf tag sensing signal, and finally form a dynamic association system between the intelligent shelf tag and the stored items, which includes the core associated units, auxiliary associated units and the connection relationship between the units.

[0052] Besides the core association units, there are also association records that appear within discontinuous time windows. While these records do not have continuous matching relationships, they may reflect temporary or special correspondences between smart shelf tags and stored items in certain situations; therefore, they are considered auxiliary association units. To integrate these auxiliary association units into the overall association system, indirect connections are established between them by analyzing the similarity between the waveform characteristics of the smart shelf tag sensing signals of the auxiliary association units and those of the core association units. For example, if the waveform characteristics of an auxiliary association unit are similar to those of a core association unit, then the auxiliary association unit is connected to the association system through that core association unit. Ultimately, the core association units, auxiliary association units, and the various connections between them together constitute a dynamic association system between smart shelf tags and stored items. This dynamic association system can dynamically reflect the complex correspondences between smart shelf tags and stored items.

[0053] Step S130: Based on the intelligent shelf label and the dynamic association system of stored goods and the environmental monitoring data of the storage area, construct an intelligent shelf label status prediction model. The intelligent shelf label status prediction model is used to predict the changes in the sensing status of the intelligent shelf label and the duration of attribute retention of the stored goods according to the changing trend of the environmental monitoring data.

[0054] In this embodiment, constructing a smart shelf label status prediction model is an important means to predict the status of smart shelf labels and stored goods. By combining the dynamic association system between smart shelf labels and stored goods with environmental monitoring data of the storage area, a relationship model can be established between environmental changes and the sensing status of smart shelf labels and the attributes of stored goods, thereby enabling the prediction of future states.

[0055] Step S131: Obtain environmental monitoring data of the storage area. The environmental monitoring data includes temperature change records, humidity change records, light intensity change records, and airflow speed change records. All records are arranged into a continuous data sequence in chronological order.

[0056] Multiple environmental monitoring sensors are installed within the storage area, collecting environmental parameters such as temperature, humidity, light intensity, and airflow velocity in real time. Temperature change records reflect temperature fluctuations over time, including periods of increase, decrease, and stability. Humidity change records track changes in air humidity, also arranged chronologically. Light intensity change records show variations in light intensity within the storage area, such as differences between day and night light levels and changes caused by switching lights on and off. Airflow velocity change records reflect changes in airflow velocity within the storage area, such as changes in airflow velocity when ventilation equipment is turned on and off. This environmental monitoring data is continuously transmitted to the warehouse management system, organized chronologically into a continuous data sequence, and stored in a database.

[0057] Step S132: Extract the data of overlapping areas of the sensing range of the smart shelf label from the real-time interactive data corresponding to each core association unit in the dynamic association system between the smart shelf label and the stored goods, and the environmental sensitivity index of the stored goods in the corresponding static attribute data. The environmental sensitivity index of the stored goods includes the temperature tolerance range, humidity tolerance range, light tolerance range and airflow tolerance range of the stored goods.

[0058] In the dynamic association system between smart shelf labels and stored goods, each core association unit corresponds to specific real-time interaction data and static attribute data. Data on the overlapping area of ​​the smart shelf label sensing range corresponding to that core association unit is extracted from the real-time interaction data. This data reflects the mutual influence of the smart shelf label sensing ranges. Simultaneously, corresponding environmental sensitivity indicators for stored goods are extracted from the static attribute data. These indicators specify the tolerance range of stored goods to various environmental factors. Specifically, the temperature tolerance range defines the temperature range within which stored goods can be stored normally; exceeding this range may lead to damage or spoilage. The humidity tolerance range specifies the permissible humidity range. The light tolerance range indicates the degree of tolerance of stored goods to light intensity. The airflow tolerance range describes the adaptability range of stored goods to airflow speed.

[0059] Step S133: Analyze the relationship between the temperature change records in the environmental monitoring data and the overlapping area data of the smart shelf tag sensing range of the core associated unit. By comparing continuous time series, determine the first influence law of temperature change on the overlapping area data of the smart shelf tag sensing range. The first influence law is used to represent the expansion or contraction pattern of the overlapping area data of the smart shelf tag sensing range when the temperature is in different change ranges.

[0060] Step S1331: Extract continuous temperature change records from the environmental monitoring data, divide them into multiple temperature data segments at fixed time intervals, and each temperature data segment contains the maximum temperature, minimum temperature and average temperature within that time interval.

[0061] First, a continuous temperature change record is selected from environmental monitoring data. Then, this temperature change record is divided into multiple temperature data segments at fixed time intervals, such as every hour or every two hours. Each temperature data segment contains the maximum, minimum, and average temperature values ​​collected within that time interval. The maximum temperature is the highest temperature recorded during that time period, the minimum temperature is the lowest temperature, and the average temperature is the arithmetic mean of all temperature data within that time period. In this way, the continuous temperature change record can be transformed into a series of temperature data segments with statistical characteristics, facilitating subsequent analysis.

[0062] Step S1332: Extract the overlapping area data of the smart shelf label sensing range corresponding to each time interval from the core association unit of the smart shelf label and the dynamic association system of the stored goods. Divide the data into multiple overlapping area data segments according to the same time interval. Each overlapping area data segment includes the maximum area, minimum area and average area of ​​the overlapping area of ​​the smart shelf label sensing range within the time interval.

[0063] For each time interval corresponding to a temperature data segment, the overlapping area data of the smart shelf label sensing range within the corresponding time interval is extracted from the core association unit of the smart shelf label and warehouse item dynamic association system. Similarly, according to the same time interval, the above overlapping area data is divided into multiple overlapping area data segments. Each overlapping area data segment contains the maximum area, minimum area, and average area of ​​the overlapping area of ​​the smart shelf label sensing range within that time interval. The maximum area is the largest value reached by the overlapping area within that time period, the minimum area is the smallest value, and the average area is the average area of ​​the overlapping area within that time period.

[0064] Step S1333: Pair each temperature data segment with the corresponding overlapping area data segment to form a temperature range pairing group. Each temperature range pairing group includes a time interval identifier, temperature data, and the corresponding smart shelf tag sensing range overlapping area data.

[0065] The segmented temperature data is paired one-to-one with the corresponding overlapping area data segments to form temperature range pairing groups. Each pairing group has a unique time interval identifier to indicate the time range it corresponds to. The temperature data includes the maximum, minimum, and average temperature values ​​within that time interval, while the overlapping area data of the smart shelf tag sensing range includes the maximum, minimum, and average area of ​​the overlapping area. Through this pairing, the temperature changes within a specific time interval can be correlated with the overlapping area data of the smart shelf tag sensing range.

[0066] Step S1334: For each temperature range pairing group, analyze the relationship between the average temperature and the average area of ​​the overlapping area of ​​the smart shelf label sensing range, and statistically analyze the direction of change of the average area of ​​the overlapping area of ​​the smart shelf label sensing range when the average temperature changes.

[0067] For each temperature range pairing group, the focus is on analyzing the relationship between the average temperature and the average area of ​​the overlapping region of the smart shelf tag sensing range. By comparing the average area of ​​the overlapping region under different average temperatures, we can statistically determine whether the average area of ​​the overlapping region increases or decreases as the average temperature increases or decreases. For example, when the average temperature increases within a certain range, we observe whether the average area of ​​the overlapping region increases accordingly, or whether it decreases when the average temperature decreases. Statistical analysis of these trends helps to initially understand the impact of temperature on the area of ​​the overlapping region.

[0068] Step S1335: Calculate the correspondence between the maximum temperature value and the maximum area of ​​the overlapping region of the smart shelf label sensing range, and determine the variation range of the maximum area of ​​the overlapping region of the smart shelf label sensing range when the maximum temperature value is in any value range.

[0069] Besides the average temperature, the maximum temperature may also affect the maximum area of ​​the overlapping region of the smart shelf label sensing range. The variation in the maximum area of ​​the overlapping region of the smart shelf label sensing range was statistically analyzed within different ranges of maximum temperature values. For example, when the maximum temperature is in a lower range, the variation in the maximum area of ​​the overlapping region is smaller; while when the maximum temperature is in a higher range, the variation in the maximum area of ​​the overlapping region is larger. These statistics clearly demonstrate the degree of influence of the maximum temperature on the maximum area of ​​the overlapping region.

[0070] Step S1336: Calculate the correspondence between the minimum temperature value and the minimum area of ​​the overlapping region of the smart shelf label sensing range, and determine the variation range of the minimum area of ​​the overlapping region of the smart shelf label sensing range when the minimum temperature value is in any numerical range.

