Material replenishment methods, devices, equipment, and storage media for beverage robots

Abstract

This invention relates to the field of replenishment calculation, and more particularly to a material replenishment method, apparatus, equipment, and storage medium for a beverage robot. The method includes the following steps: collecting real-time sales logs; performing rolling forecasts based on the real-time sales logs to generate short-term sales forecasts for different beverages; calculating the material consumption rate based on the short-term sales forecasts to generate a material consumption dataset; calculating the maximum dispensing capacity for each beverage based on the material consumption dataset to generate a dispensing capacity assessment result; performing multi-material collaborative replenishment analysis based on the dispensing capacity assessment result, and outputting a replenishment report to be sent to the beverage robot for background replenishment processing. This invention achieves proactive and automated replenishment decisions for beverage robots, reducing the risk of material shortages, minimizing material waste, and ensuring continuous and stable dispensing capacity.

Description

Technical Field

[0001] This invention relates to the field of replenishment calculation, and in particular to a material replenishment method, apparatus, equipment and storage medium for a beverage robot. Background Technology

[0002] During the long-term operation of beverage robots, sales of various beverages are affected by a variety of factors, including time of day, customer traffic, weather changes, and promotional activities, exhibiting significant fluctuations and uncertainties. Different beverages consume significantly different types and quantities of materials, and multiple materials often participate in the preparation of a single drink. If the inventory of a key material is insufficient, the beverage may not be dispensed properly, or even the equipment may shut down. Delayed replenishment can easily lead to order failures and service interruptions; conversely, excessive replenishment increases inventory pressure, resulting in expired materials, waste, and other adverse consequences, severely impacting the overall operational efficiency of the beverage robot. Existing beverage robot material replenishment methods mostly rely on preset inventory thresholds, fixed replenishment cycles, or human experience, lacking in-depth analysis and forecasting capabilities based on real-time sales data. These methods typically only trigger replenishment reminders when inventory is close to or below a safety threshold, exhibiting significant lag and failing to respond promptly to sudden sales increases or changes in sales structure. Existing replenishment strategies are mostly managed on a single material basis, failing to comprehensively consider the synergistic consumption relationships between multiple materials. This can easily lead to situations where some materials are replenished sufficiently, but the overall cup-making capacity is still limited. The efficiency and accuracy of replenishment need to be further improved. Summary of the Invention

[0003] To address the aforementioned technical problems, this invention proposes a material replenishment method, apparatus, equipment, and storage medium for a beverage robot, thereby resolving at least one of the aforementioned technical problems.

[0004] To achieve the above objectives, the present invention provides a material replenishment method for a beverage robot, comprising the following steps: Step S1: Collect real-time sales logs; perform rolling forecasts based on real-time sales logs to generate short-term sales forecasts for different beverages; Step S2: Calculate the material consumption rate based on the short-term sales forecast and generate a material consumption dataset; Step S3: Calculate the maximum dispensing capacity for each beverage based on the material consumption dataset, and generate a dispensing capacity assessment result; Step S4: Based on the cup dispensing assessment results, perform multi-material collaborative replenishment analysis, output a replenishment report and send it to the beverage robot for background replenishment processing.

[0005] This specification provides a material replenishment device for a beverage robot, used to execute the material replenishment method for the beverage robot as described above, including: The data acquisition unit is used to collect real-time sales logs; based on the real-time sales logs, it performs rolling forecasts and generates short-term sales forecasts for different beverages. The material calculation unit is used to calculate the consumption rate based on the short-term sales forecast and generate a material consumption dataset. The dispensing evaluation unit is used to calculate the maximum dispensing capacity for each beverage based on the material consumption dataset and generate dispensing evaluation results. The replenishment unit is used to perform multi-material collaborative replenishment analysis based on the cup dispensing assessment results, and output a replenishment report to be sent to the beverage robot for back-end replenishment processing.

[0006] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the material replenishment method for the beverage robot described in any of the above claims.

[0007] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the material replenishment method for the beverage robot described in any of the preceding claims.

[0008] The beneficial effects of this invention are specifically as follows: It continuously and dynamically grasps the actual sales trends of various beverages within the current time window. Compared to static forecasting methods based on fixed historical periods, this rolling forecasting mechanism can continuously correct forecast results as sales data is updated in real time, thus effectively addressing complex scenarios such as peak periods, sudden order increases, or rapid sales declines. It breaks down abstract beverage sales demand into the consumption volume of specific raw materials, enabling a clear understanding of the consumption rate and intensity of various materials within the forecast period. It accurately assesses the actual dispensing capacity of the beverage robot from the perspective of the overall material combination. By comprehensively considering the inventory status and consumption rate of multiple raw materials, it avoids biases caused by judging dispensing capacity solely based on the inventory of a single material, thereby accurately identifying bottleneck materials that limit the continuous dispensing of a particular beverage. Through the dispensing capacity assessment results, potential beverage supply disruption risks can be identified in advance, providing early warnings for replenishment decisions and operational scheduling. A replenishment report is generated and sent to the beverage robot's backend for replenishment processing, upgrading replenishment decisions from single-material replenishment to collaborative optimization of overall dispensing capacity. This step comprehensively considers the impact of different materials on the dispensing capacity of various beverages, avoiding ineffective replenishment situations where a single material is adequately replenished but the overall dispensing capacity remains limited. By automatically generating replenishment reports and sending them to the backend, the replenishment process is automated and controlled in a closed loop. This not only reduces the cost of manual intervention but also significantly improves the accuracy and timeliness of replenishment decisions, thereby ensuring the beverage robot operates continuously and efficiently within the predicted sales period. Attached Figure Description

[0009] Figure 1 This is a schematic diagram of the steps of a material replenishment method for a beverage robot according to the present invention; Figure 2 This is a detailed flowchart illustrating the implementation steps of step S1. Figure 3 This is a flowchart illustrating the detailed implementation steps of step S2. Detailed Implementation

[0010] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0011] This application provides a material replenishment method, apparatus, equipment, and storage medium for a beverage robot. The execution entities of the material replenishment method, apparatus, equipment, and storage medium for the beverage robot include, but are not limited to, mechanical equipment, data processing platforms, cloud server nodes, and network upload devices mounted on the system, which can be considered as general computing nodes in this application. The data processing platform includes, but is not limited to, at least one of an audio-visual management system, an information management system, and a cloud-based data management system.

