Food sales data operation and maintenance analysis method and system based on big data

Abstract

The invention discloses a food sales data operation and maintenance analysis method and system based on big data, and the method comprises the steps: collecting food sales data, and carrying out the preprocessing of the data; establishing a food sales prediction model, and calculating a sales volume characteristic change coefficient; food quality indexes are calculated through time sequence analysis; through a lightweight gradient elevator algorithm, screening core factors influencing the sales volume, and calculating a food marketing influence coefficient; calculating an inventory supply coefficient through a statistical analysis method; according to the sales volume characteristic change coefficient, the food quality index, the food marketing influence coefficient and the inventory supply coefficient, calculating a comprehensive sales analysis value and issuing a food sales analysis report; according to the method, problems in traditional sales data analysis can be effectively solved, a visual, comprehensive and accurate analysis result is provided for an enterprise, the enterprise is helped to quickly make a scientific decision, and the marketing management level and market competitiveness of the enterprise in the field of food sales are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of big data analytics, and in particular to a method and system for analyzing and maintaining food sales data based on big data. Background Technology

[0002] In today's digital age, competition in the food industry is increasingly fierce. The operation and analysis of food sales data is crucial for enterprises to make informed decisions and enhance their market competitiveness. Currently, existing technologies for food sales data operation and analysis mostly employ traditional statistical analysis methods. These methods typically only summarize simple data and perform basic statistical calculations. Some enterprises may use basic time series analysis models to make short-term forecasts of sales data, but these models are often structurally simple and cannot fully consider complex market factors and the potential relationships between data. When screening factors affecting sales, traditional methods mostly rely on human experience and judgment, lacking objectivity and accuracy. Traditional technologies do not provide comprehensive data preprocessing, making it difficult to effectively handle the large number of missing and outlier values, and also failing to fully uncover hidden information within the data.

[0003] Due to limitations in analytical methods, it is impossible to comprehensively and deeply analyze food sales data, making it difficult to accurately grasp the changing patterns and inherent characteristics of sales data. This leads to significant errors in predicting sales trends and fails to provide reliable decision-making support for businesses. Furthermore, manual screening of factors influencing sales is highly subjective, prone to overlooking important factors, and inefficient, failing to adapt to a rapidly changing market environment. Insufficient data preprocessing severely impacts the accuracy and reliability of subsequent analysis, causing biased results that fail to accurately reflect the market situation.

[0004] In terms of data processing, this solution effectively cleans and transforms data through comprehensive data preprocessing, improving data quality. It establishes a food sales forecasting model and calculates sales characteristic change coefficients to more accurately capture the dynamic patterns of sales data. Time series analysis is used to calculate food quality indicators, linking food quality with sales data and providing a more comprehensive analytical perspective. A lightweight gradient boosting machine algorithm is employed to screen core factors and calculate food marketing impact coefficients, achieving objectivity and efficiency in factor selection. Statistical analysis methods are used to calculate inventory supply coefficients, making supply chain management more scientific and rational. Finally, a comprehensive sales and operational value is calculated, providing enterprises with intuitive, comprehensive, and accurate analytical results, helping them make rapid and scientific decisions, and effectively improving their marketing management level and market competitiveness in the food sales field. Summary of the Invention

[0005] This invention provides a method for the operation and maintenance analysis of food sales data based on big data, including:

[0006] Collect food sales data and preprocess the data;

[0007] Establish a food sales forecasting model and calculate the coefficient of change of sales characteristics based on food sales data;

[0008] Based on food sales data, time series analysis is used to calculate food quality indicators;

[0009] The lightweight gradient booster algorithm is used to screen the core factors affecting sales and calculate the impact coefficient of food marketing.

[0010] Calculate the inventory supply coefficient based on food sales data using statistical analysis methods;

[0011] Calculate the comprehensive sales analysis value based on the sales characteristic change coefficient, food quality indicators, operational efficiency coefficient, and inventory supply coefficient.

[0012] A food sales analysis report is obtained based on the comprehensive sales analysis values.

[0013] The above-mentioned big data-based food sales data operation and analysis method includes establishing a food sales forecasting model and calculating the sales characteristic change coefficient based on food sales data, including:

[0014] Based on the characteristics of food sales data, a food sales forecasting model is constructed using cluster analysis, association rule mining, and time series analysis methods.

[0015] Based on the food sales forecasting model, input food sales data and calculate the coefficient of change of sales characteristics.