[0071] Similarly, the correlation between the minimum temperature and the minimum overlapping area of ​​the smart shelf tag sensing range was analyzed. The variation in the minimum overlapping area was statistically analyzed when the minimum temperature fell within different value ranges. For example, the variation in the minimum overlapping area might be more pronounced when the minimum temperature was lower, and relatively smaller when the minimum temperature was higher.

[0072] Step S1337: Classify the statistical results of all temperature range pairing groups according to temperature change intervals, and summarize the influence patterns of temperature changes in different intervals on the data of overlapping areas of the smart shelf tag sensing range.

[0073] The statistical results of all temperature range pairs are categorized according to temperature variation intervals, such as low temperature, normal temperature, and high temperature intervals. Within each temperature variation interval, the impact of the average, maximum, and minimum temperature values ​​on the overlapping area data of the smart shelf tag sensing range is comprehensively analyzed, summarizing the influence pattern of temperature changes on the overlapping area data within that interval. For example, in the high temperature interval, an increase in temperature may cause the overlapping area of ​​the smart shelf tag sensing range to expand; while in the low temperature interval, a decrease in temperature may cause the overlapping area to shrink.

[0074] Step S1338: Organize the influence pattern into a regularized description to form the first influence law of temperature change on the data of the overlapping area of ​​the sensing range of the smart shelf label.

[0075] The observed influence patterns within different temperature ranges are organized and summarized, described using standardized language to form the first influence law. This first influence law clearly explains how the data in the overlapping area of ​​the smart shelf tag's sensing range changes under different temperature variations—whether it expands or contracts, and the degree and trend of change. For example, when the temperature rises slowly within the normal temperature range, the overlapping area of ​​the smart shelf tag's sensing range will expand slowly at a set rate; when the temperature rises sharply within the high temperature range, the overlapping area will expand rapidly, and so on.

[0076] Step S134: Analyze the relationship between the humidity change records in the environmental monitoring data and the environmental sensitivity indicators of the stored goods in the core associated unit. By comparing continuous time series, determine the second influence law of humidity change on the properties of stored goods. The second influence law is used to represent the pattern of change in the material property description of stored goods when the humidity exceeds the humidity tolerance range of the stored goods in the environmental sensitivity indicators of stored goods.

[0077] Humidity change records were extracted from environmental monitoring data and compared with the humidity tolerance range of stored goods in the environmental sensitivity indicators of the core related units. Through continuous time series comparison, the changes in the material properties of stored goods when humidity exceeds their tolerance range were observed. For example, when humidity exceeds the upper limit, some paper stored goods may become damp, soften, or deform, resulting in a decrease in the hardness parameter in their material property description; for metal stored goods, oxidation and rust may occur, changing the surface condition from smooth to rough. Analysis of a large amount of such data led to the summarization of patterns in the changes in material property descriptions when humidity exceeds the tolerance range, forming a second influencing law.

[0078] Step S135: Analyze the relationship between the light intensity change records and the environmental sensitivity indicators of the stored goods, and the airflow speed change records and the overlapping area data of the intelligent shelf label sensing range, respectively, to determine the third influence law of light intensity change on the attributes of stored goods and the fourth influence law of airflow speed change on the sensing range of intelligent shelf label.

[0079] To analyze the relationship between light intensity variation records and environmental sensitivity indicators of stored goods, the light intensity variation records were compared with the light tolerance range of the stored goods in the environmental sensitivity indicators. When the light intensity exceeded the light tolerance range of the stored goods, the changes in the properties of the stored goods were observed. For example, some light-sensitive plastic stored goods may fade or age under strong light, and the color and intensity parameters in their material property descriptions will change. By analyzing these changes, a third influence law on the properties of stored goods based on light intensity variation was summarized.

[0080] To investigate the relationship between airflow velocity variation records and the overlapping area data of the smart shelf label's sensing range, a time-series comparison was performed. Changes in airflow velocity may affect the propagation of the smart shelf label's sensing signal, thus leading to changes in the sensing range. For example, when the airflow velocity increases, it may interfere with the smart shelf label's sensing signal, causing the sensing range to shrink, and consequently, the overlapping area to decrease. By analyzing the changes in the overlapping area data of the sensing range under different airflow velocities, a fourth influence law on the sensing range of the smart shelf label was determined.

[0081] Step S136: Integrate the first influence law, the second influence law, the third influence law, and the fourth influence law into a basic inference module. The input of the basic inference module is the change in environmental monitoring data, and the output is the predicted value of the sensing state change of the smart shelf tag and the predicted value of the attribute retention time of the stored items.

[0082] The previously identified first, second, third, and fourth influencing factors are integrated to construct a basic inference module. The core function of this module is to comprehensively analyze and infer changes in the input environmental monitoring data using these four influencing factors, thereby predicting changes in the sensing status of smart shelf tags and the duration of attribute retention for stored goods. For example, if the temperature rises significantly and the humidity exceeds the tolerance range in the input environmental monitoring data, the basic inference module will predict changes in the overlapping area of ​​the smart shelf tag sensing range based on the first influencing factor, predict changes in the material properties of the stored goods based on the second influencing factor, and combine these changes to predict the duration of attribute retention for the stored goods.

[0083] Step S137: The core association units in the dynamic association system between the smart shelf label and the stored items are connected to the basic inference module in chronological order. The output of the previous core association unit is used as the input supplement of the next core association unit to form a chain inference structure.

[0084] In the dynamic association system between intelligent shelf tags and stored items, the core association units are formed in chronological order. These core association units are sequentially connected to the basic inference module. In this chain structure, the predicted values ​​of sensor state changes and attribute retention durations output by the basic inference module after processing by the previous core association unit serve as supplementary inputs for the next core association unit. Therefore, when the next core association unit performs its inference, it not only considers the current changes in environmental monitoring data but also references the inference results of the previous core association unit, making the entire inference process more coherent and accurate, and able to reflect the continuous changes in the status of intelligent shelf tags and stored items over time.

[0085] Step S138: Add a time decay factor to the chain-like extrapolation structure. The time decay factor is used to correct the degree of change in the influence of environmental monitoring data over time. The time decay factor is used to reflect the gradual decrease in the influence of historical environmental data on the current prediction results over time.

[0086] Because the impact of environmental monitoring data on smart shelf labels and the status of stored goods gradually weakens over time, a time decay factor is incorporated into the chain-like extrapolation structure. The time decay factor is a parameter determined based on time intervals, and its value gradually decreases with increasing time. During the extrapolation process, the impact of historical environmental data on the current prediction result is multiplied by the corresponding time decay factor, ensuring that environmental data from further back in time has a lower impact on the current prediction result. For example, the impact of yesterday's environmental data on today's prediction result is corrected by the time decay factor, making its impact less than that of today's latest environmental data.

[0087] Step S139: Verify the chain-like simulation structure using historical warehouse data, adjust the values ​​of the influence law parameters and time decay factors of the basic simulation module, and finally form an intelligent shelf label status simulation model.

[0088] The constructed chain-like extrapolation structure was validated using a large amount of historical warehousing data stored in the warehouse management system. Historical environmental monitoring data was input into the chain-like extrapolation structure to obtain prediction results, which were then compared with actual historical changes in the smart shelf tag sensing status and the duration of stored item attribute retention. Based on the comparison results, the magnitude and causes of prediction errors were analyzed, and parameters of various influencing factors in the basic extrapolation module, such as the influence degree coefficient, change rate parameter, and time decay factor, were adjusted. Through multiple iterations of validation and parameter adjustments, the prediction results of the chain-like extrapolation structure accurately reflected the actual situation, ultimately forming a stable and reliable smart shelf tag status extrapolation model.

[0089] Step S140: Based on the flow plan data of the stored goods and the prediction results of the intelligent shelf label status inference model, mark the associated units that need to be adjusted in the dynamic association system between the intelligent shelf label and the stored goods, and generate a dynamic management instruction that includes the exclusive identifier of the target intelligent shelf label, the exclusive identifier of the target stored goods, the adjustment execution order, and the operation time sequence.

[0090] In the process of warehouse management, by combining the information from the flow plan data of stored goods and the prediction results of the intelligent shelf label status inference model, it is possible to accurately identify the related units that need to be adjusted and generate corresponding dynamic management instructions to ensure the safe storage of stored goods and the normal operation of intelligent shelf labels.

[0091] Step S141: Analyze the circulation plan data of the stored goods, extract the unique identifier of each stored goods, the planned outbound time of the stored goods, the target circulation area of ​​the stored goods, and the circulation quantity requirements of the stored goods, and organize them into a circulation plan list of stored goods according to the unique identifier of the stored goods.