[0012] Please see Figures 1 to 3 This invention provides a material replenishment method for a beverage robot, comprising the following steps: Step S1: Collect real-time sales logs; perform rolling forecasts based on real-time sales logs to generate short-term sales forecasts for different beverages; Step S2: Calculate the material consumption rate based on the short-term sales forecast and generate a material consumption dataset; Step S3: Calculate the maximum dispensing capacity for each beverage based on the material consumption dataset, and generate a dispensing capacity assessment result; Step S4: Based on the cup dispensing assessment results, perform multi-material collaborative replenishment analysis, output a replenishment report and send it to the beverage robot for background replenishment processing.

[0013] In the embodiments of the present invention, see Figure 1 This is a flowchart illustrating the steps of a material replenishment method for a beverage robot according to the present invention. In this example, the steps of the material replenishment method for the beverage robot include: Step S1: Collect real-time sales logs; perform rolling forecasts based on real-time sales logs to generate short-term sales forecasts for different beverages; In this embodiment, real-time sales log data generated by the self-service beverage robot during actual operation is collected. This sales log data includes the precise timestamp of order generation, beverage category identification code, cup size specifications, completion time, payment method, order interval duration, and order density fluctuations within a unit time period. The collected sales log data is processed into a time series, establishing a sales event flow in chronological order. A sliding time window mechanism is set, with the time window configured as an adjustable multi-scale mode, including a short-cycle 5-minute window, a medium-cycle 15-minute window, and a long-cycle 60-minute window, to capture sales variation patterns at different time granularities. Within each time window, beverage sales data is statistically aggregated to calculate the real-time sales volume of various beverages such as Americano, Latte, Cappuccino, Mocha, and Matcha Latte. A time-series sales dataset for each beverage category is constructed, recording the sales value for each time window over the past 24 hours. An ARIMA autoregressive moving average model is introduced as the basic prediction framework, with model parameters p=3 (autoregressive term order), d=1 (difference order), and q=2 (moving average term order). The system acquires current environmental factors, including real-time temperature data, weather condition classification (sunny, cloudy, rainy, snowy), relative humidity percentage, and current time period (morning peak 7:00-9:00, midday 11:30-13:30, afternoon tea 14:30-16:30, evening peak 17:30-19:30). It also extracts historical sales data from the past 30 days under similar weather conditions during the same time period from the historical sales database and performs pattern matching analysis. Finally, it uses multiple linear regression to establish a correlation between weather factors and sales volume. The correlation model between the two factors is used to fit the regression coefficients using the least squares method. The ARIMA time series forecast results and the multivariate regression correction results are weighted and fused, with the time series forecast weight set at 0.7 and the external factor correction weight set at 0.3. Rolling forecasts are made for multiple future time windows (future 30 minutes, future 60 minutes, future 120 minutes) to generate short-term sales forecasts for different beverages in each forecast period. The structured forecast results are output, including the predicted sales volume of each beverage category, the prediction confidence interval (set at 95% confidence level), and the prediction error estimation range.

[0014] Step S2: Calculate the material consumption rate based on the short-term sales forecast and generate a material consumption dataset; In this embodiment, a mapping relationship between beverage categories and material requirements is established; a beverage recipe database is constructed, recording in detail the standard preparation recipe for each beverage; the Americano recipe is analyzed, revealing that a single cup requires 30 ml of espresso (corresponding to 7 g of coffee beans), 120 ml of hot water, and one paper cup; the latte recipe is analyzed, revealing that a single cup requires 30 ml of espresso (corresponding to 7 g of coffee beans), 200 ml of milk, 50 ml of milk foam, and one paper cup; the cappuccino recipe is analyzed, revealing that a single cup requires 30 ml of espresso (corresponding to 7 g of coffee beans), 100 ml of milk, 150 ml of milk foam, and one paper cup; the mocha recipe is analyzed, revealing that a single cup requires 30 ml of espresso (corresponding to 7 g of coffee beans), 150 ml of milk, 20 ml of chocolate syrup, 30 ml of whipped cream, and one paper cup.