[0016] The above-mentioned big data-based food sales data operation and analysis method includes calculating food quality indicators based on food sales data through time series analysis, including:

[0017] By using time series analysis methods, we can extract the characteristics of sales data and obtain the time series analysis results of food sales data.

[0018] Based on the time series analysis results and the relationship between food sales and quality, food quality indicators are calculated.

[0019] The above-mentioned big data-based food sales data operation and analysis method uses a lightweight gradient boosting machine algorithm to screen core factors affecting sales and calculate the food marketing impact coefficient, including:

[0020] The model is built using the lightweight gradient booster algorithm to obtain the importance scores of each feature and to obtain a list of core factors.

[0021] Calculate the food marketing impact coefficient based on the list of core factors.

[0022] The above-mentioned big data-based food sales data operation and maintenance analysis method includes, through statistical analysis, calculating the inventory supply coefficient based on food sales data, including:

[0023] Statistical analysis of food sales data is conducted to obtain key inventory indicators based on turnover.

[0024] Calculate the inventory supply coefficient based on key inventory indicators.

[0025] This invention provides a food sales data operation and maintenance analysis system based on big data, comprising:

[0026] The data hub module is used to collect food sales data and preprocess the data;

[0027] The sales forecasting module is used to build a food sales forecasting model and calculate the coefficient of change of sales characteristics based on food sales data.

[0028] The quality analysis module is used to calculate food quality indicators based on food sales data through time series analysis.

[0029] The marketing analytics module uses a lightweight gradient booster algorithm to filter out the core factors affecting sales and calculate the impact coefficient of food marketing.

[0030] The inventory analysis module is used to calculate the inventory supply coefficient based on food sales data using statistical analysis methods.

[0031] The comprehensive analysis module is used to calculate the comprehensive sales analysis value based on the sales characteristic change coefficient, food quality indicators, food marketing impact coefficient, and inventory supply coefficient; and to obtain a food sales analysis report based on the comprehensive sales analysis value.

[0032] As described above, in a big data-based food sales data operation and analysis system, the sales forecasting module is used to establish a food sales forecasting model and calculate the sales characteristic change coefficient based on food sales data, including:

[0033] The model building submodule is used to build a food sales forecasting model based on the characteristics of food sales data, using cluster analysis, association rule mining, and time series analysis methods.

[0034] The coefficient of change calculation submodule is used to calculate the coefficient of change of sales characteristics based on the food sales forecasting model and input food sales data.

[0035] As described above, in a big data-based food sales data operation and analysis system, the quality analysis module is used to calculate food quality indicators based on food sales data through time series analysis, including:

[0036] The time series analysis submodule is used to extract features from sales data using time series analysis methods to obtain time series analysis results for food sales data.

[0037] The quality index calculation submodule is used to calculate food quality indicators based on the time series analysis results and the relationship between food sales and quality.

[0038] As described above, in a big data-based food sales data operation and analysis system, the marketing analysis module uses a lightweight gradient boosting machine algorithm to filter core factors affecting sales and calculate the food marketing impact coefficient, including:

[0039] The feature extraction submodule is used to build a model using the lightweight gradient boosting machine algorithm, obtain the importance score of each feature, and obtain a list of core factors.

[0040] The marketing efficiency calculation submodule is used to calculate the food marketing impact coefficient based on the list of core factors.

[0041] As described above, in a big data-based food sales data operation and analysis system, the inventory analysis module is used to calculate the inventory supply coefficient based on food sales data using statistical analysis methods, including:

[0042] The inventory analysis submodule is used to perform statistical analysis on food sales data and obtain key inventory indicators based on turnover.

[0043] The inventory supply calculation submodule is used to calculate the inventory supply coefficient based on key inventory indicators.

[0044] The beneficial effects achieved by this invention are as follows:

[0045] In terms of data processing, this solution effectively cleans and transforms data through comprehensive data preprocessing, improving data quality. It establishes a food sales forecasting model and calculates sales characteristic change coefficients to more accurately capture the dynamic patterns of sales data. Time series analysis is used to calculate food quality indicators, linking food quality with sales data and providing a more comprehensive analytical perspective. A lightweight gradient boosting machine algorithm is employed to screen core factors and calculate food marketing impact coefficients, achieving objectivity and efficiency in factor selection. Statistical analysis methods are used to calculate inventory supply coefficients, making supply chain management more scientific and rational. Finally, a comprehensive sales and operational value is calculated, providing enterprises with intuitive, comprehensive, and accurate analytical results, helping them make rapid and scientific decisions, and effectively improving their marketing management level and market competitiveness in the food sales field. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0047] Figure 1 This application provides a method for the operation and maintenance analysis of food sales data based on big data, as described in Embodiment 1.