[0092] Warehouse item movement planning data typically exists in the form of documents or electronic data, containing a wealth of information about item movement. This data is analyzed to extract key information for each item, including its unique identifier; planned outbound time (the specific time the item is scheduled to leave the warehouse); target destination area; and quantity requirements. This extracted information is then categorized and organized according to the unique identifiers of each item, creating a detailed warehouse item movement plan list. This facilitates subsequent matching and analysis with the prediction results of the intelligent shelving label status extrapolation model.

[0093] Step S142: The prediction results of the smart shelf label status inference model are classified according to the unique identifier of the stored item. The prediction result corresponding to each unique identifier of the stored item includes the predicted value of the attribute retention time of the stored item and the predicted value of the sensing status change of the corresponding smart shelf label.

[0094] The prediction results of the smart shelf tag status inference model are specific to each stored item and smart shelf tag. These prediction results are categorized according to the unique identifier of each stored item, so that each unique identifier corresponds to a set of prediction results. In this set of prediction results, the predicted duration of attribute retention for a stored item represents the estimated length of time that the item can maintain its normal attributes under the current environmental conditions; the predicted value of the sensing status change of the corresponding smart shelf tag reflects the changes in the sensing status of the smart shelf tag associated with that stored item over a future period, such as changes in sensing range or signal strength.

[0095] Step S143: Compare the planned outbound time of each warehouse item in the warehouse item circulation plan list with the predicted value of the attribute retention time corresponding to the warehouse item. If the predicted value of the attribute retention time is less than the difference between the planned outbound time of the warehouse item and the current time, then mark the associated unit corresponding to the warehouse item as an associated unit that needs to be adjusted in the intelligent shelf label and warehouse item dynamic association system.

[0096] For each stored item in the warehouse item flow plan list, calculate the difference between the planned outbound time and the current time to determine the remaining storage time of the item in the warehouse. Compare this time with the predicted attribute retention time for the corresponding item. If the predicted attribute retention time is less than this difference, it indicates that, based on the current environmental conditions and storage status, the item's attributes may change before the planned outbound date, failing to meet normal flow requirements. In this case, in the intelligent shelf label and warehouse item dynamic association system, the associated unit corresponding to the item is marked as an adjustment unit, requiring adjustments to its storage status or environmental conditions.

[0097] Step S144: If the predicted value of attribute retention duration is greater than or equal to the difference between the planned outbound time of the stored item and the current time, further compare the predicted value of the sensing status change of the smart shelf label corresponding to the stored item with the preset standard value of the sensing status of the smart shelf label. If the predicted value of the sensing status change exceeds the standard value of the sensing status of the smart shelf label, then mark the associated unit corresponding to the stored item as an associated unit that needs to be adjusted in the dynamic association system between the smart shelf label and the stored item.

[0098] If the predicted duration of attribute retention is greater than or equal to the difference between the planned outbound time and the current time, it indicates that the attributes of the stored goods will remain normal before the planned outbound departure. However, further inspection of the sensing status of the smart shelving tag associated with this stored goods is required. Compare the predicted change in the sensing status of the smart shelving tag with the preset standard value for the smart shelving tag's sensing status. The standard value for the smart shelving tag's sensing status specifies the range of sensing status when the smart shelving tag is working normally, such as the size of the sensing range and the upper and lower limits of signal strength. If the predicted change in sensing status exceeds this standard value, it indicates that the smart shelving tag may not be able to accurately sense the status of the stored goods, affecting the monitoring and management of the stored goods. Therefore, this associated unit also needs to be marked as an associated unit requiring adjustment in the dynamic association system between the smart shelving tag and the stored goods.

[0099] Step S145: Mark the associated units that need to be adjusted in the dynamic association system between the smart shelf label and the stored items. Each marked associated unit includes a unique identifier for the stored item, a unique identifier for the corresponding smart shelf label, and the reason for adjustment.

[0100] For the associated units identified as requiring adjustment through the above steps, they are clearly marked in the dynamic association system between smart shelf labels and stored items. Each marked associated unit contains necessary information: a unique identifier for the stored item and a unique identifier for the corresponding smart shelf label to accurately identify the object requiring adjustment. The reason for adjustment explains why the associated unit needs adjustment, such as insufficient attribute retention time or abnormal smart shelf label sensing status. This allows managers to clearly understand the status of each marked associated unit.

[0101] Step S146: For all marked associated units, determine the adjustment execution order based on the urgency attribute of the adjustment reason, and generate dynamic management instructions that include the exclusive identifier of the target smart shelf label, the exclusive identifier of the target stored item, the adjustment execution order, and the operation time sequence.

[0102] Step S1461: The tag association units are divided into two categories according to the adjustment reason. The first category is the tag association units whose adjustment reason involves the duration of the storage item attribute retention. The second category is the tag association units whose adjustment reason involves the change of the smart shelf tag sensing status.

[0103] Based on the reasons for adjusting the associated units, they are divided into two categories. The first category consists of associated units that need adjustment due to insufficient retention time of stored item attributes. These adjustments directly affect the quality and availability of stored items and are therefore highly urgent. The second category consists of associated units that need adjustment due to abnormal sensing status of smart shelf tags. Although this also affects management, its urgency is relatively lower.

[0104] Step S1462: Preset execution order rules to determine that the basic execution order of the first type of tag-associated units takes precedence over the basic execution order of the second type of tag-associated units.

[0105] To ensure that urgent adjustment tasks are prioritized, a pre-defined execution order rule is established, explicitly stating that the basic execution order of first-type tag-associated units generally takes precedence over second-type tag-associated units. That is, all first-type tag-associated units are processed first, followed by second-type tag-associated units, to guarantee the security of stored item attributes.

[0106] Step S1463: For the first type of tag association unit, extract the signal transmission consumption time between smart shelf tags corresponding to each tag association unit from the real-time interactive data. The signal transmission consumption time between smart shelf tags is the signal transmission time between the smart shelf tag corresponding to the tag association unit and the adjacent smart shelf tag.

[0107] For the first type of tag association unit, the signal transmission time between the smart shelf tag corresponding to each tag association unit and its adjacent smart shelf tags is extracted from the real-time interaction data. This signal transmission time reflects the efficiency and distance of communication between smart shelf tags; the shorter the signal transmission time, the closer the smart shelf tags are or the better the communication.

[0108] Step S1464: Sort the first type of tag association units in order of the signal transmission time between smart shelf tags from shortest to longest, assign the tag association unit with the shortest transmission time to the first execution order, and arrange them in order.

[0109] The first type of tag association units are sorted according to the signal transmission time between the extracted smart shelf tags. The shorter the signal transmission time, the better the communication between the corresponding smart shelf tag and its adjacent tags, or the closer they are, making adjustment operations easier. Therefore, they are assigned a higher execution order. The internal execution order of the first type of tag association units is formed by arranging them in ascending order of transmission time.

[0110] Step S1465: For the second type of tag association unit, extract the signal transmission time between smart shelf tags corresponding to each tag association unit, sort them from shortest to longest consumption time, assign the tag association unit with the shortest consumption time to the first execution order, and arrange them in order.

[0111] For the second type of tag association unit, the same method as the first type is used to extract the signal transmission time between smart shelf tags and sort them in ascending order to determine the execution order within the second type of tag association unit.

[0112] Step S1466: If there are two or more smart shelf tags of the tag association unit with the same signal transmission time, then further extract the current humidity value from the environmental monitoring data corresponding to the tag association unit, and assign the tag association unit whose current humidity value is closer to the upper limit of the humidity tolerance range of the stored goods in the environmental sensitivity index of the stored goods to a higher execution order.

[0113] When multiple smart shelf tags with associated tags consume the same amount of time for signal transmission, other factors need to be introduced to determine the execution order. In this case, the current humidity value from the environmental monitoring data corresponding to these associated tags is extracted and compared with the upper limit of the humidity tolerance range in the environmental sensitivity indicators for stored goods. The closer the current humidity value of an associated tag is to the upper limit, the higher the humidity risk faced by the stored goods, and adjustments should be prioritized to prevent further humidity increases and damage to the stored goods.

[0114] Step S1467: If the current humidity value is also the same, extract the storage item entry time corresponding to the marked associated unit, and assign the marked associated unit with the earlier storage item entry time to the earlier execution order.

[0115] If the current humidity values ​​are also the same, and the execution order cannot be distinguished by humidity, then the warehouse entry time corresponding to the tagged associated unit is extracted. Warehouse items with earlier entry times have been stored in the warehouse for a longer period and may require priority adjustment to ensure their normal flow according to plan. Therefore, the tagged associated units with earlier entry times are assigned to a higher execution order.