[0015] The recipe for a matcha latte was analyzed, revealing that a single cup requires 3 grams of matcha powder, 220 ml of milk, 10 ml of syrup, and one paper cup. The recipe information was then matrixed to construct a beverage-material consumption coefficient matrix. For a 30-minute forecast window, based on the predicted sales volume of each beverage, the predicted sales volume for Americano was calculated as follows: 18 cups; for latte, 25 cups; for cappuccino, 12 cups; for mocha, 8 cups; and for matcha latte, 15 cups. Based on the beverage-material consumption coefficient matrix, the cumulative consumption of each material was calculated. The predicted consumption of coffee beans in the next 30 minutes was calculated as (18 + 25 + 12 + 8) × 7 grams = 44 grams; the predicted consumption of milk in the next 30 minutes was calculated as 25 × 200 + 12 × 100 + 8 × 150 + 15 × 220 = 9700 ml. The predicted consumption of chocolate syrup in the next 30 minutes is calculated as 8 × 20 = 160 ml; the predicted consumption of matcha powder in the next 30 minutes is calculated as 15 × 3 = 45 g; the predicted consumption of paper cups in the next 30 minutes is calculated as 18 + 25 + 12 + 8 + 15 = 78. The same material consumption calculation is performed for the prediction windows of the next 60 minutes and the next 120 minutes. The consumption of each material is normalized over time, and the material consumption rate is calculated. The consumption rates of coffee beans are 882 g / hour, milk is 19.4 liters / hour, chocolate syrup is 320 ml / hour, matcha powder is 90 g / hour, and paper cups are 156 / hour. The material consumption dataset is then generated, including the unit time consumption rate of each material, the cumulative consumption in different time windows, and the trend of consumption rate changes.

[0016] Step S3: Calculate the maximum dispensing capacity for each beverage based on the material consumption dataset, and generate a dispensing capacity assessment result; In this embodiment, the real-time inventory status of the beverage robot is obtained; sensor data from each material container is read to determine the current remaining quantity: coffee beans 3500g, milk 15 liters, chocolate syrup 800ml, matcha powder 200g, paper cups 300, and hot water is plentiful and unlimited; the maximum number of cups that can be made is calculated separately for each beverage category; the maximum number of cups dispensed for Americano is calculated, and its material constraint factors, including coffee beans (7g per cup) and paper cups (7g per cup), are analyzed. (Consuming 1); Based on the coffee bean inventory, the maximum number of cups that can be made is 3500 ÷ 7 = 500 cups; Based on the paper cup inventory, the maximum number of cups that can be made is 300 cups; Taking the minimum of the two, we find that the maximum number of Americano cups that can be made under the current inventory constraints is 300 cups, with the key limiting material being the paper cup; For lattes, the maximum number of cups that can be made is calculated, and its material constraint factors include coffee beans (consuming 7 grams per cup), milk (consuming 200 ml per cup), and paper cups (consuming 1 per cup); Based on the coffee bean inventory, the maximum number of cups that can be made is 3500 ÷ 7 = 500 cups.

[0017] Based on milk inventory, the maximum number of cups that can be made is calculated to be 15000 ÷ 200 = 75 cups; based on paper cup inventory, the maximum number of cups that can be made is calculated to be 300 cups; taking the minimum of the three, we find that the maximum number of lattes that can be made under the current inventory constraints is 75 cups, with milk being the key limiting material; For cappuccino, the maximum number of cups that can be made is calculated, and its material constraint factors include coffee beans (7 grams per cup), milk (100 ml per cup for milk foam making), and paper cups (1 per cup); based on coffee bean inventory, the maximum number of cups that can be made is calculated to be 500 cups; based on milk inventory, the maximum number of cups that can be made is calculated to be 15000 ÷ 100 = 150 cups; based on paper cup inventory, the maximum number of cups that can be made is 3... 00 cups; taking the minimum of the three, we find that the maximum number of cappuccinos that can be made under the current inventory constraints is 150 cups, with milk being the key limiting material; for mocha, we calculate the maximum number of cups that can be made, analyzing its material constraint factors, which include coffee beans, milk, chocolate syrup, and paper cups; based on the coffee bean inventory, the maximum number of cups that can be made is calculated to be 500 cups; based on the milk inventory, the maximum number of cups that can be made is calculated to be 15000÷150=100 cups; based on the chocolate syrup inventory, the maximum number of cups that can be made is calculated to be 800÷20=40 cups; based on the paper cup inventory, the maximum number of cups that can be made is calculated to be 300 cups; taking the minimum of the four, we find that the maximum number of mochas that can be made under the current inventory constraints is 40 cups, with chocolate syrup being the key limiting material.

[0018] The maximum number of cups that can be made for matcha lattes is calculated, and the material constraints include matcha powder, milk, syrup, and paper cups. Based on the matcha powder inventory, the maximum number of cups that can be made is calculated to be 200 ÷ 3 = 66 cups; based on the milk inventory, the maximum number of cups that can be made is calculated to be 15000 ÷ 220 = 68 cups; based on the paper cup inventory, the maximum number of cups that can be made is calculated to be 300 cups. Taking the minimum value, the maximum number of matcha lattes that can be made under the current inventory constraints is 66 cups, with matcha powder being the key limiting material. Combining the material consumption rate prediction in step S2, the sustainable supply time for each beverage is calculated. The sustainable supply time for Americano coffee under the predicted sales volume (18 cups / half hour) is calculated to be 300 ÷ 18 × 0.5 = 8.3 hours. The sustainable supply time for latte is calculated to be... The sustainable supply time for the following beverages is calculated as follows: 75 ÷ 25 × 0.5 = 1.5 hours based on the projected sales volume (25 cups / half hour); 150 ÷ ​​12 × 0.5 = 6.25 hours for cappuccino (12 cups / half hour); 40 ÷ 8 × 0.5 = 2.5 hours for mocha (8 cups / half hour); and 66 ÷ 15 × 0.5 = 2.2 hours for matcha latte (15 cups / half hour). The system generates a production capacity assessment, outputting the maximum number of cups that can be made for each beverage, key limiting materials, the estimated time when inventory will be depleted, and indicating that lattes will be unavailable in as little as 1.5 hours.

[0019] Step S4: Based on the cup dispensing assessment results, perform multi-material collaborative replenishment analysis, output a replenishment report and send it to the beverage robot for background replenishment processing.

