Mineral water production process management method

Abstract

The invention belongs to the technical field of production management, and discloses a mineral water production process management method. The method comprises the following steps: step 1, acquiring associated data in a mineral water production process; 2, carrying out preprocessing operation on the associated data, and carrying out data transmission by adopting a dual-channel network transmission technology; 3, network data parameters are detected in real time in the transmission process, and automatic channel switching is carried out; step 4, processing and analyzing based on the data in the data set, and performing corresponding management on the mineral water production process; and 5, making and executing a personalized response strategy based on a management result. When raw material replenishment needs to be carried out, replenishment is carried out according to the actually lacked replenishment quantity, analysis is carried out in combination with the recent replenishment trend condition, the future replenishment trend condition and the recent activity influence condition, the final raw material replenishment quantity is corrected, the replenishment quantity is more accurate, and the replenishment efficiency is improved. Therefore, the phenomenon of raw material insufficiency or raw material accumulation is effectively reduced.

Description

Technical Field

[0001] This invention relates to the field of production management technology, and more specifically, to a method for managing the production process of mineral water. Background Technology

[0002] With the improvement of people's living standards and the enhancement of health awareness, the market demand for mineral water, as a natural and healthy beverage, has shown a continuous growth trend. Mineral water production enterprises are constantly expanding in scale, increasing the number of production lines, and making production processes increasingly complex. Therefore, the management of mineral water production processes is becoming increasingly important.

[0003] There are still many shortcomings in the current production management of mineral water. For example, when managing raw materials, replenishment is generally based on the production plan. However, due to the irregularity of the market, replenishment based solely on the production plan can easily lead to insufficient or excessive raw materials, thus affecting mineral water production. When managing mineral water production equipment, the corresponding operating values ​​of the equipment are usually obtained from the installed sensors. When the alarm value is exceeded, the equipment is judged to be abnormal. Although this management method can detect abnormal equipment, by the time the abnormality is detected, it has already affected the production of mineral water, resulting in a certain lag in alarms.

[0004] In view of this, the present invention proposes a mineral water production process management method to solve the above problems. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art and achieve the above objectives, the present invention provides the following technical solution: A method for managing the production process of mineral water, the method comprising: Step 1: Collect relevant data in the mineral water production process. The relevant data includes raw material management data, process parameters, and production equipment operating parameters. Step 2: Preprocess the associated data to form a dataset, and use dual-channel network transmission technology for data transmission; Step 3: Monitor network data parameters in real time during transmission and perform channel conversion based on the network data parameters; Step 4: Process and analyze the data within the dataset to manage the mineral water production process accordingly. This management includes raw material anomaly management, production equipment anomaly management, and product quality anomaly management. Step 5: Develop and implement personalized response strategies based on management results. Personalized response strategies include determining the quantity of raw materials to replenish when raw materials are abnormal, issuing tiered alarms when production equipment is abnormal, and adjusting process parameters when product quality is abnormal.

[0006] Furthermore, the method for managing product quality anomalies in step four is as follows: The production process of mineral water is divided into multiple production steps, and each production step has corresponding standard process parameters. The actual process parameters corresponding to each production process segment are extracted from the dataset. The actual process parameters are compared with the corresponding standard process parameters. When the difference between the actual process parameters and the standard process parameters is greater than the preset product quality judgment threshold, the product is judged to have a quality abnormality.

[0007] Furthermore, the method for managing production equipment anomalies in step four is as follows: Each production process segment has corresponding standard operating parameters for the production equipment. The actual operating parameters of the corresponding production equipment in each production process segment are extracted from the dataset. The actual operating parameters are compared with the corresponding standard operating parameters. When the difference between the actual operating parameters and the standard operating parameters is greater than the preset equipment anomaly judgment threshold, the production equipment is judged to be abnormal.

[0008] Furthermore, the methods for managing production equipment anomalies in step four also include: If it is determined that there is no abnormality in the production equipment, a first risk coefficient is generated based on the operating parameters of the production equipment, and a second risk coefficient is generated based on the process parameters. The first risk coefficient and the second risk coefficient are weighted and summed to obtain a risk score. When the risk score is greater than the predetermined risk score judgment threshold, it is determined that there is a potential abnormal risk in the production equipment.

[0009] Furthermore, the method for generating the first risk coefficient is as follows: Assuming there are no abnormalities in the production equipment, under time series t, an actual parameter change function is generated based on the actual operating parameters of the production equipment, and a standard parameter change function is generated based on the standard operating parameters of the production equipment; historical operating parameters of the production equipment are obtained, and historical parameter change functions are generated based on the historical operating parameters of the production equipment under normal operation; the integral values ​​of the actual parameter change function and the standard parameter change function, and the actual parameter change function and the historical parameter change function under time series t are calculated respectively. The deviation value of the operating parameters is obtained by the difference between the actual operating parameters and the corresponding standard operating parameters. The deviation value of the operating parameters under the time series t is used to generate the operating deviation parameter change function. The first change coefficient is obtained by the change function of the deviation parameter within the time series t. The first risk value is obtained by combining the first change coefficient and all integral values, and the first risk value is then standardized to obtain the first risk coefficient.

[0010] Furthermore, the method for generating the second risk coefficient is as follows: Obtain the number of product quality anomalies that occur in the production process segment of the corresponding production equipment within the time series t, and calculate the standard deviation of the process parameters based on the actual process parameters of the corresponding production process segment within the time series t; The deviation value of the process parameters is calculated based on the difference between the actual process parameters and the standard process parameters. A process deviation change function is generated based on the process parameter deviation value under the time series t. The second change coefficient is obtained based on the change of the process deviation change function within the time series t. The second risk value is obtained by combining the number of product quality anomalies, the standard deviation of process parameters, and the second variation coefficient. The second risk value is then standardized to obtain the second variation coefficient.