[0116] Step S1468: The sorted first-class label association units and second-class label association units are integrated according to the basic execution order to form an adjusted execution order. The first-class label association units are placed before the second-class label association units as a whole, and the execution order within the same category is determined according to the sorting result.

[0117] The sorted first and second category of marker association units are integrated according to a preset basic execution order, meaning the first category of marker association units is placed before the second category. Within each category, the specific execution order of each marker association unit is determined according to the previously established sorting result. This forms a complete adjustment execution order, clarifying the order in which each marker association unit is adjusted.

[0118] Step S1469: Mark the execution order level corresponding to each marked associated unit in the adjusted execution order in the marked associated unit information.

[0119] To facilitate subsequent management and operation, the execution order level corresponding to each tagged unit in the execution order adjustment will be marked in the information of that tagged unit. The execution order level can be represented by numbers or other methods, such as level 1 indicating execution first, level 2 indicating second, and so on. After marking the execution order level, managers and the execution system can clearly understand the adjustment priority of each tagged unit.

[0120] Step S14610: After determining the adjustment execution order, based on the spatial layout of the storage area and the planning of the storage item flow path, an operation time sequence is assigned to each marked associated unit. The operation time sequence should avoid conflicts between the adjustment operations of different marked associated units in the same area or the same time period.

[0121] After determining the execution sequence of adjustments, a specific operation time sequence is assigned to each tagged unit, taking into account the spatial layout of the storage area and the planned flow path of stored goods. The spatial layout of the storage area determines the location of each intelligent shelf and the aisle conditions between them, while the planned flow path of stored goods specifies the route for moving stored goods within the warehouse. When assigning operation time sequences, these factors must be fully considered to avoid adjustment operations of different tagged units being performed in the same area or at the same time, which could lead to operational conflicts, aisle blockages, or safety hazards. For example, for two tagged units located on opposite sides of the same aisle, their adjustment operation times should be staggered to avoid simultaneous movement operations that could block the aisle.

[0122] Step S14611: Integrate the target smart shelf label unique identifier, target stored item unique identifier, adjustment execution order and operation time sequence of each tag associated unit to generate dynamic management instructions. The dynamic management instructions also include specific execution requirements corresponding to each operation. The specific execution requirements include the adjustment direction of the smart shelf label sensing parameters and the target shelf position of the stored item movement.

[0123] The relevant information for each tag-associated unit is integrated, including the unique identifier of the target smart shelf label (i.e., the unique identifier of the smart shelf label to which the stored item will be associated after adjustment); the unique identifier of the target stored item (i.e., the unique identifier of the stored item that needs adjustment); the adjustment execution order (i.e., the execution order of the tag-associated unit among all adjustment operations); and the operation time sequence (i.e., the specific start and end times of the operation). Simultaneously, the dynamic management instruction also includes the specific execution requirements for each operation, such as the adjustment direction of the smart shelf label sensing parameters (whether to increase or decrease the sensing range, the magnitude of signal strength adjustment, etc.); and the target shelf location for the stored item to be moved to (i.e., which target shelf the stored item needs to be moved to from the current shelf). This information is integrated to form a complete dynamic management instruction, which is then issued to the warehouse execution system for execution.

[0124] Step S150: Execute the dynamic management instruction to synchronously update the intelligent shelf label sensing status, the connection relationship between the intelligent shelf label and the storage item dynamic association system, and the input parameters of the intelligent shelf label status inference model in the real-time interactive data.

[0125] Once the dynamic management instructions are generated, they need to be executed by the warehouse execution system. During execution, this involves changes in the sensing status of smart shelf tags, adjustments to the association between smart shelf tags and stored items, and updates to the input parameters of the smart shelf tag status deduction model, to ensure that the overall warehouse management system's status remains consistent with the actual situation.

[0126] Step S151: The dynamic management instruction is sent to the warehouse execution system. The warehouse execution system, based on the target smart shelf label unique identifier, the target stored item unique identifier, and the operation time sequence in the instruction, schedules the handling equipment to perform the stored item movement operation, and simultaneously schedules the smart shelf label control module to perform the smart shelf label sensing parameter adjustment operation.

[0127] The warehouse management system sends dynamically generated management instructions to the warehouse execution system via the internal network. Upon receiving the instructions, the warehouse execution system parses them, extracting key information such as the unique identifier of the target smart shelf label, the unique identifier of the target stored item, and the operation time sequence. Based on this information, the warehouse execution system dispatches appropriate handling equipment, such as automated forklifts and conveyor belts, to move the stored items from their current shelf to the target shelf according to the operation time sequence. Simultaneously, it dispatches the smart shelf label control module to adjust the sensing parameters of the target smart shelf label according to the specific execution requirements in the instructions, such as adjusting the sensing frequency and sensitivity, to ensure that the smart shelf label can accurately sense the moved stored items.

[0128] Step S152: During the operation, new sensing data of the smart shelf labels are collected in real time. The new sensing data includes the adjusted smart shelf label sensing signal waveform, the new signal transmission time between smart shelf labels, and the updated data of the overlapping area of ​​the smart shelf label sensing range.

[0129] During the process of material handling equipment moving stored goods and the smart shelf tag control module adjusting sensing parameters, the smart shelf tags continuously collect new sensing data in real time. The waveform of the adjusted smart shelf tag sensing signal reflects the change in the sensing signal received by the smart shelf tag after the sensing parameters are adjusted. The signal transmission time between smart shelf tags refers to the time spent on signal transmission between the adjusted smart shelf tag and adjacent smart shelf tags, which may change due to changes in sensing range or signal strength. The updated smart shelf tag sensing range overlap area data is the latest data of the overlapping area of ​​the smart shelf tag sensing range after the material movement and sensing parameter adjustment, including changes in the size and shape of the overlapping area. This new sensing data is transmitted to the warehouse management system in real time via wireless network.

[0130] Step S153: Replace the old data of the corresponding smart shelf label in the real-time interactive data with the new sensing data to complete the update of the real-time interactive data.

[0131] After receiving new sensor data, the warehouse management system locates the corresponding old data record for the smart shelf label in the real-time interactive data based on the label's unique identifier. It then replaces the corresponding portions of the old data with information from the new sensor data, such as the sensor signal waveform, signal transmission time, and overlapping sensing areas. Once this replacement is complete, the real-time interactive data is updated, accurately reflecting the current sensing status of the smart shelf label.

[0132] Step S154: After the stored item is moved from the original shelf to the target shelf, update the mapping relationship between the smart shelf label and the corresponding associated unit in the dynamic association system of the stored item, disconnect the connection relationship between the original smart shelf label exclusive identifier and the stored item exclusive identifier, establish a new connection relationship between the target smart shelf label exclusive identifier and the stored item exclusive identifier, and update the time attribute of the associated unit to the operation completion time.

[0133] Step S1541: After the stored item is moved from the original shelf to the target shelf, the operation completion signal fed back by the storage execution system and the new sensing data of the smart shelf tag are used for dual verification to verify that the stored item has arrived at the target shelf and that the target smart shelf tag can normally sense the stored item.

[0134] Once the handling equipment moves the stored item from its original shelf to its target shelf, the warehouse execution system sends an operation completion signal to the warehouse management system. Simultaneously, new sensor data collected by the target smart shelf tag is also transmitted to the warehouse management system. The warehouse management system performs dual verification on these two signals, checking the validity of the operation completion signal and whether the new sensor data contains the sensor information for the stored item, to confirm that the stored item has indeed successfully arrived at the target shelf and that the target smart shelf tag can detect the stored item correctly.

[0135] Step S1542: Search for the core association unit corresponding to the original smart shelf label exclusive identifier and the storage item exclusive identifier in the smart shelf label and storage item dynamic association system. The core association unit includes the original smart shelf label exclusive identifier, the storage item exclusive identifier, the old smart shelf label sensing signal waveform characteristics, and the old storage item material attribute matching items.

[0136] In the dynamic association system between smart shelf labels and stored items, the corresponding core association unit is located based on the unique identifiers of the original smart shelf labels and the unique identifiers of the stored items. This core association unit records the association relationship between the original smart shelf labels and the stored items before the relocation operation, including the unique identifiers of the original smart shelf labels, the unique identifiers of the stored items, the waveform characteristics of the old smart shelf label sensing signals (i.e., the waveform characteristics of the sensing signals before the relocation), and the matching items of the old stored item material attributes (i.e., the matching status of the stored item material attributes with the smart shelf label sensing signal waveforms before the relocation).

[0137] Step S1543: In the connection relationship between the core association unit and the adjacent association units in the dynamic association system between the intelligent shelf label and the stored items, delete the connection between the core association unit and the adjacent association units, including the connection with the preceding core association unit, the subsequent core association unit and the auxiliary association unit.