[0020] In this embodiment, the sustainable supply duration of each beverage is sorted and analyzed; the shortest sustainable supply duration for lattes is identified as 1.5 hours, with milk being the key limiting material; milk is marked as the material most likely to become scarce, and its replenishment priority is set to the highest level; the milk shortage point is calculated as the current time plus 1.5 hours, meaning that there is a risk of milk inventory running out at 15:30 in the afternoon; a service continuity guarantee mechanism is introduced, setting a minimum inventory safety threshold that can support 30 minutes of normal sales; based on the milk consumption rate of 19.4 liters / hour, the safety stock is calculated to be 19.4 × 0. 0.5 = 9.7 liters; The remaining 15 liters of milk are below the ideal inventory level, requiring a replenishment process. Simultaneously analyze the inventory status of other materials, identifying chocolate syrup with a sustainable supply time of 2.5 hours and matcha powder with a sustainable supply time of 2.2 hours, both of which require immediate replenishment. Conduct multi-material collaborative replenishment analysis to determine if multiple materials can be included in the same replenishment batch. Considering the labor and time costs of replenishment operations, set a replenishment batch merging strategy: when the overlap of replenishment time windows for multiple materials exceeds 70%, batch merging will be performed. Calculate the milk... The replenishment time windows for chocolate syrup and matcha powder overlap by 85%, meeting the consolidation criteria. Target replenishment quantities for each material are calculated, taking into account maximum container capacity, shelf-life constraints, and future sales forecasts. For milk, the maximum container capacity is 25 liters, with 15 liters currently remaining, leaving 10 liters available for replenishment. Based on sales forecasts for the next 4 hours, milk consumption is projected to be 19.4 × 4 = 77.6 liters, far exceeding the current inventory plus replenishment quantity. Considering the 48-hour shelf life of milk after opening, the replenishment quantity for a single transaction is set to not exceed the container's capacity. 90% of the container capacity, i.e., 22.5 liters; the target replenishment quantity for milk is determined to be 22.5 - 15 = 7.5 liters, rounded up to 8 liters (based on 1-liter packaging units); the replenishment quantity for chocolate syrup is calculated, the maximum container capacity is 1500 ml, and the current remaining quantity is 800 ml. Based on the estimated consumption of 1280 ml in the next 4 hours, the target replenishment quantity is 1500 - 800 = 700 ml, rounded up to 800 ml (based on 100 ml bottles, replenishing 8 bottles); the replenishment quantity for matcha powder is calculated, the maximum container capacity is 500 grams, and the current remaining quantity is 200 grams.

[0021] Based on the estimated consumption of 360 grams in the next 4 hours, the target replenishment quantity is 500-200=300 grams, and a replenishment of 300 grams is determined. An inventory assessment of the paper cups is conducted. Although there are currently 300 remaining, considering the estimated consumption of 624 in the next 4 hours, emergency replenishment needs to be initiated, with a target replenishment quantity of 1000 (replenishing 10 packs of 100). A structured replenishment report is generated, including a replenishment material list (8 liters of milk, 800 ml of chocolate syrup, 300 grams of matcha powder, and 1000 paper cups), a suggested replenishment time window (from the current moment to completion within the next hour), an urgency rating for each material (milk: Grade A, paper cups: Grade A, chocolate syrup: Grade B, matcha powder: Grade B), and the estimated sustainable operating time after replenishment.

[0022] The replenishment report is formatted and packaged to generate a JSON-formatted replenishment instruction data packet. This packet is then sent to the robot control center via the beverage robot's backend management system API. Upon receiving the replenishment instruction, the robot control center triggers a replenishment reminder mechanism, displaying a replenishment notification pop-up on the administrator's interface and simultaneously pushing a replenishment reminder message to the mobile device. After confirming the replenishment instruction, the administrator prepares the corresponding materials and activates the robot's replenishment mode. The robot enters a replenishment preparation state, suspends new order acceptance, and completes the remaining orders in the current production queue. The material storage door is opened, and a material loading interface provides guidance. The administrator sequentially completes the replenishment loading of milk, chocolate syrup, matcha powder, and paper cups. After replenishment, the robot automatically detects the increased material quantity using material sensors and updates the inventory database. The system automatically performs inventory verification, confirming that the deviation between the replenishment quantity of each material and the instruction requirements is within 5%. After completing the backend replenishment processing, the robot exits replenishment mode, resumes normal operation, and reopens the order acceptance function.

[0023] In this embodiment, see Figure 2 The diagram below illustrates the detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include: The beverage robot collects real-time sales logs, which include order generation time, coffee type, cup size, preparation time, and changes in order density per unit time. The real-time sales logs are timestamped and standardized to extract standardized sales data. Perform multi-beverage sales statistics on standardized sales data to generate sales volume per unit time for different beverages; Time-series smoothing filtering is applied to the sales volume per unit time to construct sales curves for multiple beverages; Rolling forecasts are generated based on the sales curves of multiple beverages to produce short-term sales forecasts for different beverages.

[0024] In this embodiment, the beverage robot continuously records user orders and beverage preparation activities during operation, forming continuous real-time sales log data. The real-time sales log includes at least order generation time, beverage category information, cup size specifications, beverage preparation completion time, and order density changes per unit time. Specifically, order generation time characterizes the temporal distribution of user demand; beverage category distinguishes the consumption type corresponding to different ingredient combinations; cup size specifications reflect the differences in ingredient usage per cup; beverage preparation completion time characterizes the cupping rhythm and preparation load; and order density changes per unit time describe the trend of sales activity over time. To address the issues of inconsistent time recording accuracy and timelines in the real-time sales log, order generation time and beverage preparation completion time are standardized with unified timestamps, mapping all time information to the same standard timeline. This standardization includes unified time format, unified time accuracy, and time window division, ensuring that each sales record can be accurately assigned to a preset standard time slice, such as using 1 minute or 5 minutes as the basic statistical unit. Simultaneously, the order density changes per unit time are recalculated to maintain a strict correspondence with the standard time slice.