[0048] Figure 2 This is a food sales data operation and maintenance analysis system based on big data, provided in Embodiment 2 of this application. Detailed Implementation

[0049] 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, not all, of the embodiments of the present invention. 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.

[0050] Example 1

[0051] like Figure 1 As shown, Embodiment 1 of this application provides a method for the operation and maintenance analysis of food sales data based on big data, including:

[0052] S110: Collect food sales data and preprocess the data;

[0053] Food sales data was acquired from multiple data sources. After collection, order details, inventory records, customer information, and other data were integrated. Preprocessing was performed, including data cleaning, data integration, data transformation, data dimensionality reduction, data smoothing, data standardization, and feature engineering.

[0054] Remove duplicate data: Detect and delete duplicate records;

[0055] Outlier detection: Outliers are detected using the Z-score method or box plots, and extreme outliers are deleted or corrected.

[0056] Handling missing values: For missing data, you can choose to delete records containing missing values ​​or fill in the missing values ​​with the mean, median, mode, or through a prediction algorithm;

[0057] Data normalization: Using Min-Max standardization or Z-score standardization methods, numerical data such as sales revenue and sales volume are normalized to eliminate the differences in the units of measurement between different variables.

[0058] Data discretization: Using binning techniques, continuous data is divided into discrete intervals;

[0059] Data encoding: One-hot encoding is used to convert categorical variables into numerical forms;

[0060] Data dimensionality reduction: Principal component analysis projects high-dimensional data into a low-dimensional space while retaining as much information as possible;

[0061] Data smoothing: The moving average method is used to smooth the data and reduce noisy data.

[0062] S120: Establish a food sales forecasting model and calculate the coefficient of change of sales characteristics based on food sales data;

[0063] Food categories are grouped to uncover food relationships and capture data features. After cross-validation and optimization, the data is integrated with visualization tools and early warning thresholds for dynamic monitoring, and sales characteristic change coefficients are calculated periodically.

[0064] Establish a food sales forecasting model and calculate the coefficient of change of sales characteristics based on food sales data. This includes the following sub-steps:

[0065] S121: Based on the characteristics of food sales data, cluster analysis, association rule mining, and time series analysis methods are used to construct a food sales forecasting model;

[0066] The K-means clustering algorithm is used to group food categories, customer groups, or sales regions. The elbow rule is used to plot the sum of squared errors curves corresponding to different cluster sizes K, and the inflection point of the curve is identified to determine the optimal K value. Attributes such as food price, sales volume, and gross profit margin are used as feature variables. Centroids are iteratively calculated to divide food categories, customer groups, or sales regions into different clusters. The sales characteristics of each cluster are analyzed, for example, high-volume, low-profit clusters and low-volume, high-price clusters. The silhouette coefficient is then used to evaluate the clustering quality.

[0067] The Apriori algorithm is used to mine the associations between food items. Support, confidence, and lift thresholds are set to identify frequent itemsets and generate strong association rules.

[0068] Time series analysis, based on food sales data, captures trend, seasonality, and cyclical characteristics. The ADF test is used to determine data stationarity, and non-stationary data undergoes differencing. Autocorrelation and partial autocorrelation functions are used to determine model parameters, and external variables are introduced to improve prediction accuracy.

[0069] After each model is built, its performance is evaluated through cross-validation. By comparing metrics such as the silhouette coefficient of clustering, the lift of association rules, and the MAE of time series, the parameters are optimized and integrated to form a comprehensive analytical model covering customer segmentation, cross-selling, and trend prediction.

[0070] S122: Based on the food sales forecasting model, input food sales data and calculate the sales characteristic change coefficient;

[0071] Establish an operation and maintenance monitoring function based on the food sales forecasting model, and calculate the sales characteristic change coefficient by monitoring the changes and trends of food sales data in real time.

[0072] Food sales data is buffered in a message queue, and the data is cleaned and transformed in real time based on a streaming computing framework. The results are then dynamically compared with historical data and the output of the prediction model.

[0073] Use visualization tools to display real-time changes and trends in sales data, set early warning thresholds, and trigger alarms when data deviates from the thresholds or prediction deviations exceed the set range.

[0074] The sales volume characteristic change coefficient is calculated periodically using a food sales forecasting model, and the calculation results are stored in a time series database.