[0138] After locating the core associated unit, in the connection relationship between the intelligent shelf label and the dynamic association system of stored items, delete the connections between this core associated unit and other adjacent associated units. These connections include connections with preceding core associated units (i.e., core associated units that precede this core associated unit in time); connections with subsequent core associated units (i.e., core associated units that follow this core associated unit in time); and connections with auxiliary associated units. Deleting these connections breaks the connection relationship of this core associated unit in the association system.

[0139] Step S1544: Extract the waveform features of the smart shelf label sensing signal corresponding to the exclusive identifier of the target smart shelf label from the updated real-time interactive data, and combine them with the material attribute description of the stored item exclusive identifier to generate a new core association unit. The new core association unit includes the exclusive identifier of the target smart shelf label, the exclusive identifier of the stored item, the new smart shelf label sensing signal waveform features, the new material attribute matching item of the stored item, and the initial time attribute.

[0140] From the updated real-time interactive data, the waveform features of the smart shelf tag sensing signal corresponding to the unique identifier of the target smart shelf tag are extracted. These waveform features are new waveform features collected by the target smart shelf tag after the stored item is moved to the target shelf. This waveform feature is matched with the material attribute description of the stored item corresponding to its unique identifier, generating a new material attribute matching item. Then, the unique identifier of the target smart shelf tag, the unique identifier of the stored item, the new smart shelf tag sensing signal waveform features, the new material attribute matching item, and the initial time attribute (i.e., the operation start time) are combined to generate a new core association unit.

[0141] Step S1545: Insert the new core association unit into the corresponding position of the intelligent shelf label and the dynamic association system of stored items. Based on the waveform characteristics of the intelligent shelf label sensing signal of the new core association unit and the similarity of the intelligent shelf label sensing signal waveform of the adjacent core association units, establish a new connection relationship. Adjacent association units with similarity reaching the preset standard are connected to the new core association unit.

[0142] The newly generated core association units are inserted into the corresponding time positions in the dynamic association system between smart shelf tags and stored items to maintain the temporal sequence integrity of the association system. Then, the similarity between the waveform characteristics of the smart shelf tag sensing signal of the new core association unit and the waveform characteristics of adjacent core association units is calculated. When the similarity reaches a preset standard, new connections are established between the new core association unit and these adjacent core association units, thus forming a new association chain. For example, if the waveform characteristics of the new core association unit have high similarity to the waveform characteristics of its immediate and subsequent adjacent core association units, both reaching the preset standard, then connections are established with these two adjacent core association units respectively, integrating the new core association unit into the continuous association unit chain.

[0143] Step S1546: Check if there is an auxiliary associated unit corresponding to the new core associated unit. If there is an auxiliary associated unit that meets the association conditions, that is, the sensing data of the auxiliary associated unit partially matches the storage item attributes of the new core associated unit, then establish a connection between the new core associated unit and the auxiliary associated unit that meets the association conditions. If it is detected during the operation that the original auxiliary associated unit no longer meets the association conditions, then remove the auxiliary associated unit from the intelligent shelf label and storage item dynamic association system. If a new auxiliary associated unit that meets the association conditions is generated, then add the new auxiliary associated unit that meets the association conditions to the intelligent shelf label and storage item dynamic association system and establish a connection with the core associated unit.

[0144] After a new core association unit is inserted into the association system, it checks whether there are any auxiliary association units related to the new core association unit. For existing auxiliary association units, it is determined whether their sensing data partially matches the stored item attributes of the new core association unit. If a partial match exists and the association conditions are met, an indirect connection is established between the new core association unit and the auxiliary association unit. If, during operation, the original auxiliary association unit no longer meets the association conditions due to reasons such as the movement of stored items or changes in the environment, for example, if the similarity of the sensing signal waveform decreases below a preset standard, the auxiliary association unit is removed from the association system. Simultaneously, if a new auxiliary association unit that meets the association conditions is generated in a new location or under new environmental conditions, meaning its sensing data partially matches the stored item attributes of a core association unit, the new auxiliary association unit is added to the intelligent shelf tag and stored item dynamic association system, and an indirect connection is established with the corresponding core association unit.

[0145] Step S1547: Update the time attribute of the new core associated unit to the operation completion time.

[0146] After inserting the new core associated unit and establishing the connection relationship, the time attribute of the new core associated unit is updated from the initial operation start time to the operation completion time to accurately reflect the actual time when the associated unit was established.

[0147] Step S155: Extract key features from the updated real-time interactive data and the associated unit attributes of the updated smart shelf label and warehouse item dynamic association system, and use them as new input parameters to replace the original input parameters in the smart shelf label state inference model.

[0148] For example, in step S1551, key features are extracted from the updated real-time interactive data. The key features include the peak shape change of the smart shelf tag sensing signal waveform of each smart shelf tag, the average value of the signal transmission time between smart shelf tags, the average area of ​​the overlapping area data of the smart shelf tag sensing range, and the overlap frequency of the sensing range of different smart shelf tags.

[0149] Feature extraction is performed on the updated real-time interactive data to obtain key features that reflect the status of the smart shelf tags. The peak shape change of the smart shelf tag sensing signal waveform reflects the changes in the shape and height of the waveform peak over time; the average signal transmission time between smart shelf tags is the arithmetic mean of the transmission times of multiple signals within a set time, reflecting the overall efficiency of signal transmission; the average area of ​​the overlapping region data of the smart shelf tag sensing range is the average area of ​​the overlapping region over a set time; and the overlap frequency of different smart shelf tag sensing ranges refers to the number of times the sensing ranges overlap per unit time.

[0150] Step S1552: Organize the key features and classify them according to the unique identifier of the smart shelf label. The key features corresponding to each unique identifier of the smart shelf label form a set of feature data, which are used as input parameters related to the smart shelf label in the smart shelf label state inference model.

[0151] The extracted key features are categorized and organized according to the unique identifier of the smart shelf label, so that each unique identifier of the smart shelf label corresponds to a set of data containing the above key features. The above data describes the current state of the smart shelf label from different perspectives and will be used as the input parameters related to the smart shelf label in the smart shelf label state inference model.

[0152] Step S1553: Extract the associated unit attributes from the updated intelligent shelf label and the dynamic association system of stored items. The associated unit attributes include the unique identifier of the core associated unit for the stored item, the description of the material attributes of the stored item, the current value of the storage item attribute retention time, the current value of the environmental sensitivity index of the stored item, and the stability description of the connection relationship of the associated unit.

[0153] The attribute information of associated units is extracted from the updated intelligent shelf label and dynamic association system of stored items. The unique identifier of the core associated unit is used to associate the corresponding stored item; the material attribute description of the stored item is the material characteristic of the current stored item; the current value of the attribute retention time of the stored item refers to the estimated time that the stored item can maintain normal attributes in the current state; the current value of the environmental sensitivity index of the stored item reflects the current tolerance range of the stored item to environmental factors; the stability description of the connection relationship of the associated unit describes the stability of the connection between this associated unit and other associated units, such as connection duration and connection strength.

[0154] Step S1554: Classify the associated unit attributes according to the unique identifier of the stored item. The associated unit attributes corresponding to each unique identifier of the stored item form a set of attribute data, which serves as the input parameters related to the stored item in the intelligent shelf label status inference model.

[0155] The extracted associated unit attributes are classified according to the unique identifier of the stored item. Each unique identifier of the stored item corresponds to a set of data containing the above attributes. These data describe the current status and association of the stored item and serve as the input parameters related to the stored item in the intelligent shelf label status inference model.

[0156] Step S1555: Simultaneously extract the latest data segment from the updated environmental monitoring data. The latest data segment includes temperature, humidity, light, and airflow data within a preset time after the operation is completed, which are used as environmental-related input parameters in the intelligent shelf label status inference model.

[0157] Environmental data such as temperature, humidity, light intensity, and airflow speed are extracted from the updated environmental monitoring data within a preset time after the operation is completed, forming the latest data segment. The above data reflects the latest environmental conditions of the storage area after the operation is completed, and serves as the environment-related input parameters in the intelligent shelf label status inference model.

[0158] Step S1556: Integrate the input parameters related to the smart shelf label, the input parameters related to the stored items, and the input parameters related to the environment according to a preset format to form a new set of input parameters.

[0159] Following the preset format required by the intelligent shelving label status deduction model, the input parameters related to the intelligent shelving labels, the stored items, and the environment are integrated into a complete new set of input parameters. The preset format specifies the order and data type of each parameter to ensure that the model can correctly read and process these parameters.