[0025] Based on standardized sales data, order records are categorized and statistically analyzed according to beverage category. Sales volume for different beverages within a unit of time is calculated under a unified time scale. During the statistical process, a standard time slice is used as the statistical window to summarize the order quantity for each beverage within its corresponding time slice. The sales data is then converted based on cup size parameters to reflect differences in raw material consumption for different beverage sizes; for example, large cups are converted to high-weighted sales values, and small cups to low-weighted sales values. Due to the randomness and short-term fluctuations in sales behavior, a time-series smoothing filter is applied to the unit-time sales sequence to reduce the impact of occasional fluctuations on trend judgment. This smoothing process uses a continuous time window to weight and integrate sales data, making sales changes between adjacent time slices more continuous and stable. It also suppresses abnormal sales points that significantly deviate from the historical normal range, preventing single-point anomalies from interfering with the overall trend. After time-series smoothing, a continuous and smooth sales change curve is generated for each beverage. Based on the constructed multi-beverage sales curves, a rolling time window approach is used to predict and analyze sales trends, generating predicted sales values ​​for different beverages in the near future. During the forecasting process, the continuous sales curve up to the current moment is used as the input interval. Its direction of change, growth rate, and characteristics of the same historical time period are comprehensively evaluated to estimate the sales level within the next forecast window. The forecast period is set to a relatively short time interval to reflect upcoming changes in sales demand, thereby obtaining short-term sales forecasts for each beverage.

[0026] In this embodiment, the specific steps for generating short-term sales forecasts for different beverages based on rolling forecasts using multiple beverage sales curves are as follows: Based on the sales curves of multiple beverages, the frequency characteristics of sales change rate, peak value, fluctuation period and short-term surge characteristics are calculated to obtain the sales rhythm feature set of multiple beverages. Acquire real-time environmental meteorological data; based on the real-time environmental meteorological data, perform historical database matching to extract historical sales records with similar weather conditions; Rolling forecasts are generated for different beverages based on historical sales records and sales rhythm feature sets.

[0027] In this embodiment, the sales rate characteristics of different beverages are evaluated based on the changes in sales figures within adjacent time windows to characterize the speed of sales growth or decline. A grading interval for the rate of change is set to distinguish between steady growth, rapid increase, and rapid decrease. Secondly, local maxima in the sales curve are identified, and the peak sales values ​​and their corresponding time distributions for each beverage within the statistical period are extracted to reflect the concentration of high-demand periods. Simultaneously, by analyzing the recurring fluctuation patterns of the sales curve over a longer time span, the fluctuation cycle and its frequency characteristics are extracted. For example, 30-minute, 60-minute, or peak business hours can be set as reference cycles in the experimental parameters to identify periodic sales rhythms. Furthermore, sudden increases in sales within short time windows are analyzed, calculating the deviation of the increase from the historical average sales, and a sudden increase identification threshold is set to distinguish between normal fluctuations and abnormal demand growth. In the sales forecasting process, the impact of environmental meteorological factors on user beverage demand is introduced. Real-time environmental meteorological data corresponding to the current moment is collected, including at least parameters such as ambient temperature range, humidity level, precipitation status, weather type, and the trend of changes in perceived temperature. Based on the acquired real-time meteorological parameters, a similarity match is performed with the meteorological records stored in the historical database. By setting tolerance ranges for meteorological parameter deviations, such as allowable deviation ranges for temperature, similar humidity levels, or consistent weather types, historical meteorological samples that are highly similar to the current meteorological conditions are selected from the historical data. During the matching process, different matching weights can be assigned to different meteorological parameters. For example, temperature and weather type are given higher weights, while humidity and precipitation status are given secondary weights, in order to improve the relevance of the matching results to the impact on beverage demand.

[0028] After obtaining the sales rhythm feature set and historical sales records under similar weather conditions, rolling forecast analysis is performed on the short-term sales of different beverages. During the forecasting process, continuous sales data up to the current moment is used as the basic input interval. The forecast is then combined with the rate of change, peak distribution characteristics, periodic frequency characteristics, and short-term surge tendency reflected in the sales rhythm feature set to evaluate the sales trend within the short-term window. Simultaneously, historical sales records matching the current weather conditions are introduced to correct the forecast results for weather scenarios, ensuring that the forecast results reflect the impact of environmental changes on users' beverage selection preferences. The forecast window length is set according to replenishment response needs, for example, set to the next 15 minutes, 30 minutes, or 1 hour, and the forecast sensitivity parameter is adjusted based on historical forecast errors. When the forecast result continuously exceeds the upper limit threshold of the historical average within a short period, it is determined to be a high-demand forecast state; when the forecast result is consistently below the lower limit threshold, it is determined to be a low-demand forecast state.

[0029] In this embodiment, see Figure 3 The diagram below illustrates the detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include: Based on the beverage robot, beverage types are clustered to obtain multiple beverage types; The recipes for multiple beverage types are analyzed one by one to generate material recipe information for different beverages; The material formula information is standardized to generate material formula coefficients for multiple beverages; The material consumption rate is calculated based on the short-term sales forecast and material formulation coefficient to generate a material consumption dataset; the material consumption dataset includes the material consumption rate and consumption quantity within the forecast period.