[0011] Furthermore, the method for managing raw material anomalies in step four is as follows: Real-time monitoring of raw material inventory levels and setting inventory warning thresholds. When the raw material inventory level is lower than the warning threshold, it is determined that there is an anomaly in the raw material inventory.

[0012] Furthermore, the method for developing personalized response strategies based on management results in step five is as follows: The method for determining the replenishment quantity of raw materials when there is an anomaly is as follows: A mathematical model for replenishment quantity is constructed, and the replenishment quantity is generated based on this model; raw material management data is extracted from the dataset, and fixed procurement time points are obtained according to the pre-set procurement plan in the raw material management data; when the raw material inventory is less than the warning threshold, the current raw material inventory and the warning threshold are obtained; the replenishment quantity between the last procurement time point and the current procurement time point, and the replenishment quantity between the current procurement time point and the next procurement time point are obtained from the raw material management data; sales impact information at the current procurement time point is obtained, and influencing factors are determined based on this information; the current raw material inventory, the warning threshold, the replenishment quantity between the last procurement time point and the current procurement time point, the replenishment quantity between the current procurement time point and the next procurement time point, and the influencing factors are input into the replenishment quantity mathematical model to obtain the raw material replenishment quantity. The method for graded alarms when production equipment malfunctions is as follows: when the risk score is greater than the predetermined risk score judgment threshold, the location of the malfunction is determined and a first-level alarm signal is generated; when the difference between the actual operating parameters and the standard operating parameters is greater than the preset equipment malfunction judgment threshold, the location of the malfunction is determined and a second-level alarm signal is generated, wherein the urgency of the second-level alarm signal is higher than that of the first-level alarm signal. When product quality is abnormal, the method for adjusting process parameters is as follows: when the difference between the actual process parameters and the standard process parameters is greater than the preset product quality judgment threshold, the difference is recorded as the difference value. When the actual process parameters are higher than the standard process parameters, the process parameters are lowered according to the corresponding difference value. When the actual process parameters are lower than the standard process parameters, the process parameters are raised according to the corresponding difference value.

[0013] Furthermore, the method for determining the impact factor is as follows: The method for determining the impact factor is as follows: Sales impact information is obtained, which refers to relevant information that can significantly affect mineral water sales. This includes feature impact information corresponding to different key features. The obtained sales impact information is input into a trained key feature impact scoring model to obtain the impact score of each key feature. The key feature impact scoring model is a deep neural network model. Each key feature is assigned a corresponding weight coefficient, and the impact scores of each key feature are weighted and summed to obtain the impact factor.

[0014] Furthermore, the method for channel conversion based on network data parameters in step three is as follows: The channel includes a main channel and a secondary channel. Multiple network data parameters of the main channel are acquired in real time. Each network data parameter has a corresponding standard operating data range. When the acquired network data parameter is not within the set standard operating data range, the transmission is switched from the main channel to the secondary channel. When the network data parameters are all restored to the standard operating data range, the transmission is switched back to the main channel. When the real-time acquired network data parameters are within the set standard operating data range, the optimal operating data value for each network data parameter is determined based on the standard operating data range for each network data parameter. The quality score for each network data parameter is derived based on the difference between the real-time acquired network data parameter and its corresponding optimal operating data value. After standardizing the quality scores of each network data parameter, a weighted summation is performed to obtain the total quality score. A Cartesian coordinate system is constructed with time as the x-axis and the total quality score as the y-axis. A curve of the total quality score is constructed in the Cartesian coordinate system, and the area enclosed by the total quality score curve and the x-axis within a time period T is extracted. The standard deviation of the score within time T is calculated, and the area is multiplied by the standard deviation to obtain the scoring coefficient. When the scoring coefficient exceeds a preset scoring coefficient threshold, the main channel transmission is switched to the secondary channel transmission.

[0015] The technical effects and advantages of the mineral water production process management method of this invention are as follows: This invention can derive a first variation coefficient by comprehensively analyzing the deviation between the actual operating parameters and standard operating parameters of the production equipment, the deviation between the actual operating parameters and historical operating parameters, and the changing trend of the actual operating parameters. It can also derive a second risk coefficient by comprehensively analyzing the changing trend and fluctuation of the process parameters under the corresponding production equipment and the number of product anomalies. By combining the first and second risk coefficients for comprehensive analysis, the potential abnormal risks of the production equipment can be assessed more accurately, thereby providing early warning and handling, reducing the occurrence of production equipment anomalies, and ensuring product production efficiency.

[0016] When raw material replenishment is needed, this invention not only replenishes the material based on the actual shortage, but also conducts a comprehensive analysis of recent replenishment trends, future replenishment trends, and the impact of recent activities to adjust the final raw material replenishment quantity, making the replenishment quantity more accurate and effectively reducing the occurrence of raw material shortages or stockpiles. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the steps of a mineral water production process management method according to Embodiment 1 of the present invention. Figure 2 This is a flowchart of a mineral water production process management method according to Embodiment 1 of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example 1

[0020] Please see Figure 1 , Figure 2 As shown in this embodiment, a mineral water production process management method includes: Step 1: Collect relevant data in the mineral water production process. The relevant data includes raw material management data, process parameters, and production equipment operating parameters. Step 2: Preprocess the associated data to form a dataset, and use dual-channel network transmission technology for data transmission; Step 3: Monitor network data parameters in real time during transmission and perform channel conversion based on the network data parameters; Step 4: Process and analyze the data within the dataset to manage the mineral water production process accordingly. This management includes raw material anomaly management, production equipment anomaly management, and product quality anomaly management. Step 5: Develop and implement personalized response strategies based on management results. Personalized response strategies include determining the quantity of raw materials to replenish when raw materials are abnormal, issuing tiered alarms when production equipment is abnormal, and adjusting process parameters when product quality is abnormal.