[0160] Step S1557: Extract the original set of input parameters from the intelligent shelf label state inference model, compare the differences between the new set of input parameters and the original set of input parameters, and mark the differences and their degree.

[0161] The original set of input parameters currently in use is obtained from the intelligent shelf label state inference model. The new set of input parameters is then compared item by item with the original set. For parameter items that differ, the name of the differing item and the degree of difference, such as the magnitude of the parameter value change, are recorded.

[0162] Step S1558: Replace the corresponding items in the original input parameter set with the difference items in the new input parameter set, load the replaced input parameter set into the intelligent shelf label state inference model, and complete the update of the input parameters of the intelligent shelf label state inference model.

[0163] Based on the comparison results, items in the new input parameter set that differ from the original input parameter set are replaced with their corresponding items in the original parameter set. After the replacement, the new input parameter set is loaded into the intelligent shelf label state inference model, enabling the model to make predictions based on the latest state data, ensuring the accuracy and timeliness of the prediction results.

[0164] Step S156: Based on the new input parameters, adjust the influence law parameters of the basic inference module in the smart shelf tag status inference model. If the new sensing data shows that the degree of influence of any environmental factor on the sensing range of the smart shelf tag changes, then correct the corresponding influence law description.

[0165] After updating the input parameters of the smart shelf label status inference model, the influence law parameters in the basic inference module are adjusted based on the changes in the relationship between environmental factors and the sensing range of the smart shelf labels and the attributes of stored items, as reflected in the new input parameters. For example, if the new input parameters show that the influence of temperature on the sensing range of the smart shelf labels has increased, then the temperature influence coefficient in the first influence law is increased accordingly. If the new sensing data indicates that the influence pattern of a certain environmental factor on the sensing range of the smart shelf labels has changed, such as the temperature increase originally causing the sensing range to expand, but now causing the sensing range to shrink, then the corresponding influence law description needs to be revised to accurately reflect the actual influence pattern.

[0166] In step S157, the time decay factor in the intelligent shelf label state inference model is adjusted. The value of the time decay factor is updated according to the interval between the operation completion time and the current time, so that the prediction result of the intelligent shelf label state inference model matches the latest state.

[0167] The time decay factor is adjusted based on the interval between the operation completion time and the current time. As the time interval increases, the value of the time decay factor gradually decreases to reduce the influence of historical environmental data on the current prediction results. By reasonably adjusting the time decay factor, the intelligent shelf label state inference model can better adapt to the latest state, improving the accuracy of the prediction results. For example, when the time interval between the operation completion time and the current time is short, the value of the time decay factor is large, and the influence of historical data is relatively large; when the interval is long, the value of the time decay factor is small, the influence of historical data is weakened, and the model relies more on the latest input parameters for prediction.

[0168] Furthermore, Figure 2 A schematic diagram of the hardware structure of a warehouse management system 100 based on smart shelf tags for implementing the methods provided in the embodiments of this application is shown. Figure 2 As shown, the warehouse management system 100 based on smart shelf tags may include at least one processor 102 (the processor 102 may be, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, a transmission device 106 for communication functions, and a controller 108. Those skilled in the art will understand that... Figure 2 The structure shown is for illustrative purposes only and does not limit the structure of the smart shelf tag-based warehouse management system 100. For example, the smart shelf tag-based warehouse management system 100 may also include... Figure 2 The more or fewer components shown, or having the same Figure 2 The different configurations shown.

[0169] The memory 104 can be used to store software programs and modules for application software, such as the program instructions corresponding to the method embodiments described above in this application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the above-described warehouse management method based on intelligent shelf labels. The transmission device 106 is used to acquire or send data via a network.

[0170] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

Claims

1. A warehouse management method based on intelligent shelf labels, characterized in that, The method includes: The system acquires real-time interactive data of smart shelf labels within the storage area and static attribute data of the corresponding shelf-borne stored items. The real-time interactive data includes the unique identifier of the smart shelf label, the waveform of the smart shelf label sensing signal, the signal transmission time between smart shelf labels, and the data of the overlapping area of ​​the smart shelf label sensing range. The static attribute data includes the unique identifier of the stored item, the description of the material properties of the stored item, the stacking restrictions of the stored item, and the environmental sensitivity indicators of the stored item. The real-time interactive data and the static attribute data are dynamically associated to generate a dynamic association system between smart shelf labels and stored items. Each association unit in the dynamic association system corresponds to a mapping relationship between a set of real-time interactive data and the static attribute data of a stored item. The association units are connected by association rules based on the similarity of the smart shelf label sensing signal waveform and the matching of the material attribute description of the stored item. Based on the dynamic association system between the smart shelf labels and stored goods and the environmental monitoring data of the storage area, a smart shelf label status prediction model is constructed. The smart shelf label status prediction model is used to predict the changes in the sensing status of the smart shelf labels and the duration of attribute retention of the stored goods according to the changing trend of the environmental monitoring data. Based on the flow plan data of the stored goods and the prediction results of the intelligent shelf label status inference model, the associated units that need to be adjusted are marked in the intelligent shelf label and stored goods dynamic association system, and a dynamic management instruction is generated that includes the exclusive identifier of the target intelligent shelf label, the exclusive identifier of the target stored goods, the adjustment execution order and the operation time sequence. Execute the dynamic management command to synchronously update the intelligent shelf label sensing status, the connection relationship between the intelligent shelf label and the storage item dynamic association system, and the input parameters of the intelligent shelf label status inference model in the real-time interactive data. The step of dynamically associating the real-time interactive data with the static attribute data to generate a dynamic association system between intelligent shelf labels and stored items includes: Extract the unique identifier of each smart shelf label and the corresponding smart shelf label sensing signal waveform from the real-time interactive data. Arrange the smart shelf label sensing signal acquisition time sequence into a smart shelf label waveform sequence. Each smart shelf label waveform sequence contains the smart shelf label sensing signal waveform characteristics of the same smart shelf label at different acquisition time points. Extract the unique identifier and corresponding material attribute description of each stored item from the static attribute data, and organize them into a storage item attribute sequence according to the storage item's entry time. Each storage item attribute sequence contains the storage item's material attribute description and the storage item's status change record after entry into the warehouse. A continuous time window is defined. Within each time window, the waveform features of the intelligent shelf label sensing signal in the intelligent shelf label waveform sequence are extracted and matched with the material attribute description of the stored goods in the stored goods attribute sequence. The matching basis is a preset association rule between the waveform features of the intelligent shelf label sensing signal and the material attribute description of the stored goods. The association rule is determined based on the reflection and absorption characteristics of different materials to the intelligent shelf label sensing signal. For each time window, a matching record is generated between the unique identifier of the smart shelf label and the unique identifier of the stored item. The matching record includes the unique identifier of the smart shelf label, the unique identifier of the stored item, the time window identifier, the waveform feature matching item of the smart shelf label, and the material attribute matching item of the stored item. Arrange all associated records of all time windows in chronological order, filter out the combination of smart shelf label exclusive identifier and warehouse item exclusive identifier that maintains a matching relationship within a continuous time window, and use the combination as the core association unit; Based on the similarity of the sensing signal waveforms of the smart shelf labels with unique identifiers corresponding to the core associated units, connections are established between adjacent core associated units, and core associated units with similarity reaching the preset association standard form a continuous associated unit chain. Supplementing non-continuous time windows with associated records serves as auxiliary associated units. These auxiliary associated units are then indirectly connected to the core associated units via the waveform characteristics of the intelligent shelf tag sensing signals. Ultimately, this forms a dynamic association system between the intelligent shelf tag and stored items, encompassing the core associated units, auxiliary associated units, and the connections between these units.