[0030] In this embodiment, based on the complete set of beverages that the beverage robot can produce, the beverages are categorized into types, grouping beverages with similar material composition and production characteristics into the same beverage type. During the categorization process, the main types of raw materials, the raw material combination structure, the proportion of liquid and additive raw materials, the complexity of the production process, and the distribution characteristics of the production time per cup are comprehensively considered. By performing a similarity comparison analysis on these characteristics, beverages with highly consistent material structure characteristics are aggregated. After categorization, each beverage type corresponds to a set of beverages with relatively stable material consumption patterns, thereby reducing the uncertainty caused by differences in individual beverages during subsequent material calculations. After completing the beverage type categorization, each specific beverage within each type undergoes individual recipe analysis to extract its complete material composition information. During recipe analysis, according to the standard beverage production process, the required materials are broken down item by item, identifying basic liquid raw materials, functional additive raw materials, and decorative or auxiliary raw materials, and recording the relative usage of each type of raw material in a single cup of beverage. Simultaneously, the changes in raw materials caused by different cup sizes, concentrations, or flavor options are uniformly organized, ensuring that the formula information covers material usage across various production specifications. Through this analysis process, each beverage is transformed into material formula information consisting of multiple materials and their relative usage relationships.

[0031] To achieve a unified expression of material consumption relationships across different beverages, the material formulation information of beverages is standardized, transforming previously fragmented formulation descriptions into material formulation coefficients at a unified scale. During standardization, a standard single-cup beverage is used as a reference to normalize the usage relationships of various materials in different beverages, ensuring comparability in material consumption across different beverages. Differences in usage due to changes in cup size or flavor are uniformly converted using specification correction relationships, allowing material consumption to be expressed in a stable proportional form. Based on short-term sales forecasts for different beverages and combined with corresponding material formulation coefficients, a comprehensive assessment of the consumption of various materials within the forecast period is conducted. In the calculation process, the predicted sales values ​​of different beverages within the forecast period are correlated and mapped with their material formulation coefficients to obtain the consumption contribution of each beverage to different materials. The consumption contributions of the same material in multiple beverages are then summarized to form the overall consumption situation of that material within the forecast period. Simultaneously, considering the length of the forecast period, the material consumption situation is expanded over a time scale, reflecting both the consumption rate characteristics per unit time and the cumulative consumption quantity within the forecast period.

[0032] In this embodiment, step S3 includes the following steps: Obtain the front-end material inventory data of the beverage robot; the front-end material inventory data includes the remaining weight of coffee beans, the remaining volume of milk, the remaining milliliters of syrup, and the remaining number of paper cups; The front-end material inventory data is digitally recorded and processed to output the front-end material inventory curve; Based on the front-end material inventory curve, the maximum dispensing capacity of each beverage in the material consumption dataset is calculated, and a dispensing assessment result is generated.

[0033] In this embodiment, the inventory status of materials directly involved in cup making within the beverage robot's front-end area is acquired to form a front-end material inventory data set. This data includes key material information such as the remaining weight of coffee beans, the remaining volume of milk, the remaining milliliters of syrup, and the remaining number of paper cups. Specifically, the remaining weight of coffee beans reflects the basic raw material reserves sufficient to support grinding and extraction; the remaining volume of milk characterizes the continuous supply capacity of dairy beverages; the remaining milliliters of syrup measure the upper limit of flavored beverage production; and the remaining number of paper cups constrains the overall cup-making capacity. During the recording process, different materials are measured using methods matching their physical properties. For example, coffee beans are represented by weight ranges, liquid materials by volume ranges, and consumable materials by quantity ranges. Abnormal states are identified and processed, such as inventory levels approaching the minimum safety margin or showing a sudden downward trend. After acquiring the front-end material inventory data, the inventory status of various materials is continuously and digitally recorded, enabling the formation of an analyzable inventory change trajectory over time. During the digital processing, inventory values ​​at different times are recorded according to a unified time scale, and the inventory status is updated at fixed time intervals, creating a continuous sequence of changes in coffee bean weight, milk volume, syrup volume, and paper cup quantity over time. By organizing the inventory data into a time series, the impact of the consumption rate, consumption rhythm, and replenishment behavior of materials during the cup-making process on the inventory can be clearly reflected. When a certain material shows a significant downward trend or abnormal fluctuation in a short period of time, it can be represented by a change in slope or a breakpoint in the inventory curve.

[0034] Based on the front-end material inventory curve and the generated material consumption dataset, the maximum number of cups that can be dispensed for different beverages under the current inventory conditions is evaluated and calculated. During the evaluation process, for each beverage, considering its corresponding material consumption relationships, dispensing limitations are determined from multiple material dimensions, including coffee beans, milk, syrup, and paper cups, with the current available inventory level reflected in the inventory curve serving as the constraint basis. When a certain material has the lowest dispensing capacity within the forecast period, it is used as the upper limit dispensing limit for that beverage. By comprehensively comparing various material limitations, the maximum number of cups that can be dispensed for different beverages under the current inventory state can be obtained, further generating a dispensing evaluation result that includes the beverage name, dispensing limit, and restricted material types.

[0035] In this embodiment, step S4 includes the following steps: Based on the cup-dispensing assessment results, identify the material with the fastest shortage and mark the critical limiting material; The shortage time points of the key restricted materials are accurately calculated to obtain the shortage time points; Calculate the time buffer for shortage points to generate replenishment time windows; Based on the replenishment time window, multi-material collaborative demand calculations are performed to generate a replenishment report.