[0021] Through the above technical solution, this application obtains relevant data related to mineral water production management from multiple data points in the mineral water production process. This relevant data includes, but is not limited to, raw material management data, process parameters for each production stage, and operating parameters of the corresponding production equipment for each production stage. After preprocessing, the relevant data is compiled into a dataset. Preprocessing methods include data denoising, time-series synchronization, outlier removal, missing value imputation, and format standardization, etc., to obtain clearer and more accurate data for subsequent analysis. The data in the dataset is transmitted using IoT transmission technology. The data can be transmitted to the management center platform for storage and analysis. To ensure data transmission integrity, a dual-channel transmission technology is used, prioritizing the main channel. During transmission, multiple network data parameters of the main channel, such as signal strength and transmission rate, are acquired in real time. Each network data parameter has a corresponding standard operating data range. When the real-time acquired network data parameters are not... If the data transmission on the main channel is abnormal within the set standard operating data range, then to ensure transmission efficiency, the main channel transmission is switched to the secondary channel. The main channel network data is continuously monitored, and once it recovers to the corresponding standard operating data range, the transmission is switched back to the main channel, ensuring stable data transmission. The management center platform includes a raw material management unit, an equipment management unit, and a product quality management unit. The raw material management unit manages mineral water raw materials, including raw material procurement, raw material inspection, and raw material inventory. The equipment management unit analyzes the operating parameters of the production equipment in the dataset to identify equipment anomalies and issues timely alarms when an anomaly is detected. The product quality management unit analyzes the process parameters in the dataset to identify product quality anomalies and issues timely adjustments and alarms once an anomaly is detected in the corresponding process parameters. Finally, based on the management results, corresponding responses are taken to any abnormal phenomena. For example, in raw material management, when a procurement warning is triggered, the procurement department promptly contacts the corresponding supplier based on the warning information and places a purchase order according to the replenishment quantity to ensure raw material supply. If problems are found during raw material inspection, the production department isolates, reworks, or destroys products produced using that raw material, while unused raw materials are returned or exchanged. In terms of production equipment anomaly management, when equipment maintenance personnel receive an anomaly warning, they can quickly determine the location and cause of the anomaly, repair or replace parts of the production equipment, and restore normal operation of the production equipment as soon as possible, reducing production downtime. In terms of product quality anomaly management, when an anomaly warning is generated, the process parameters are quickly adjusted according to the production process segment with abnormal process parameters to ensure production quality. This constitutes a complete mineral water production process management method to ensure overall production efficiency and quality.

[0022] The method for managing product quality anomalies in step four is as follows: The production process of mineral water is divided into multiple production steps, and each production step has corresponding standard process parameters. The actual process parameters corresponding to each production process segment are extracted from the dataset. The actual process parameters are compared with the corresponding standard process parameters. When the difference between the actual process parameters and the standard process parameters is greater than the preset product quality judgment threshold, the product is judged to have a quality abnormality.

[0023] The above technical solution provides a specific method for managing product quality anomalies. First, based on the mineral water production process, it is divided into multiple production stages. Each stage has corresponding standard process parameters. The actual process parameters for each stage are extracted from the dataset and compared with the corresponding standard parameters. When the difference between the actual and standard parameters exceeds a preset product quality judgment threshold, a quality anomaly is identified. This threshold is set by those skilled in the art based on the specific circumstances. For example, a production stage involves sterilizing the mineral water. This stage has a standard sterilization temperature. During production, the system automatically compares the real-time sterilization temperature with the standard sterilization temperature and records the deviation. When the sterilization temperature deviation exceeds the maximum allowable temperature deviation judgment threshold, it indicates an abnormal sterilization temperature, thus indicating a quality problem in the product. This ensures that the process parameters for each production stage are within a reasonable range, thereby guaranteeing product quality.

[0024] The method for managing production equipment anomalies in step four is as follows: Each production process segment has corresponding standard operating parameters for the production equipment. The actual operating parameters of the corresponding production equipment in each production process segment are extracted from the dataset. The actual operating parameters are compared with the corresponding standard operating parameters. When the difference between the actual operating parameters and the standard operating parameters is greater than the preset equipment anomaly judgment threshold, the production equipment is judged to be abnormal.

[0025] The above technical solution provides a specific method for managing production equipment anomalies. First, standard operating parameters are set for corresponding production equipment under each production process segment. The actual operating parameters for each production equipment under each production process segment are extracted from the dataset. These actual operating parameters are compared with the corresponding standard operating parameters. When the difference between the actual and standard operating parameters exceeds a preset equipment anomaly judgment threshold, the production equipment is judged to be abnormal. The equipment anomaly judgment threshold is set by those skilled in the art based on actual conditions. For example, if the difference between the operating speed of the production equipment in a certain production process and the set standard operating speed exceeds the maximum allowable speed deviation judgment threshold, it indicates that the production equipment speed is abnormal. This enables timely alarms, facilitates rapid repair of the production equipment, and thus ensures production efficiency.