[0075] The formula for calculating the coefficient of change in sales characteristics is as follows:

[0076]

[0077] S n S0 represents the sales data for period n, i.e., the sales amount at the end of the analysis period; S0 represents the initial sales data, which represents the sales amount at the beginning of the analysis period; T represents the time period, which refers to the time span of the sales data analysis; μ is the average sales growth rate, which represents the percentage increase in sales per unit time; μ(t) reflects the dynamic trend of sales growth within time t.

[0078] The average sales amount within period T is represented by the following formula: S i σ represents the sales amount at the i-th moment within the time period; σ represents the standard deviation of the sales data, which measures the dispersion of the sales data relative to the mean and reflects the severity of sales fluctuations.

[0079] α represents the market volatility impact coefficient, used to adjust the degree of impact of external market fluctuations on sales volume; β represents the seasonal adjustment coefficient, which controls the period length of seasonal fluctuations in sales data. β represents a sine wave with a generation period of β; γ represents the price elasticity coefficient, which measures the degree of impact of price changes on sales volume; ΔP represents the amount of price change; ΔS represents the ratio of the amount of sales volume change; θ is the external intervention coefficient, reflecting the impact of non-market natural fluctuation factors on sales volume.

[0080] S130: Calculate food quality indicators based on food sales data through time series analysis;

[0081] Time series analysis is performed on food sales data to extract features such as trends, seasonality, and volatility, and these features are combined with quality data to calculate food quality indicators.

[0082] Based on food sales data, time series analysis is used to calculate food quality indicators, including the following sub-steps:

[0083] S131: Using time series analysis methods, extract the characteristics of sales data to obtain the time series analysis results of food sales data;

[0084] Import food sales data into the analytics system. This data must include at least two key fields: timestamp and sales volume.

[0085] Use timestamps as indexes and sales volume as values ​​to construct time series objects, and set the frequency of the time series according to the time granularity of the data.

[0086] Next, linear regression is used to fit the trend of the time series data. Seasonal decomposition methods, such as classic seasonal decomposition, are used to decompose the time series into trend, seasonal, and residual components through data analysis. The results of the seasonal decomposition are observed, and a periodogram is used to detect the length of the seasonal cycle. The strength of the seasonal component is calculated as a seasonal characteristic value. The volatility index of the time series is calculated, and the rate of change of volatility is calculated as a volatility characteristic value through data analysis.

[0087] The extracted trend features, seasonal features, periodic features, and volatility features are summarized into a data table and the data is stored.

[0088] Use visualization tools to create raw data plots, trend plots, seasonal breakdown plots, periodic plots, and volatility plots for time series data.

[0089] Specifically, this includes drawing trend lines for time series data to show the overall trend changes; drawing trend, seasonality, and residual component plots after seasonal decomposition to visually display seasonal characteristics; drawing periodicity plots to show the periodic frequency distribution of the data; and drawing volatility index plots for time series data to show the volatility changes of the data.

[0090] The feature extraction results and visualization charts are integrated to obtain the final time series analysis results.

[0091] S132: Based on the time series analysis results and the relationship between food sales and quality, calculate food quality indicators;

[0092] The time series analysis results should include at least food quality-related data, such as quality testing indicators.

[0093] To verify whether the timestamps of sales data and quality data are aligned, the two sets of data are integrated into one data table by timestamp through data merging to obtain a comprehensive dataset containing sales characteristics and quality indicators.

[0094] We conduct data analysis to calculate sales characteristics and combine the comprehensive dataset to calculate food quality indicators.

[0095] The formulas for calculating food quality indicators are as follows:

[0096]

[0097] n represents the number of categories of quality-influencing factors, which comprehensively consider the combined effects of multiple different factors on food quality through multiplication; i is the counting variable in the multiplication operation, used to traverse different categories of quality-influencing factors.

[0098] λ i (t) represents the function of the i-th quality influencing factor changing with time t, reflecting the strength of the factor's effect on food quality over time; t represents the time variable; t1 and t2 represent the lower and upper limits of the integral, defining the time range of the quality influencing factor's effect within the time interval; μ i (s) represents a certain attenuation or suppression effect, where s is the integral variable; Indicates the relationship with λ i (t) decay correction over the time interval [t1,t2], taking into account the decay characteristics of quality factors over time;