[0036] In this embodiment, after assessing the production capacity of each beverage, the production capacity of all beverages under current inventory conditions is compared and analyzed to identify materials that are likely to be depleted in the shortest time. During the analysis, the production capacity of various materials such as coffee beans, milk, syrup, and paper cups within the forecast period is ranked, and the material with the lowest production capacity is marked as a critical constraint material. Critical constraint materials are the bottleneck materials most likely to cause beverage production interruptions under current inventory and sales forecast conditions; these materials directly affect production capacity and sales continuity. For the marked critical constraint materials, the shortage time point is accurately calculated by combining their current inventory level and the material consumption rate within the forecast period. During the calculation, the current remaining quantity of materials is continuously compared with the predicted consumption rate to estimate the earliest time node when the materials will be depleted, which is taken as the material shortage time point. The shortage time point also considers the cumulative consumption impact of the sales forecast curves of each beverage on the materials to ensure that the calculation reflects the actual consumption under mixed sales of different beverages. The fluctuation of the material consumption rate over time is evaluated so that the shortage time point reflects the pressure on inventory during peak sales periods.

[0037] After determining the shortage time points of critical restricted materials, a buffer time calculation is performed to ensure timely replenishment, forming a replenishment time window. This window comprises the time required for material delivery, the time for inventory replenishment operations, and a buffer time to prevent sudden sales fluctuations, ensuring materials are replenished before depletion and preventing interruptions in beverage production. The buffer time is calculated based on the average delivery cycle and operation time, while also considering potential accelerated consumption during peak sales periods, extending the window's start time to guarantee a safety margin. The generation of replenishment time windows clarifies the replenishable time period for each critical restricted material. After obtaining the replenishment time windows for each critical material, collaborative demand calculations are performed based on the consumption relationships and replenishment priorities among multiple materials. These collaborative calculations are based on the shortage time points, available beverage production demand, and material consumption rates for each material, while also considering the shared dependencies of different materials in beverage production, determining the optimal replenishment quantity and order for each material within the replenishment window. By sorting and summarizing the overall demand for critical and non-critical materials, a complete replenishment report is generated. The report includes the replenishment time window, replenishment quantity recommendations, and priority replenishment order for each material.

[0038] In this embodiment, the specific steps for calculating the collaborative demand of multiple materials based on the replenishment time window and generating a replenishment report are as follows: Based on the replenishment time window and material consumption dataset, perform linked calculations to extract the inventory quantity of other materials; Based on the inventory quantity of other materials, perform multi-material collaborative demand calculations to generate the replenishment quantity for all materials; A replenishment report is generated based on the replenishment quantity and replenishment time window. Dynamic replenishment instructions are generated based on the replenishment report and sent to the beverage robot; The beverage robot receives a dynamic replenishment instruction and enters replenishment preparation mode to process the replenishment in the background.

[0039] In this embodiment, after obtaining the replenishment time window and material consumption dataset for critical materials, linked calculations are performed on non-critical materials to comprehensively understand the inventory status of various materials. The linked calculations estimate the remaining quantity of each material within the replenishment window by comparing its current inventory level, consumption rate per unit time, and projected sales volume within the forecast period. For different materials, the potential inventory change trend during the critical material replenishment time window is calculated based on their frequency and proportion of use in beverage production. After obtaining the inventory quantity and consumption information for various materials, multi-material collaborative demand calculations are performed to ensure that all materials can continuously meet the beverage production demand under the predicted sales conditions. The collaborative calculations consider the interdependence between critical and non-critical materials, as well as the proportion of each material in different beverages. By comparing the material consumption rate with the inventory level, the optimal replenishment quantity for each material within the replenishment period is calculated. The calculations simultaneously consider peak material consumption periods and the length of the replenishment time window to ensure that the replenishment quantity meets continuous production demand without causing inventory backlog. After completing the multi-material collaborative calculations, the replenishment quantity and replenishment time window are integrated to form a complete replenishment report. The report records the expected replenishment quantity, replenishment priority, and corresponding replenishment time period for each material. It also indicates the inventory level and expected consumption rate of each material, providing a clear reference for replenishment scheduling. By displaying the consumption status and replenishment requirements of each material within the replenishment window, the report enables front-line managers to understand the execution sequence and urgency of the overall replenishment plan, thereby ensuring that material supply can meet the continuity of beverage production under short-term sales forecast conditions.

[0040] The replenishment report's information on replenishment quantity, time window, and priority for each material is transformed into dynamic replenishment instructions that the beverage robot can recognize. These instructions include the replenishment time, quantity, and order for each material, enabling the robot to adjust its replenishment plan and schedule tasks accordingly. During instruction transmission, material inventory curves and predicted consumption rates are used to ensure dynamic replenishment is matched to the current material status in real time, guaranteeing replenishment operations are completed before actual cup-making demand occurs. The generated dynamic replenishment instructions cover both critical and secondary materials, ensuring the continuity of supply across the entire material mix. Upon receiving the dynamic replenishment instructions, the beverage robot enters replenishment preparation mode and performs material scheduling according to the instructions. The robot executes material replenishment operations sequentially based on the replenishment time window, quantity, and priority, including retrieving materials from the storage area, transporting materials to the front-end replenishment location, and completing material shelving or allocation to the production module. During the replenishment process, the robot monitors inventory changes and cup dispensing in real time, and makes adjustments for possible replenishment delays or material abnormalities. This ensures that the replenishment operation is synchronized with the predicted sales volume, and that key and auxiliary materials are replenished in a short period of time, thereby maintaining the continuity of beverage production and the stability of service.