[0026] Step four of the methods for managing production equipment anomalies also includes:

[0027] If it is determined that there is no abnormality in the production equipment, a first risk coefficient is generated based on the operating parameters of the production equipment, and a second risk coefficient is generated based on the process parameters. The first risk coefficient and the second risk coefficient are weighted and summed to obtain a risk score. When the risk score is greater than the predetermined risk score judgment threshold, it is determined that there is a potential abnormal risk in the production equipment.

[0028] The method for generating the first risk coefficient is as follows: under the condition that there is no abnormality in the production equipment, under the time series t, generate the actual parameter change function based on the actual operating parameters of the production equipment, and generate the standard parameter change function based on the standard operating parameters of the production equipment; obtain the historical operating parameters of the production equipment, generate the historical parameter change function based on the historical operating parameters of the production equipment under normal operation, and calculate the integral values ​​of the actual parameter change function and the standard parameter change function, and the actual parameter change function and the historical parameter change function under the time series t respectively. The deviation value of the operating parameters is obtained by the difference between the actual operating parameters and the corresponding standard operating parameters. The deviation parameter change function is generated based on the deviation value of the operating parameters under the time series t. The first change coefficient is obtained by the change of the deviation parameter change function within the time series t. The first risk value is obtained by combining the first change coefficient and all integral values, and the first risk value is standardized to obtain the first risk coefficient. The method for generating the second risk coefficient is as follows: obtain the number of product quality abnormalities that occur in the production process segment of the corresponding production equipment within the time series t, and calculate the standard deviation of the process parameters based on the actual process parameters of the corresponding production process segment within the time series t. The deviation value of the process parameters is calculated based on the difference between the actual process parameters and the standard process parameters. A process deviation change function is generated based on the process parameter deviation value under the time series t. The second change coefficient is obtained based on the change of the process deviation change function within the time series t. The second risk value is obtained by combining the number of product quality anomalies, the standard deviation of process parameters, and the second variation coefficient. The second risk value is then standardized to obtain the second variation coefficient.

[0029] The above technical solution provides another method for judging equipment anomaly management, namely, assessing the potential anomaly risks of production equipment. First, a first risk coefficient is obtained. Assuming no equipment anomalies are detected, under time series t, a function is generated based on the actual operating parameters of the production equipment, showing how the actual operating parameters change over time, and this function is labeled as the actual parameter change function. A function is also generated based on the standard operating parameters of the production equipment, showing how the standard operating parameters change over time, and this function is labeled as the standard parameter change function. Historical operating parameters of the production equipment are obtained. Based on these historical operating parameters under normal operating conditions, a function is generated showing how the historical operating parameters change over time, and this function is labeled as the historical parameter change function. The integral values ​​of the actual parameter change function and the standard parameter change function, as well as the integral values ​​of the actual parameter change function and the historical parameter change function, are calculated under time series t. Simultaneously, the operating parameter deviation value is calculated based on the difference between the actual operating parameters and the corresponding standard operating parameters. A function is generated based on the operating parameter deviation value under time series t, showing how the operating parameter deviation value changes over time, and this function is labeled as the deviation parameter change function. The first change coefficient is obtained based on the change of the deviation parameter change function within time series t. The first change coefficient and all integral values ​​are then combined to obtain the first risk value. This first risk value is then standardized to obtain the first risk coefficient. Specifically, the actual parameter change function can be set as follows: The standard parameter variation function is set to The historical parameter change function is set to The deviation value of the operating parameters, calculated from the difference between the actual operating parameters and the corresponding standard operating parameters, is set as follows: If the difference is calculated as an absolute value difference, then the deviation parameter variation function can be set as follows: Meanwhile, the first coefficient of change is set to The first risk value is set as ; First coefficient of change The expression is: ; First risk value The expression is: ; In the formula, Let t be the start time of the time series. t represents the end time of the time series t.

[0030] From mathematical expressions It can be seen that when When the value is large, it indicates that the deviation of the operating parameters of the production equipment within the time series t shows an increasing trend. The larger the value, the greater the possibility of potential anomalies in the production equipment, and vice versa. When this occurs, it indicates that the deviation of the operating parameter within the time series t does not show an increasing trend, so its value is set to 0; similarly, from the mathematical expression... It can be seen that, This represents the difference between the actual operating parameters and the standard operating parameters obtained within the time series t. This represents the difference between the actual operating parameters obtained within the time series t and the historical operating parameters. The larger the value of both, the greater the possibility of potential equipment anomalies. Finally, the first risk value is calculated by integrating the first change coefficient. It can be seen that the larger the value of the first risk value, the greater the potential for abnormalities in the production equipment; finally, the first risk value is standardized to obtain the first risk coefficient. .

[0031] Then, the second risk coefficient is obtained. This involves acquiring the number of product quality anomalies occurring in the corresponding production process segment within a time series t, and calculating the standard deviation of the process parameters based on the actual process parameters of the corresponding production process segment within time series t. The process parameter deviation value is then calculated based on the difference between the actual process parameters and the standard process parameters. A process deviation change function is generated based on the process parameter deviation value under time series t, and the second change coefficient is derived based on the change of the process deviation change function within time series t. Finally, the number of product quality anomalies, the standard deviation of the process parameters, and the second change coefficient are combined to obtain the second risk value. This second risk value is then standardized to obtain the second change coefficient. Specifically, the number of product quality anomalies is set as... Set the standard deviation of the process parameters to The deviation value of the process parameters calculated based on the difference between the actual process parameters and the standard process parameters is set as follows: The second coefficient of change is obtained using the same calculation method as the first coefficient of change. The second risk value is set as ; Second risk value The expression is: ; In the formula, The anomaly count comparison value is set based on historical data of the number of product quality anomalies. For the proposed standard deviation comparison value, These are the proposed second coefficient of change comparison values. The magnitude of these comparison values ​​can be determined manually and used for dimensionless measurement and weight adjustment.