[0099] m represents the number of detection indicators related to the i-th type of quality influencing factor; j is the counting variable in the summation operation, used to iterate through the different detection indicators related to the i-th type of quality influencing factor; β ij ρ represents the weighting coefficient, reflecting the degree of correlation between the i-th quality influencing factor and the j-th detection indicator, such as the correlation weight between the taste factor and the appearance detection indicator of a certain food; ij σ represents the density of the j-th detection index under the i-th type of quality influencing factor, used to describe the distribution characteristics of this detection index under the i-th type of quality influencing factor; ij This represents the fluctuation function of the j-th detection index under the i-th type of quality influencing factors, reflecting the fluctuation of this detection index; It represents the ratio of the gradients of two functions, reflecting the relative change between the distribution and fluctuation of the detection index;

[0100] γ(t) represents the impact of environmental factors on food quality as time t changes; the erf(t) error function works together with γ(t) to reflect the probability and uncertainty of the impact of environmental factors on food quality.

[0101] S140: Using a lightweight gradient booster algorithm, core factors affecting sales are screened and the influence coefficient of food marketing is calculated.

[0102] Features are extracted, categorical features are encoded, and irrelevant features are initially filtered out. LightGBM (Lightweight Gradient Boosting Machine) model parameters are configured, and the monitoring status of indicators is recorded. Feature importance scores are obtained and ranked, core factors are selected, and after verification, a final list of core factors is generated for calculating the impact of food marketing.

[0103] The lightweight gradient boosting machine algorithm is used to screen the core factors affecting sales and calculate the influence coefficient of food marketing. The specific steps include the following:

[0104] S141: Use the lightweight gradient booster algorithm to build a model, obtain the importance scores of each feature, and obtain a list of core factors;

[0105] Feature engineering is performed to extract and transform features from the original data, and categorical features are encoded.

[0106] Meanwhile, the system performs preliminary screening of features, removing obviously irrelevant features; and configures the parameters of the LightGBM (Lightweight Gradient Boosting Machine) model according to business needs and data scale.

[0107] The parameters include at least the tree depth, learning rate, and number of iterations, and the data is divided into training and testing sets.

[0108] The LightGBM (Lightweight Gradient Boosting Machine) model is trained using the training set data. During training, various metrics are recorded to monitor the model's status. The model's feature importance evaluation function is then invoked to obtain an importance score for each feature, which is stored in a data structure. The score results are then sorted and visualized.

[0109] Based on business requirements, set a threshold for feature importance scoring, retaining features with importance scores above a certain value or the top N most important features. Filter the core factors from the sorted feature list and store them. Perform business logic validation on the selected core factors to obtain the final list of core factors.

[0110] The list of core factors includes at least the marketing cost index and the cross-channel synergy effect index, which are key factors affecting the impact coefficient of food marketing.

[0111] S142: Calculate the food marketing impact coefficient based on the list of core factors;

[0112] After obtaining the list of core factors, calculate the food marketing impact coefficient based on the list;

[0113] The formula for calculating the impact coefficient of food marketing is as follows:

[0114]

[0115] γ represents the marketing cost index, reflecting the weight of marketing investment costs on the sales and marketing impact coefficient; δ represents the marketing contact frequency decay coefficient, indicating that the effect of each contact may decrease as the number of times a certain marketing method is used; T j This represents the relevant time interval of the j-th marketing contact; This represents the average time interval of all marketing contacts; This is used to measure the degree to which consumers' sensitivity to repeated marketing messages diminishes; m represents the total number of marketing contacts, and j represents the i-th marketing contact.

[0116] ε represents the cross-channel synergy index, which measures the weight of the impact coefficient of synergy between different marketing channels on sales and marketing. This represents brand loyalty-related metrics, such as customer repeat purchase rate, customer retention rate, and brand recommendation willingness score; k represents the k-th metric measuring brand loyalty; p represents that there are p different dimensions of this metric.

[0117] ζ represents the purchase decision delay factor, reflecting the time delay between consumers being exposed to marketing information and making a purchase decision; This section is used to assess the balance between the effectiveness and stability of marketing activities; ω represents the regional difference index, reflecting the differences between different sales regions; R l This represents the marketing response metrics relevant to the l-th region; The average of marketing response metrics across all regions; q represents the total number of marketing regions;

[0118] μ refers to the average conversion rate of all marketing activities of an enterprise within a specific period, reflecting the basic efficiency of marketing activities in driving users to take targeted actions, such as purchasing, registering, and repeat purchases; σ measures the degree of fluctuation in conversion rates between different marketing activities or regions; S is used to distinguish the priority of different marketing activity types; μS combines the average conversion rate with the activity type weight to reflect the weighted average conversion efficiency; σS combines the activity type weight to adjust the standard deviation of the conversion rate, reflecting the amplification or reduction effect of activity type differences on conversion stability.