[0041] In this embodiment, a material replenishment device for a beverage robot is provided, used to execute the material replenishment method for the beverage robot as described above, including: The data acquisition unit is used to collect real-time sales logs; based on the real-time sales logs, it performs rolling forecasts and generates short-term sales forecasts for different beverages. The material calculation unit is used to calculate the consumption rate based on the short-term sales forecast and generate a material consumption dataset. The dispensing evaluation unit is used to calculate the maximum dispensing capacity for each beverage based on the material consumption dataset and generate dispensing evaluation results. The replenishment unit is used to perform multi-material collaborative replenishment analysis based on the cup dispensing assessment results, and output a replenishment report to be sent to the beverage robot for back-end replenishment processing.

[0042] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the material replenishment method for the beverage robot described in any of the above claims.

[0043] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the material replenishment method for the beverage robot described in any of the preceding claims.

[0044] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0045] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein are implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for replenishing materials for a beverage robot, characterized in that, Includes the following steps: Step S1: Collect real-time sales logs; perform rolling forecasts based on real-time sales logs to generate short-term sales forecasts for different beverages; Step S2: Calculate the material consumption rate based on the short-term sales forecast and generate a material consumption dataset; Step S3: Calculate the maximum dispensing capacity for each beverage based on the material consumption dataset, and generate a dispensing capacity assessment result; Step S4: Based on the cup dispensing assessment results, perform multi-material collaborative replenishment analysis, output a replenishment report and send it to the beverage robot for background replenishment processing.

2. The material replenishment method for the beverage robot according to claim 1, characterized in that, The specific steps of step S1 are as follows: The beverage robot collects real-time sales logs, which include order generation time, coffee type, cup size, preparation time, and changes in order density per unit time. The real-time sales logs are timestamped and standardized to extract standardized sales data. Perform multi-beverage sales statistics on standardized sales data to generate sales volume per unit time for different beverages; Time-series smoothing filtering is applied to the sales volume per unit time to construct sales curves for multiple beverages; Rolling forecasts are generated based on the sales curves of multiple beverages to produce short-term sales forecasts for different beverages.

3. The material replenishment method for the beverage robot according to claim 2, characterized in that, The specific steps for generating short-term sales forecasts for different beverages based on rolling forecasts using multiple beverage sales curves are as follows: Based on the sales curves of multiple beverages, the frequency characteristics of sales change rate, peak value, fluctuation period and short-term surge characteristics are calculated to obtain the sales rhythm feature set of multiple beverages. Acquire real-time environmental meteorological data; based on the real-time environmental meteorological data, perform historical database matching to extract historical sales records with similar weather conditions; Rolling forecasts are generated for different beverages based on historical sales records and sales rhythm feature sets.

4. The material replenishment method for the beverage robot according to claim 1, characterized in that, The specific steps of step S2 are as follows: Based on the beverage robot, beverage types are clustered to obtain multiple beverage types; The recipes for multiple beverage types are analyzed one by one to generate material recipe information for different beverages; The material formula information is standardized to generate material formula coefficients for multiple beverages; The material consumption rate is calculated based on the short-term sales forecast and material formulation coefficient to generate a material consumption dataset; the material consumption dataset includes the material consumption rate and consumption quantity within the forecast period.

5. The material replenishment method for the beverage robot according to claim 1, characterized in that, The specific steps of step S3 are as follows: Obtain the front-end material inventory data of the beverage robot; the front-end material inventory data includes the remaining weight of coffee beans, the remaining volume of milk, the remaining milliliters of syrup, and the remaining number of paper cups; The front-end material inventory data is digitally recorded and processed to output the front-end material inventory curve; Based on the front-end material inventory curve, the maximum dispensing capacity of each beverage in the material consumption dataset is calculated, and a dispensing assessment result is generated.

6. The material replenishment method for the beverage robot according to claim 1, characterized in that, The specific steps of step S4 are as follows: Based on the cup-dispensing assessment results, identify the material with the fastest shortage and mark the critical limiting material; The shortage time points of the key restricted materials are accurately calculated to obtain the shortage time points; Calculate the time buffer for shortage points to generate replenishment time windows; Based on the replenishment time window, multi-material collaborative demand calculations are performed to generate a replenishment report.

7. The material replenishment method for the beverage robot according to claim 6, characterized in that, The specific steps for calculating the collaborative demand of multiple materials based on the replenishment time window and generating a replenishment report are as follows: Based on the replenishment time window and material consumption dataset, perform linked calculations to extract the inventory quantity of other materials; Based on the inventory quantity of other materials, perform multi-material collaborative demand calculations to generate the replenishment quantity for all materials; A replenishment report is generated based on the replenishment quantity and replenishment time window. Dynamic replenishment instructions are generated based on the replenishment report and sent to the beverage robot; The beverage robot receives a dynamic replenishment instruction and enters replenishment preparation mode to process the replenishment in the background.

8. A material replenishment device for a beverage robot, characterized in that, The material replenishment method for performing the beverage robot as described in claim 1 includes: The data acquisition unit is used to collect real-time sales logs; based on the real-time sales logs, it performs rolling forecasts and generates short-term sales forecasts for different beverages. The material calculation unit is used to calculate the consumption rate based on the short-term sales forecast and generate a material consumption dataset. The dispensing evaluation unit is used to calculate the maximum dispensing capacity for each beverage based on the material consumption dataset and generate dispensing evaluation results. The replenishment unit is used to perform multi-material collaborative replenishment analysis based on the cup dispensing assessment results, and output a replenishment report to be sent to the beverage robot for back-end replenishment processing.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the material replenishment method for the beverage robot according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the material replenishment method for the beverage robot according to any one of claims 1 to 7.

Images