[0032] Since abnormalities in production equipment may also lead to abnormalities in the corresponding product process parameters, data related to the process parameters of the corresponding production equipment's production process segment within a time series t are analyzed. The more product quality abnormalities appear within time series t, the greater the likelihood of potential abnormalities in the corresponding production equipment; a larger standard deviation indicates greater fluctuations in process parameters. A larger second coefficient of variation indicates a greater trend of process parameters deviating from standard process parameters, thus increasing the likelihood of potential abnormalities in the production equipment at the corresponding process stage. Therefore, a higher second risk value... The larger the value, the greater the potential risk of abnormality in the production equipment; finally, the second risk value is standardized to obtain the second risk coefficient. .

[0033] After deriving the first risk coefficient And after the second risk factor The risk score is obtained by summing the results. Risk scoring is available This indicates that the risk is due to considerations regarding production equipment. Weight is higher than Therefore, The weight is , The weight is When the risk score exceeds a predetermined risk score judgment threshold (set by a person skilled in the art based on the actual situation), it indicates a potential abnormal risk in the equipment, and a warning signal is issued. In this way, a first variation coefficient is derived by comprehensively analyzing the deviations between the actual and standard operating parameters of the production equipment, the deviations between the actual and historical operating parameters, and the changing trends of the actual operating parameters. A second risk coefficient is derived by comprehensively analyzing the changing trends and fluctuations of the process parameters under the corresponding production equipment, as well as the number of product anomalies. Combining the first and second risk coefficients allows for a more accurate assessment of potential abnormal risks in the production equipment, enabling early warning and handling, reducing the occurrence of production equipment anomalies, and ensuring product production efficiency.

[0034] The methods for managing raw material anomalies in step four mainly include managing raw material procurement, managing raw material inspection, and managing raw material inventory.

[0035] The system manages raw material procurement by automatically generating purchase orders based on the production plan, linking them to a list of qualified suppliers, allowing procurement personnel to submit purchase requests online, and generating formal orders that are then sent to suppliers after the approval process. Simultaneously, the system tracks order progress in real time, including supplier order acceptance status, delivery time, and logistics information, and issues alerts for any abnormalities. Raw material inspection management: After the raw materials arrive, the inspectors enter the basic information of the raw material batch and quantity through the system, generate an inspection task sheet according to the preset inspection standards, and record various data during the inspection process; determine whether the inspection results of each data are qualified, enter qualified raw materials into the inventory, and trigger the abnormal handling process for unqualified raw materials, while retaining the inspection records for traceability. Manage raw material inventory: Monitor raw material inventory levels in real time and set inventory warning thresholds. When the raw material inventory level is lower than the warning threshold, it is determined that there is an anomaly in the raw material inventory.

[0036] The above technical solution provides specific methods for managing raw materials, including raw material procurement management, raw material inspection management, and raw material inventory management. Raw material procurement management primarily involves: automatically generating purchase orders based on production plans, linking them to a list of qualified suppliers, allowing procurement personnel to submit purchase requests online, and generating formal orders that are then sent to suppliers after approval. Simultaneously, order progress is tracked in real time, including supplier order acceptance status, delivery time, and logistics information, enabling procurement personnel to promptly grasp the arrival status of raw materials and triggering alarms in case of any abnormalities. Then, raw material inspection management involves: after raw materials arrive, inspection personnel record data through the system... The system inputs basic information such as raw material batch and quantity, generates inspection task sheets based on preset inspection standards (such as water source quality indicators, packaging material hygiene standards, etc.), and records various data during the inspection process (such as pH value, mineral content, microbial indicators, etc.). The system automatically determines whether the inspection results are qualified. Qualified raw materials are entered into the inventory, and unqualified raw materials trigger an abnormal handling process (such as return or exchange), and the inspection records are retained for traceability. Finally, the system manages the raw material inventory: it monitors the raw material inventory quantity in real time, sets an inventory warning threshold, and automatically issues a warning message when the raw material inventory is lower than the threshold, indicating an abnormality in the raw material inventory. At this time, it reminds the purchasing personnel to replenish the stock in time.