[0119] S150: Calculate the inventory supply coefficient based on food sales data using statistical analysis methods;

[0120] Divide food sales data by time dimension and create time series datasets, analyze sales trends, calculate growth rate and duration, investigate the reasons for decline, and calculate inventory supply coefficient.

[0121] Calculate the inventory supply coefficient using statistical analysis methods based on food sales data, including the following sub-steps:

[0122] S151: Conduct statistical analysis of food sales data and obtain key inventory indicators based on turnover.

[0123] Statistical analysis of food sales data reveals sales trends, analyzes turnover, and obtains key inventory indicators.

[0124] The collected food sales data were divided according to the time dimension, and a time series dataset was created for statistical analysis.

[0125] A line chart is generated based on time series data. The trend of the line chart is analyzed to determine whether the sales trend is upward, downward, or stable. For upward trends, the magnitude and duration of growth are analyzed to identify possible driving factors. For downward trends, possible causes are investigated. Based on the average monthly sales volume, the impact of short-term fluctuations is eliminated to obtain a stable sales trend. Inventory quantity changes over different time periods are statistically analyzed to assess inventory turnover and ultimately obtain key inventory indicators.

[0126] Key inventory indicators should include at least the analysis results of average sales volume, expected inventory, and actual inventory.

[0127] S152: Calculate the inventory supply coefficient based on key inventory indicators;

[0128] The formula for calculating the inventory supply coefficient is as follows:

[0129]

[0130] t1 is the start time of the time interval; t2 is the end time of the time interval; α represents the sales weighting factor, used to measure the importance of sales data in inventory supply assessment; the larger the value, the greater the impact of sales data on the inventory supply coefficient; S avg β represents the average sales volume, i.e., the average quantity of food sold within the statistical period, reflecting the average level of sales; I represents the inventory volatility weighting factor, measuring the importance of inventory volatility in inventory supply assessment; std The standard deviation of inventory is the standard deviation of inventory quantity within a statistical period, reflecting the dispersion of inventory data.

[0131] γ represents the demand skewness weighting factor, which measures the importance of demand distribution skewness and reflects the asymmetry of demand distribution; D skewδ represents the skewness coefficient of demand, the skewness coefficient of food demand within the statistical period, describing the degree to which the demand distribution deviates from a symmetrical distribution; δ represents the weighting factor of the time series coefficient of variation, measuring the importance of the time series coefficient of variation and reflecting the fluctuation characteristics of sales data over time; T cv The coefficient of variation of the time series represents the relative dispersion of the sales data time series within a period.

[0132] ε represents the weighting factor between expected and actual inventory, reflecting the impact of inventory plan execution on inventory supply; Q exp This represents the expected inventory level, which is the anticipated inventory level set based on sales forecasts and inventory strategies; Q act η represents the actual inventory quantity, i.e., the actual inventory quantity held at the statistical point in time; η represents the replenishment efficiency factor, the larger the value, the higher the replenishment efficiency, and the more significant the improvement effect on the inventory supply coefficient; ζ represents the price correlation weighting factor, reflecting the impact of price fluctuations on inventory supply; P corr This indicates the degree of linear correlation between food prices and inventory levels within a given period.

[0133] S160: Calculate the comprehensive sales analysis value based on the sales characteristic change coefficient, food quality indicators, food marketing impact coefficient, and inventory supply coefficient.

[0134] The formula for calculating the comprehensive sales analysis value is as follows:

[0135]

[0136] t1 is the start time of the time interval; t2 is the end time of the time interval; W represents the sales characteristic change coefficient; λ represents the time-related influence coefficient, characterizing the rate of change of sales characteristics over time. When λ is greater than 0, as time progresses, e λt An increase in the value indicates that the impact of sales fluctuations is increasing; t is a time variable; Q represents the food quality index; μ represents the quality fluctuation-related parameters, reflecting the characteristics of food quality fluctuations; δ represents the quality fluctuation frequency; E represents the food marketing impact coefficient; v represents the correlation weight between inventory and marketing, reflecting the degree of influence of inventory supply on marketing effectiveness.

[0137] K represents the inventory supply coefficient; ω represents the marketing saturation coefficient, indicating the critical state where marginal returns diminish after a certain level of marketing investment, reflecting the saturation characteristics of marketing investment; K i This represents the inventory supply coefficient for the i-th region; θ represents the average inventory supply coefficient across all regions; f is the inventory deviation index, which adjusts the intensity of inventory fluctuation factors; θ represents the category correction factor, which corrects the comprehensive sales analysis value based on the differences in various aspects of different food categories.