2. The warehouse management method based on intelligent shelf labels according to claim 1, characterized in that, Within each time window, the waveform features of the smart shelf label sensing signal in the smart shelf label waveform sequence are extracted and matched with the material attribute descriptions of the stored goods in the stored goods attribute sequence. The matching is based on preset association rules between the waveform features of the smart shelf label sensing signal and the material attribute descriptions of the stored goods, including: Within the current time window, the waveform features of the smart shelf tag sensing signal are extracted from the smart shelf tag waveform sequence. The waveform features of the smart shelf tag sensing signal include the waveform peak shape, rising edge variation trend, falling edge variation trend, and oscillation frequency distribution. Extract the material attribute description of the stored items from the attribute sequence of the stored items. The material attribute description of the stored items includes the material density parameter, surface smoothness parameter, electrical conductivity parameter, and dielectric constant category parameter of the stored items. The system invokes a preset association rule between the waveform characteristics of the smart shelf tag sensing signal and the material properties of the stored goods. In this association rule, different material density parameters of the stored goods correspond to different waveform peak shapes, different surface smoothness parameters of the stored goods correspond to different rising edge trends, different conductivity parameters of the stored goods correspond to different falling edge trends, and different dielectric constant category parameters of the stored goods correspond to different oscillation frequency distributions. The waveform peak shape in the extracted intelligent shelf tag sensing signal waveform features is compared with the material density parameter of the stored goods in the material attribute description of the stored goods to determine whether the two conform to the correspondence in the association rules. The rising edge variation trend of the extracted intelligent shelf label sensing signal waveform features is compared with the surface smoothness parameter of the stored goods in the material attribute description of the stored goods to determine whether the two conform to the correspondence in the association rules. The falling edge variation trend of the extracted intelligent shelf tag sensing signal waveform features is compared with the conductivity parameter of the stored goods in the material attribute description of the stored goods to determine whether the two conform to the correspondence in the association rules. The oscillation frequency distribution in the extracted intelligent shelf tag sensing signal waveform features is compared with the dielectric constant category parameter of the stored goods in the material attribute description of the stored goods to determine whether the two conform to the correspondence in the association rules. If all four comparisons above match the corresponding relationships in the association rules, then the unique identifier of the smart shelf label and the unique identifier of the stored item are successfully matched within the current time window. If one or more comparisons do not match, the time window is redefined and the comparison operation is performed again until a matching combination of the unique identifier of the smart shelf label and the unique identifier of the stored item is found or it is confirmed that there is no matching combination within the time window.

3. The warehouse management method based on intelligent shelf labels according to claim 1, characterized in that, The intelligent shelf label status prediction model is constructed based on the dynamic association system between the intelligent shelf label and the stored goods, and the environmental monitoring data of the storage area, including: Acquire environmental monitoring data of the storage area. The environmental monitoring data includes temperature change records, humidity change records, light intensity change records, and airflow speed change records. All records are arranged into a continuous data sequence in chronological order. Extract the data of overlapping areas of the sensing range of the smart shelf labels from the real-time interactive data of each core association unit in the dynamic association system between the smart shelf labels and the stored items, and the environmental sensitivity index of the stored items in the corresponding static attribute data. The environmental sensitivity index of the stored items includes the temperature tolerance range, humidity tolerance range, light tolerance range and airflow tolerance range of the stored items. The relationship between temperature change records in the environmental monitoring data and the overlapping area data of the smart shelf tag sensing range of the core associated unit is analyzed. By comparing continuous time series, the first influence law of temperature change on the overlapping area data of the smart shelf tag sensing range is determined. The first influence law is used to represent the expansion or contraction pattern of the overlapping area data of the smart shelf tag sensing range when the temperature is in different ranges. The relationship between humidity change records in the environmental monitoring data and the environmental sensitivity indicators of stored goods in the core associated unit is analyzed. By comparing continuous time series, the second influence law of humidity change on the properties of stored goods is determined. The second influence law is used to represent the pattern of change in the material property description of stored goods when the humidity exceeds the humidity tolerance range of stored goods in the environmental sensitivity indicators of stored goods. The relationships between light intensity change records and environmentally sensitive indicators of stored goods, and between airflow velocity change records and overlapping areas of the smart shelf tag sensing range were analyzed to determine the third influence law of light intensity change on the attributes of stored goods and the fourth influence law of airflow velocity change on the sensing range of smart shelf tags. The first influence law, the second influence law, the third influence law, and the fourth influence law are integrated into a basic inference module. The input of the basic inference module is the change in environmental monitoring data, and the output is the predicted value of the change in the sensing state of the smart shelf tag and the predicted value of the retention time of the stored items' attributes. The core association units in the intelligent shelf label and the dynamic association system of stored items are connected to the basic inference module in chronological order. The output of the previous core association unit is used as the input supplement of the next core association unit to form a chain inference structure. A time decay factor is added to the chain-like extrapolation structure. The time decay factor is used to correct the degree of change in the influence of environmental monitoring data over time. The time decay factor is used to reflect the gradual decrease in the influence of historical environmental data on the current prediction results over time. The chain-like simulation structure was verified by using historical warehouse data. The values ​​of the influence law parameters and time decay factor of the basic simulation module were adjusted to ultimately form an intelligent shelf label status simulation model.

4. The warehouse management method based on intelligent shelf labels according to claim 3, characterized in that, The analysis examines the relationship between temperature change records in the environmental monitoring data and the overlapping area data of the intelligent shelf tag sensing range of the core associated unit. Through comparison of continuous time series, the influence of temperature changes on the overlapping area data of the intelligent shelf tag sensing range is determined, including: Continuous temperature change records are extracted from the environmental monitoring data and divided into multiple temperature data segments at fixed time intervals. Each temperature data segment contains the maximum temperature, minimum temperature and average temperature within that time interval. From the core association unit of the intelligent shelf label and the dynamic association system of the stored goods, extract the data of the overlapping area of ​​the intelligent shelf label sensing range corresponding to the time interval of each temperature data segment, and divide it into multiple overlapping area data segments according to the same time interval. Each overlapping area data segment includes the maximum area of ​​the overlapping area of ​​the intelligent shelf label sensing range, the minimum area of ​​the overlapping area of ​​the intelligent shelf label sensing range, and the average area of ​​the overlapping area of ​​the intelligent shelf label sensing range within that time interval. Each temperature data segment is paired with the corresponding overlapping area data segment to form a temperature range pairing group. Each temperature range pairing group includes a time interval identifier, temperature data, and the corresponding smart shelf label sensing range overlapping area data. For each temperature range pairing group, analyze the relationship between the average temperature and the average area of ​​the overlapping region of the smart shelf label sensing range, and statistically analyze the direction of change of the average area of ​​the overlapping region of the smart shelf label sensing range when the average temperature changes. The correspondence between the maximum temperature value and the maximum area of ​​the overlapping region of the smart shelf label sensing range is statistically analyzed to determine the variation range of the maximum area of ​​the overlapping region of the smart shelf label sensing range when the maximum temperature value is in any value range. The correspondence between the minimum temperature and the minimum area of ​​the overlapping region of the smart shelf label sensing range is statistically analyzed to determine the variation range of the minimum area of ​​the overlapping region of the smart shelf label sensing range when the minimum temperature is in any numerical range. The statistical results of all temperature range pairing groups are classified according to temperature change intervals, and the influence patterns of temperature changes in different intervals on the data of overlapping areas of the smart shelf tag sensing range are summarized respectively. The aforementioned influence patterns are organized into a regularized description, forming the influence law of temperature changes on the data in the overlapping area of ​​the smart shelf label sensing range.

5. The warehouse management method based on intelligent shelf labels according to claim 1, characterized in that, Based on the flow plan data of stored goods and the prediction results of the intelligent shelf label status inference model, the system marks the units that need to be adjusted in the dynamic association system between intelligent shelf labels and stored goods, and generates dynamic management instructions that include the unique identifier of the target intelligent shelf label, the unique identifier of the target stored goods, the adjustment execution order, and the operation time sequence, including: Analyze the flow plan data of warehouse goods, extract the unique identifier of each warehouse goods, the planned outbound time of the warehouse goods, the target flow area of ​​the warehouse goods, and the flow quantity requirements of the warehouse goods, and organize them into a warehouse goods flow plan list according to the unique identifier of the warehouse goods; The prediction results of the smart shelf label status inference model are classified according to the unique identifier of the stored item. The prediction result corresponding to each unique identifier of the stored item includes the predicted value of the attribute retention time of the stored item and the predicted value of the sensing status change of the corresponding smart shelf label. The planned outbound time of each warehouse item in the warehouse item circulation plan list is compared with the predicted value of the attribute retention time corresponding to the warehouse item. If the predicted value of the attribute retention time is less than the difference between the planned outbound time of the warehouse item and the current time, the associated unit corresponding to the warehouse item is marked as an associated unit that needs to be adjusted in the intelligent shelf label and warehouse item dynamic association system. If the predicted duration of attribute retention is greater than or equal to the difference between the planned outbound time of the stored item and the current time, the predicted value of the sensor status change of the smart shelf label corresponding to the stored item is further compared with the preset standard value of the smart shelf label sensor status. If the predicted value of the sensor status change exceeds the standard value of the smart shelf label sensor status, then the associated unit corresponding to the stored item is marked as an associated unit that needs to be adjusted in the dynamic association system between the smart shelf label and the stored item. The associated units that need to be adjusted are marked in the dynamic association system between the smart shelf labels and the stored items. Each marked associated unit includes a unique identifier for the stored item, a unique identifier for the corresponding smart shelf label, and the reason for the adjustment. For all tag-associated units, the order of adjustment execution is determined based on the urgency of the adjustment reason. The execution order of tag-associated units whose adjustment reason involves the duration of storage item attribute retention takes precedence over the order of tag-associated units whose adjustment reason involves changes in the sensing status of smart shelf tags. For tag-associated units with the same adjustment reason, the order of adjustment execution is determined based on the signal transmission time between their corresponding smart shelf tags. After determining the order of adjustment, based on the spatial layout of the storage area and the planning of the flow path of stored goods, an operation time sequence is assigned to each marked associated unit. The operation time sequence should avoid conflicts between the adjustment operations of different marked associated units in the same area or the same time period. The system integrates the unique identifiers of the target smart shelf labels, the unique identifiers of the target stored items, the adjustment execution order, and the operation time sequence of each labeled unit to generate dynamic management instructions. The dynamic management instructions also include the specific execution requirements for each operation, including the adjustment direction of the smart shelf label sensing parameters and the target shelf position for the moved stored items.