[0037] The method for developing personalized response strategies based on management results in step five is as follows: The method for determining the replenishment quantity of raw materials when there is an anomaly is as follows: A mathematical model for replenishment quantity is constructed, and the replenishment quantity is generated based on this model; raw material management data is extracted from the dataset, and fixed procurement time points are obtained according to the pre-set procurement plan in the raw material management data; when the raw material inventory is less than the warning threshold, the current raw material inventory and the warning threshold are obtained; the replenishment quantity between the last procurement time point and the current procurement time point, and the replenishment quantity between the current procurement time point and the next procurement time point are obtained from the raw material management data; sales impact information at the current procurement time point is obtained, and influencing factors are determined based on this information; the current raw material inventory, the warning threshold, the replenishment quantity between the last procurement time point and the current procurement time point, the replenishment quantity between the current procurement time point and the next procurement time point, and the influencing factors are input into the replenishment quantity mathematical model to obtain the raw material replenishment quantity. The method for determining the influencing factor is as follows: Sales impact information is obtained, which refers to relevant information that can significantly affect mineral water sales. This includes the feature impact information corresponding to different key characteristics. The obtained sales impact information is input into a trained key feature impact scoring model to obtain the impact score of each key feature. The key feature impact scoring model is a deep neural network model. Each key feature is assigned a corresponding weight coefficient, and the impact scores of each key feature are weighted and summed to obtain the influencing factor. For example, if the key features are pre-defined as promotional activity impact features, weather impact features, market activity impact features, and social activity impact features, then the obtained sales impact information includes the impact of promotional activities. The key feature impact features are divided into three categories: promotional information, weather information, market activity information, and social activity information. Promotional information includes offline supermarket promotions such as "buy two get one free" for a certain brand of mineral water, and online e-commerce platform discounts. Weather information includes temperature and rainfall information. Market activity information includes offline promotional activities by the mineral water brand, and social activity information includes information on local competitions. The different feature impact information is then preprocessed (i.e., vectorized) and input into the trained key feature impact scoring model to obtain the impact score of each key feature. Finally, the impact factors are calculated by weighting the data.

[0038] The method for graded alarms when production equipment malfunctions is as follows: when the risk score is greater than the predetermined risk score judgment threshold, the location of the malfunction is determined and a first-level alarm signal is generated; when the difference between the actual operating parameters and the standard operating parameters is greater than the preset equipment malfunction judgment threshold, the location of the malfunction is determined and a second-level alarm signal is generated, wherein the urgency of the second-level alarm signal is higher than that of the first-level alarm signal. When product quality is abnormal, the method for adjusting process parameters is as follows: when the difference between the actual process parameters and the standard process parameters is greater than the preset product quality judgment threshold, the difference is recorded as the difference value. When the actual process parameters are higher than the standard process parameters, the process parameters are lowered according to the corresponding difference value. When the actual process parameters are lower than the standard process parameters, the process parameters are raised according to the corresponding difference value.

[0039] The above mathematical model for replenishment quantity is as follows: ,in, To obtain the replenishment quantity between the last purchase time and the current purchase time from the most recent replenishment data, This is a comparison of the replenishment quantity between the pre-set purchase time and the current purchase time. This refers to the replenishment quantity between the current purchase point and the next purchase time point, obtained from accumulated historical replenishment data. This is a pre-set comparison value for the replenishment quantity between the current procurement point and the next procurement time point. The comparison value can be pre-set manually based on actual experience, and is used to adjust the weights and remove dimensions. The impact factor; as can be seen from the formula, This indicates the recent trend in replenishment volume. This reflects the future replenishment trend, and the influencing factor reflects the impact of recent activities on mineral water sales. Therefore, when replenishment is needed, it is not only based on the actual shortage of replenishment quantity, but also on a comprehensive analysis of recent replenishment trends, future replenishment trends, and the impact of recent activities to adjust the final raw material replenishment quantity, making the replenishment quantity more accurate, thereby effectively reducing the occurrence of raw material shortages or stockpiles. The method for tiered alarms when production equipment malfunctions is as follows: when the risk score is greater than the predetermined risk score judgment threshold, the location of the malfunction is determined and a first-level alarm signal is generated; when the difference between the actual operating parameters and the standard operating parameters is greater than the preset equipment malfunction judgment threshold, the location of the malfunction is determined and a second-level alarm signal is generated, wherein the urgency of the second-level alarm signal is higher than that of the first-level alarm signal; in this way, the risk situation and the urgency of maintenance can be determined based on the generated alarm signals, which is convenient for managers to respond and handle. Finally, the method for adjusting process parameters when product quality is abnormal is as follows: When the difference between the actual process parameter and the standard process parameter is greater than the preset product quality judgment threshold, the difference is recorded as the difference value. When the actual process parameter is higher than the standard process parameter, the process parameter is lowered by the corresponding difference value; when the actual process parameter is lower than the standard process parameter, the process parameter is increased by the corresponding difference value. For example, if the standard process parameter for sterilization temperature in a process is 90 degrees Celsius, and the displayed sterilization temperature is 85 degrees Celsius, the difference value is 5 degrees Celsius. Therefore, the temperature is increased by 5 degrees Celsius to ensure that the process parameters for each production process are within a reasonable range.

[0040] The method for channel switching based on network data parameters in step three is as follows: the channel includes the main channel and the secondary channel. Multiple network data parameters of the main channel are acquired in real time. Each network data parameter has a corresponding standard operating data range. When the network data parameters acquired in real time are not within the set standard operating data range, the transmission of the main channel is switched to the secondary channel. When the network data parameters are all restored to the standard operating data range, the transmission of the main channel is switched back. When the real-time acquired network data parameters are within the set standard operating data range, the optimal operating data value for each network data parameter is determined based on the standard operating data range for each network data parameter. The quality score for each network data parameter is derived based on the difference between the real-time acquired network data parameter and its corresponding optimal operating data value. After standardizing the quality scores of each network data parameter, a weighted summation is performed to obtain the total quality score. A Cartesian coordinate system is constructed with time as the x-axis and the total quality score as the y-axis. A curve of the total quality score is constructed in the Cartesian coordinate system, and the area enclosed by the total quality score curve and the x-axis within a time period T is extracted. The standard deviation of the score within time T is calculated, and the area is multiplied by the standard deviation to obtain the scoring coefficient. When the scoring coefficient exceeds a preset scoring coefficient threshold, the main channel transmission is switched to the secondary channel transmission.