[0138] S170: Obtain a food sales analysis report based on comprehensive sales analysis values;

[0139] Based on the comprehensive sales and maintenance values, a final analysis is conducted to obtain the final food sales analysis report, and the analysis results are visualized.

[0140] Example 2

[0141] like Figure 2 As shown, Embodiment 2 of this application provides a food sales data operation and maintenance analysis system based on big data, including:

[0142] Data hub module 21: Collects food sales data and preprocesses the data;

[0143] Sales Forecasting Module 22: Establish a food sales forecasting model and calculate the coefficient of change of sales characteristics based on food sales data;

[0144] Model building submodule 221: Based on the characteristics of food sales data, cluster analysis, association rule mining, and time series analysis methods are used to build a food sales forecasting model;

[0145] Submodule 222 for calculating the coefficient of change: Based on the food sales forecasting model, input food sales data and calculate the coefficient of change of sales characteristics;

[0146] Quality Analysis Module 23: Calculate food quality indicators based on food sales data through time series analysis;

[0147] Time Series Analysis Submodule 231: Uses time series analysis methods to extract features from sales data and obtain time series analysis results for food sales data;

[0148] Quality Indicator Calculation Submodule 232: Calculates food quality indicators based on time series analysis results and the relationship between food sales and quality;

[0149] Marketing Analysis Module 24: Using a lightweight gradient booster algorithm, core factors affecting sales are screened and the impact coefficient of food marketing is calculated.

[0150] Feature extraction submodule 241: Uses a lightweight gradient boosting machine algorithm to build a model, obtains the importance score of each feature, and obtains a list of core factors;

[0151] Marketing efficiency calculation submodule 242: Calculate the food marketing impact coefficient based on the list of core factors;

[0152] Inventory Analysis Module 25: Calculates the inventory supply coefficient based on food sales data using statistical analysis methods;

[0153] Inventory Analysis Submodule 251: Performs statistical analysis on food sales data and obtains key inventory indicators based on turnover.

[0154] Inventory supply calculation submodule 252: Calculates the inventory supply coefficient based on key inventory indicators;

[0155] Comprehensive Analysis Module 26: Calculate the comprehensive sales analysis value based on the sales characteristic change coefficient, food quality indicators, food marketing impact coefficient, and inventory supply coefficient; obtain the food sales analysis report based on the comprehensive sales analysis value.

[0156] Corresponding to the above embodiments, the present invention provides a computer storage medium, including: at least one memory and at least one processor;

[0157] The memory is used to store one or more program instructions;

[0158] A processor is used to run one or more program instructions to execute a big data-based food sales data operation and maintenance analysis method.

[0159] Corresponding to the above embodiments, this embodiment of the invention provides a computer-readable storage medium containing one or more program instructions, which are executed by a processor to provide a big data-based food sales data operation and maintenance analysis method.

[0160] The embodiments disclosed in this invention provide a computer-readable storage medium storing computer program instructions. When the computer program instructions are executed on a computer, the computer performs the aforementioned method for the operation and maintenance analysis of food sales data based on big data.

[0161] In this embodiment of the invention, the processor can be an integrated circuit chip with signal processing capabilities. The processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0162] The various methods, steps, and logic diagrams disclosed in the embodiments of this invention can be implemented or executed. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this invention can be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The processor reads information from the storage medium and, in conjunction with its hardware, completes the steps of the above methods.

[0163] The storage medium can be memory, such as volatile memory or non-volatile memory, or may include both volatile and non-volatile memory.

[0164] Among them, non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory.

[0165] Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (Synchlink DRAM, SLDRAM), and direct memory bus RAM (DRRAM).

[0166] The storage media described in the embodiments of the present invention are intended to include, but are not limited to, these and any other suitable types of memory.

[0167] Those skilled in the art will recognize that, in one or more of the examples above, the functions described in this invention can be implemented using a combination of hardware and software. When applied as software, the corresponding functions can be stored in a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any medium that facilitates the transmission of computer programs from one place to another. Storage media can be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0168] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made on the basis of the technical solution of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for the operation and maintenance analysis of food sales data based on big data, characterized in that, include: Collect food sales data and preprocess the data; Establish a food sales forecasting model and calculate the coefficient of change of sales characteristics based on food sales data; Based on food sales data, time series analysis is used to calculate food quality indicators; The lightweight gradient booster algorithm is used to screen the core factors affecting sales and calculate the impact coefficient of food marketing. Calculate the inventory supply coefficient based on food sales data using statistical analysis methods; Calculate the comprehensive sales analysis value based on the sales characteristic change coefficient, food quality indicators, operational efficiency coefficient, and inventory supply coefficient. A food sales analysis report is obtained based on the comprehensive sales analysis values.