6. The warehouse management method based on intelligent shelf labels according to claim 5, characterized in that, For all tag-associated units, the adjustment execution order is determined based on the urgency of the adjustment reason. Tag-associated units whose adjustment reasons involve the duration of storage item attribute retention take precedence over tag-associated units whose adjustment reasons involve changes in the sensing status of smart shelf tags. For tag-associated units with the same adjustment reason, the adjustment execution order is determined based on the signal transmission time between their corresponding smart shelf tags, including: The tag association units are divided into two categories according to the reason for adjustment. The first category is the tag association units whose adjustment reason involves the duration of the storage item attribute retention. The second category is the tag association units whose adjustment reason involves the change of the smart shelf tag sensing status. A preset execution order rule is established to determine that the basic execution order of the first type of tag-associated units takes precedence over the basic execution order of the second type of tag-associated units. For the first type of tag association unit, the signal transmission time between smart shelf tags corresponding to each tag association unit is extracted from the real-time interactive data. The signal transmission time between smart shelf tags is the signal transmission time between the smart shelf tag corresponding to the tag association unit and the adjacent smart shelf tag. The first type of tag association units are sorted from shortest to longest according to the signal transmission time between smart shelf tags. The tag association unit with the shortest time is assigned the first execution order, and so on. For the second type of tag association unit, the signal transmission time between smart shelf tags corresponding to each tag association unit is also extracted, and sorted from shortest to longest consumption time. The tag association unit with the shortest consumption time is assigned the first execution order, and so on. If the signal transmission time between the smart shelf tags of two or more tag association units is the same, then the current humidity value in the environmental monitoring data corresponding to the tag association unit is further extracted, and the tag association unit whose current humidity value is closer to the upper limit of the humidity tolerance range of the stored goods in the environmental sensitivity index of the stored goods is assigned a higher execution order. If the current humidity values ​​are also the same, the warehouse entry time corresponding to the marked associated unit is extracted, and the marked associated unit with the earlier warehouse entry time is assigned a higher execution order. The sorted first-class label association units and the second-class label association units are integrated according to the basic execution order to form an adjusted execution order. The first-class label association units are placed before the second-class label association units as a whole, and the execution order within the same category is determined according to the sorting result. The execution order level corresponding to each marked associated unit in the adjusted execution order is marked in the marked associated unit information.

7. The warehouse management method based on intelligent shelf labels according to claim 1, characterized in that, The execution of the dynamic management command synchronously updates the intelligent shelf label sensing status, the connection relationship between the intelligent shelf label and the dynamic association system of stored goods, and the input parameters of the intelligent shelf label status inference model in the real-time interactive data, including: The dynamic management command is sent to the warehouse execution system. Based on the target smart shelf label unique identifier, the target stored item unique identifier, and the operation time sequence in the command, the warehouse execution system schedules the handling equipment to perform the storage item movement operation, and at the same time schedules the smart shelf label control module to perform the smart shelf label sensing parameter adjustment operation. During operation, new sensing data of the smart shelf labels are collected in real time. The new sensing data includes the adjusted smart shelf label sensing signal waveform, the new signal transmission time between smart shelf labels, and the updated data of the overlapping area of ​​the smart shelf label sensing range. The new sensing data replaces the old data of the corresponding smart shelf label in the real-time interactive data, thus completing the update of the real-time interactive data; When a stored item is moved from its original shelf to its target shelf, the mapping relationship between the smart shelf label and the corresponding associated unit in the dynamic association system of the stored item is updated. The connection between the original smart shelf label's unique identifier and the stored item's unique identifier is disconnected, and a new connection between the target smart shelf label's unique identifier and the stored item's unique identifier is established. At the same time, the time attribute of the associated unit is updated to the operation completion time. If the original auxiliary association unit is detected to no longer meet the association conditions during the operation, the auxiliary association unit is removed from the intelligent shelf label and warehouse item dynamic association system. If a new auxiliary association unit that meets the association conditions is generated, the new auxiliary association unit that meets the association conditions is added to the intelligent shelf label and warehouse item dynamic association system and a connection with the core association unit is established. Extract key features from the updated real-time interactive data and the associated unit attributes of the updated smart shelf label and the dynamic association system of stored items, and use them as new input parameters to replace the original input parameters in the smart shelf label state inference model. Based on the new input parameters, adjust the influence law parameters of the basic inference module in the smart shelf label status inference model. If the new sensing data shows that the degree of influence of any environmental factor on the sensing range of the smart shelf label changes, then correct the corresponding influence law description. At the same time, the time decay factor in the intelligent shelf label status inference model is adjusted. The value of the time decay factor is updated according to the interval between the operation completion time and the current time, so that the prediction result of the intelligent shelf label status inference model matches the latest status.

8. The warehouse management method based on intelligent shelf labels according to claim 7, characterized in that, When a stored item is moved from its original shelf to its target shelf, the mapping relationship between the smart shelf label and the corresponding associated unit in the dynamic association system of the stored item is updated. The connection between the original smart shelf label's unique identifier and the stored item's unique identifier is broken, and a new connection between the target smart shelf label's unique identifier and the stored item's unique identifier is established. Simultaneously, the time attribute of the associated unit is updated to the operation completion time, including: Once the operation to move stored goods is confirmed to be complete, the system verifies the completion of the operation through both the operation completion signal fed back by the storage execution system and the new sensing data from the smart shelf label. This verifies that the stored goods have arrived at the target shelf and that the target smart shelf label can sense the stored goods normally. From the dynamic association system between the smart shelf label and the stored item, find the core association unit corresponding to the original smart shelf label exclusive identifier and the stored item exclusive identifier. The core association unit includes the original smart shelf label exclusive identifier, the stored item exclusive identifier, the old smart shelf label sensing signal waveform characteristics and the old stored item material attribute matching items. In the connection relationship between the core association unit and the dynamic association system of smart shelf labels and warehouse items, delete the connection between the core association unit and the adjacent association units, including the connection with the preceding core association unit, the following core association unit and the auxiliary association unit. Extract the waveform features of the smart shelf label sensing signal corresponding to the unique identifier of the target smart shelf label from the updated real-time interactive data, and combine them with the material attribute description of the stored item with the unique identifier of the stored item to generate a new core association unit. The new core association unit includes the unique identifier of the target smart shelf label, the unique identifier of the stored item, the new smart shelf label sensing signal waveform features, the new material attribute matching item of the stored item, and the initial time attribute. The new core association unit is inserted into the corresponding position of the intelligent shelf label and the dynamic association system of stored items. Based on the waveform characteristics of the intelligent shelf label sensing signal of the new core association unit and the similarity of the intelligent shelf label sensing signal waveform of the adjacent core association units, a new connection relationship is established. Adjacent association units with similarity reaching the preset standard are connected to the new core association unit. Check if there is an auxiliary associated unit corresponding to the new core associated unit. If there is an auxiliary associated unit that meets the association conditions, that is, the sensing data of the auxiliary associated unit partially matches the storage item attributes of the new core associated unit, then establish a connection between the new core associated unit and the auxiliary associated unit that meets the association conditions. Update the time attribute of the new core associated unit to the operation completion time, and at the same time update the time identifier in the connection relationship between the adjacent associated units and the new core associated unit; By comparing the changes in the number of associated units and connection relationships of the intelligent shelf label and the dynamic association system of stored items before and after the update, a unit update report is generated. The unit update report includes the number of deleted associated units, the number of newly added associated units, the number of changes in connection relationships, and the details of changes to key associated units.

9. A warehouse management system based on intelligent shelf labels, characterized in that, The device includes a processor and a readable storage medium storing a program that, when executed by the processor, implements the warehouse management method based on intelligent shelf labels as described in any one of claims 1-8.

Images