[0041] The above scheme provides a specific method for channel switching. First, multiple network data parameters of the main channel are acquired in real time. Each network data parameter has a corresponding standard operating data range. When the acquired network data parameters are outside the set standard operating data range, the main channel transmission is switched to the secondary channel transmission. When the main channel network data parameters recover to the standard operating data range, the transmission is switched back to the main channel. This allows for real-time monitoring of the quality of each network data parameter, enabling timely channel switching to ensure transmission quality upon detection of anomalies. When the acquired network data parameters are within the set standard operating data range, the optimal operating data value is determined based on the standard operating data range for each network data parameter. The optimal operating data value for each network data parameter is taken from the center value of the standard operating data range. The quality score R for each network data parameter is calculated based on the difference between the acquired network data parameter and the optimal operating data value. After standardization of the quality scores for each network data parameter, a weighted average is performed to obtain the total quality score CR. A Cartesian coordinate system is constructed with time as the x-axis and the total quality score as the y-axis. A curve of the total quality score is then plotted in the Cartesian coordinate system. The area enclosed by the total quality score curve and the x-axis within a time period T is calculated. And calculate the standard deviation of the scores over time T. The scoring coefficient is obtained by multiplying the area by the natural index of the standard deviation of the score. , When the scoring coefficient exceeds a preset scoring coefficient threshold, the main channel transmission is switched to the secondary channel transmission. The scoring coefficient threshold is set by those skilled in the art based on actual conditions. In a Cartesian coordinate system, the area enclosed by the total quality scoring curve and the x-axis represents the deviation of network data parameters from the optimal operating data value within time T. The smaller the enclosed area, the closer the network data parameters are to the optimal operating data value; conversely, the larger the enclosed area, the further the network data parameters deviate from the optimal operating data value within time T. This indicates that the network quality of the main channel is gradually deteriorating. Similarly, the larger the scoring standard deviation, the greater the fluctuation of network data parameters within time T. Therefore, a larger scoring coefficient indicates that the network quality of the main channel is gradually deteriorating. To ensure data transmission quality, a scoring coefficient threshold is set. When the scoring coefficient exceeds the preset scoring coefficient threshold, the main channel transmission is switched to the secondary channel transmission. This method allows for channel switching before network quality deteriorates, making channel switching more sensitive and ensuring network transmission quality and stability.

[0042] This invention can comprehensively analyze the deviations between the actual and standard operating parameters of production equipment, the deviations between the actual and historical operating parameters, and the changing trends of the actual operating parameters to derive a first variation coefficient. It then comprehensively analyzes the changing trends and fluctuations of the process parameters under the corresponding production equipment, as well as the number of product anomalies, to derive a second risk coefficient. Combining the first and second risk coefficients allows for a more accurate assessment of potential anomaly risks in production equipment, enabling early warning and handling, reducing the occurrence of equipment anomalies, and ensuring product production efficiency. When raw material replenishment is required, this invention not only replenishes based on the actual shortage quantity but also considers recent replenishment trends, future replenishment trends, and the impact of recent activities to adjust the final raw material replenishment quantity, making the replenishment quantity more accurate and effectively reducing the occurrence of raw material shortages or stockpiles.

[0043] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0044] All formulas in this manual are dimensionless and calculated numerically. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0045] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for managing the production process of mineral water, characterized in that, The methods include: Step 1: Collect relevant data in the mineral water production process. The relevant data includes raw material management data, process parameters, and production equipment operating parameters. Step 2: Preprocess the associated data to form a dataset, and use dual-channel network transmission technology for data transmission; Step 3: Monitor network data parameters in real time during transmission and perform channel conversion based on the network data parameters; Step 4: Process and analyze the data within the dataset to manage the mineral water production process accordingly. This management includes raw material anomaly management, production equipment anomaly management, and product quality anomaly management. Step 5: Develop and implement personalized response strategies based on management results. Personalized response strategies include determining the quantity of raw materials to replenish when raw materials are abnormal, issuing tiered alarms when production equipment is abnormal, and adjusting process parameters when product quality is abnormal.

2. The mineral water production process management method according to claim 1, characterized in that, The method for managing product quality anomalies in step four is as follows: The production process of mineral water is divided into multiple production steps, and each production step has corresponding standard process parameters. The actual process parameters corresponding to each production process segment are extracted from the dataset. The actual process parameters are compared with the corresponding standard process parameters. When the difference between the actual process parameters and the standard process parameters is greater than the preset product quality judgment threshold, the product is judged to have a quality abnormality.

3. The mineral water production process management method according to claim 2, characterized in that, The method for managing production equipment anomalies in step four is as follows: Each production process segment has corresponding standard operating parameters for the production equipment. The actual operating parameters of the corresponding production equipment in each production process segment are extracted from the dataset. The actual operating parameters are compared with the corresponding standard operating parameters. When the difference between the actual operating parameters and the standard operating parameters is greater than the preset equipment anomaly judgment threshold, the production equipment is judged to be abnormal.

4. The mineral water production process management method according to claim 3, characterized in that, Step four of the methods for managing production equipment anomalies also includes: If it is determined that there is no abnormality in the production equipment, a first risk coefficient is generated based on the operating parameters of the production equipment, and a second risk coefficient is generated based on the process parameters. The first risk coefficient and the second risk coefficient are weighted and summed to obtain a risk score. When the risk score is greater than the predetermined risk score judgment threshold, it is determined that there is a potential abnormal risk in the production equipment.