2. The method for operation and maintenance analysis of food sales data based on big data according to claim 1, characterized in that, Establish a food sales forecasting model and calculate the coefficient of change of sales characteristics based on food sales data, including: Based on the characteristics of food sales data, a food sales forecasting model is constructed using cluster analysis, association rule mining, and time series analysis methods. Based on the food sales forecasting model, input food sales data and calculate the coefficient of change of sales characteristics.

3. The method for operation and maintenance analysis of food sales data based on big data according to claim 1, characterized in that, Based on food sales data, time series analysis is used to calculate food quality indicators, including: By using time series analysis methods, we can extract the characteristics of sales data and obtain the time series analysis results of food sales data. Based on the time series analysis results and the relationship between food sales and quality, food quality indicators are calculated.

4. The method for operation and maintenance analysis of food sales data based on big data according to claim 1, characterized in that, Using a lightweight gradient booster algorithm, we screen the core factors influencing sales and calculate the impact coefficient of food marketing, including: The model is built using the lightweight gradient booster algorithm to obtain the importance scores of each feature and to obtain a list of core factors. Calculate the food marketing impact coefficient based on the list of core factors.

5. The method for operation and maintenance analysis of food sales data based on big data according to claim 1, characterized in that, Using statistical analysis methods, the inventory supply coefficient is calculated based on food sales data, including: Statistical analysis of food sales data is conducted to obtain key inventory indicators based on turnover. Calculate the inventory supply coefficient based on key inventory indicators.

6. A food sales data operation and maintenance analysis system based on big data, characterized in that, include: The data hub module is used to collect food sales data and preprocess the data; The sales forecasting module is used to build a food sales forecasting model and calculate the coefficient of change of sales characteristics based on food sales data. The quality analysis module is used to calculate food quality indicators based on food sales data through time series analysis. The marketing analytics module uses a lightweight gradient booster algorithm to filter out the core factors affecting sales and calculate the impact coefficient of food marketing. The inventory analysis module is used to calculate the inventory supply coefficient based on food sales data using statistical analysis methods. The comprehensive analysis module is used to calculate the comprehensive sales analysis value based on the sales characteristic change coefficient, food quality indicators, food marketing impact coefficient, and inventory supply coefficient. A food sales analysis report is obtained based on the comprehensive sales analysis values.

7. The food sales data operation and maintenance analysis system based on big data according to claim 6, characterized in that, The sales forecasting module is used to build food sales forecasting models and calculate the coefficients of change in sales characteristics based on food sales data, including: The model building submodule is used to build a food sales forecasting model based on the characteristics of food sales data, using cluster analysis, association rule mining, and time series analysis methods. The coefficient of change calculation submodule is used to calculate the coefficient of change of sales characteristics based on the food sales forecasting model and input food sales data.

8. A food sales data operation and maintenance analysis system based on big data according to claim 6, characterized in that, The quality analysis module is used to calculate food quality indicators based on food sales data through time series analysis, including: The time series analysis submodule is used to extract features from sales data using time series analysis methods to obtain time series analysis results for food sales data. The quality index calculation submodule is used to calculate food quality indicators based on the time series analysis results and the relationship between food sales and quality.

9. A food sales data operation and maintenance analysis system based on big data according to claim 6, characterized in that, The marketing analytics module uses a lightweight gradient booster algorithm to filter out core factors influencing sales and calculate the food marketing impact coefficient, including: The feature extraction submodule is used to build a model using the lightweight gradient boosting machine algorithm, obtain the importance score of each feature, and obtain a list of core factors. The marketing efficiency calculation submodule is used to calculate the food marketing impact coefficient based on the list of core factors.

10. A food sales data operation and maintenance analysis system based on big data according to claim 6, characterized in that, The inventory analysis module is used to calculate the inventory supply coefficient based on food sales data using statistical analysis methods, including: The inventory analysis submodule is used to perform statistical analysis on food sales data and obtain key inventory indicators based on turnover. The inventory supply calculation submodule is used to calculate the inventory supply coefficient based on key inventory indicators.

Images