5. The mineral water production process management method according to claim 4, characterized in that, The method for generating the first risk coefficient is as follows: Assuming there are no abnormalities in the production equipment, under time series t, an actual parameter change function is generated based on the actual operating parameters of the production equipment, and a standard parameter change function is generated based on the standard operating parameters of the production equipment; historical operating parameters of the production equipment are obtained, and historical parameter change functions are generated based on the historical operating parameters of the production equipment under normal operation; the integral values ​​of the actual parameter change function and the standard parameter change function, and the actual parameter change function and the historical parameter change function under time series t are calculated respectively. The deviation value of the operating parameters is obtained by the difference between the actual operating parameters and the corresponding standard operating parameters. The deviation parameter change function is generated based on the deviation value of the operating parameters under the time series t. The first change coefficient is obtained by the change of the deviation parameter change function within the time series t. The first risk value is obtained by combining the first change coefficient and all integral values, and the first risk value is then standardized to obtain the first risk coefficient.

6. The mineral water production process management method according to claim 5, characterized in that, The method for generating the second risk coefficient is as follows: Obtain the number of product quality anomalies that occur in the production process segment of the corresponding production equipment within the time series t, and calculate the standard deviation of the process parameters based on the actual process parameters of the corresponding production process segment within the time series t; The deviation value of the process parameters is calculated based on the difference between the actual process parameters and the standard process parameters. A process deviation change function is generated based on the process parameter deviation value under the time series t. The second change coefficient is obtained based on the change of the process deviation change function within the time series t. The second risk value is obtained by combining the number of product quality anomalies, the standard deviation of process parameters, and the second variation coefficient. The second risk value is then standardized to obtain the second variation coefficient.

7. A method for managing the production process of mineral water according to claim 6, characterized in that, The method for managing raw material anomalies in step four is as follows: Real-time monitoring of raw material inventory levels and setting inventory warning thresholds. When the raw material inventory level is lower than the warning threshold, it is determined that there is an anomaly in the raw material inventory.

8. A method for managing the production process of mineral water according to claim 7, characterized in that, The method for developing personalized response strategies based on management results in step five is as follows: The method for determining the replenishment quantity of raw materials when raw materials are abnormal is as follows: construct a mathematical model of replenishment quantity, generate the raw material replenishment quantity based on the mathematical model of replenishment quantity; extract raw material management data from the dataset, and obtain the fixed procurement time point according to the preset procurement plan in the raw material management data; When the raw material inventory is less than the warning threshold, obtain the current raw material inventory and the warning threshold. Obtain the replenishment quantity between the last purchase time and the current purchase time from the raw material management data, and the replenishment quantity between the current purchase time and the next purchase time. Obtain sales impact information at the current procurement time point, and determine influencing factors based on the sales impact information; input the current raw material inventory, early warning threshold, replenishment quantity between the last procurement time point and the current procurement time point, replenishment quantity between the current procurement time point and the next procurement time point, and influencing factors into the replenishment quantity mathematical model to obtain the raw material replenishment quantity; The method for graded alarms when production equipment malfunctions is as follows: when the risk score is greater than the predetermined risk score judgment threshold, the location of the malfunction is determined and a first-level alarm signal is generated; when the difference between the actual operating parameters and the standard operating parameters is greater than the preset equipment malfunction judgment threshold, the location of the malfunction is determined and a second-level alarm signal is generated, wherein the urgency of the second-level alarm signal is higher than that of the first-level alarm signal. When product quality is abnormal, the method for adjusting process parameters is as follows: when the difference between the actual process parameters and the standard process parameters is greater than the preset product quality judgment threshold, the difference is recorded as the difference value. When the actual process parameters are higher than the standard process parameters, the process parameters are lowered according to the corresponding difference value. When the actual process parameters are lower than the standard process parameters, the process parameters are raised according to the corresponding difference value.

9. A method for managing the production process of mineral water according to claim 8, characterized in that, The method for determining the impact factor is as follows: Acquire sales impact information, which includes feature impact information corresponding to different key features. Input the acquired sales impact information into a trained key feature impact scoring model to obtain the impact score of each key feature. The key feature impact scoring model is a deep neural network model. Each key feature is assigned a corresponding weight coefficient, and the impact scores of each key feature are weighted and summed to obtain the impact factor.

10. A method for managing the production process of mineral water according to claim 9, characterized in that, The method for channel conversion based on network data parameters in step three is as follows: The channel includes a main channel and a secondary channel. Multiple network data parameters of the main channel are acquired in real time. Each network data parameter has a corresponding standard operating data range. When the acquired network data parameter is not within the set standard operating data range, the transmission is switched from the main channel to the secondary channel. When the network data parameters are all restored to the standard operating data range, the transmission is switched back to the main channel. When the real-time acquired network data parameters are within the set standard operating data range, the optimal operating data value for each network data parameter is determined based on the standard operating data range for each network data parameter. The quality score for each network data parameter is derived based on the difference between the real-time acquired network data parameter and its corresponding optimal operating data value. After standardizing the quality scores of each network data parameter, a weighted summation is performed to obtain the total quality score. A Cartesian coordinate system is constructed with time as the x-axis and the total quality score as the y-axis. A curve of the total quality score is constructed in the Cartesian coordinate system, and the area enclosed by the total quality score curve and the x-axis within a time period T is extracted. The standard deviation of the score within time T is calculated, and the area is multiplied by the standard deviation to obtain the scoring coefficient. When the scoring coefficient exceeds a preset scoring coefficient threshold, the main channel transmission is switched to the secondary channel transmission.

Images