Retail promotion optimization method and system based on big data

Abstract

The invention relates to the technical field of marketing data analysis, in particular to a retail promotion optimization method and system based on big data, and the method comprises the following steps: carrying out the processing and encryption of data through promotion and transaction data collected from a retailer network, synchronously updating the data on a block chain network, and generating shared promotion activity data. According to the method, the behavior of the consumer is subjected to grouping analysis, and the behavior pattern and consumption habit of the consumer are predicted according to the purchase history record, so that the pertinence and individuation degree of the promotion activity are improved. The method not only helps retailers to better understand different consumer groups, but also enables promotion resources to be distributed more reasonably, thereby optimizing the sales effect and improving the customer satisfaction. The introduction of the intelligent contract enables the execution of the promotion activity to be automatic, and improves the efficiency and accuracy. Through fuzzification processing of the response data of the consumer, the evaluation of the promotion strategy becomes more detailed and comprehensive.

Description

Technical Field

[0001] The present invention relates to the technical field of marketing data analysis, and in particular to a retail promotion optimization method and system based on big data. Background Art

[0002] Marketing data analysis refers to the use of data collection, mining and analysis techniques to help companies better understand market trends, consumer behavior and sales performance. By processing and analyzing large amounts of data collected from e-commerce platforms, social media, online advertising and other channels, companies can optimize decision-making processes and enhance the effectiveness of marketing strategies. This field usually covers quantitative and qualitative analysis, using data visualization, statistical modeling, machine learning and other methods to identify market opportunities or improve customer conversion rates.

[0003] Among them, the retail promotion optimization method uses data analysis technology to optimize promotion strategies through online marketing and e-commerce platforms. It analyzes the performance data of promotion activities and reasonably allocates promotion resources to maximize sales results and customer satisfaction. Through precise promotion design and optimization, the promotion effect can be improved and marketing costs can be reduced.

[0004] In terms of promotion resource allocation and promotion effect evaluation, the existing technology is prone to data untimely update, which affects the real-time response capability and accuracy of promotion activities. In addition, the analysis of consumer behavior in the existing technology is difficult to accurately predict behavior based on different consumer groups, resulting in promotion activities unable to accurately meet the needs of target customers, affecting promotion effects and customer satisfaction. In addition, it is difficult to evaluate promotion strategies based on consumer reactions, which is not conducive to retailers accurately adjusting promotion strategies to cope with market changes. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a retail promotion optimization method and system based on big data.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a retail promotion optimization method based on big data, comprising the following steps:

[0007] S1: Collect promotion and transaction data from the retail merchant network, perform data processing and encryption operations, synchronize and update it on the blockchain network, and generate shared promotion activity data;

[0008] S2: using the shared promotion activity data and combining it with the consumer purchase history records to group consumers, annotate the purchase behavior patterns and consumption habits of each group, and generate consumer behavior grouping results;

[0009] S3: Based on the consumer behavior grouping results, a smart contract is constructed on the blockchain platform through the merchant node, promotion conditions are set, the verification and management of promotion activities are automatically executed, and a promotion activity execution record is generated;

[0010] S4: by analyzing the execution records of the promotion activities, fuzzy processing is performed on the purchase frequency, price sensitivity, and promotion response data of consumers, the reaction of consumer groups to the promotion strategy is evaluated, and a promotion effect prediction result is generated;

[0011] S5: using the promotion effect prediction results, setting multiple market scenarios for the promotion strategy, performing random simulation, analyzing the fluctuation of the promotion effect under each scenario, and generating promotion strategy performance data;

[0012] S6: Adjust the promotion discount depth, duration and target consumer group through the promotion strategy performance data, reallocate promotion resources and budget according to the simulation results, and generate an optimized promotion strategy.

[0013] The present invention has the following improvements: the shared promotion activity data includes the timestamp of the promotion activity, the amount of the promotion transaction, and the transaction verification information of the merchant node; the consumer behavior grouping results include the average purchase frequency of the consumer group, the preferred product type, and the number of historical promotion participation times; the promotion activity execution record includes the number of participants in each promotion activity, the number of completed transactions, and the promotion response speed; the promotion effect prediction results include the response level of the consumer group, the sensitivity range of price discounts, and the expected promotion conversion rate; the promotion strategy performance data includes the promotion cost, revenue fluctuation range, market response rate, and stability of strategy effect under various market scenarios; the optimized promotion strategy includes the optimized promotion discount percentage, the adjusted promotion duration, the re-positioned target consumer group, and the resource allocation results.

[0014] The present invention is improved in that the specific steps of collecting promotion and transaction data from the retail merchant network, performing data processing and encryption operations, and synchronously updating it on the blockchain network to generate shared promotion activity data are as follows:

[0015] S101: Based on the promotion and transaction data collected from the retail merchant network, standardize the data in multiple formats, handle missing values ​​and inconsistent records, integrate information from all sources, and generate formatted promotion transaction data;

[0016] S102: Based on the formatted promotion transaction data, perform data encryption operations, archive and classify all encrypted data, and generate encrypted promotion transaction data;

[0017] S103: Based on the encrypted promotional transaction data, synchronize nodes in the blockchain network, verify the promotion and transaction data of each node with the main network, and synchronously update all records to generate shared promotion activity data.

[0018] The present invention is improved in that the shared promotion activity data is used to group consumers in combination with consumer purchase history records, and the purchase behavior patterns and consumption habits of each group are marked. The specific steps of generating consumer behavior grouping results are as follows:

[0019] S201: extracting the consumer's purchase behavior data based on the shared promotion activity data and the consumer's purchase history records, and generating the consumer's purchase behavior data according to multiple dimensions such as consumption frequency, product preference, and consumption amount;

[0020] S202: Based on the consumer purchase behavior data, label and stratify the consumption habits of each group to generate consumer classification data;

[0021] S203: Based on the consumer classification data, the promotion feedback data of each group is associated with its purchase behavior data, and the promotion response and purchase habits of each group are sorted and analyzed to generate consumer behavior grouping results.

[0022] The present invention is improved in that, based on the consumer behavior grouping results, a smart contract is constructed on the blockchain platform through the merchant node, promotion conditions are set, and verification and management of promotion activities are automatically performed. The specific steps of generating promotion activity execution records are as follows:

[0023] S301: Based on the consumer behavior grouping result, extract the characteristic data of the group, configure the promotion conditions, match the promotion conditions with the group characteristics by setting the discount ratio, participating product categories and promotion period parameters, and generate the set promotion condition information;

[0024] S302: Based on the set promotion condition information, deploy the promotion contract through the blockchain node, set the automatic execution rules of the promotion activity, and generate the promotion contract execution data according to the progress of the promotion activity and the participation information of each consumer group;

[0025] S303: Based on the promotion contract execution data, record the participation of consumer groups in the promotion activities, as well as the purchasing behavior and response frequency of each group, analyze the transaction amount and promotion participation according to the promotion conditions, and generate a promotion activity execution record.

[0026] The present invention is improved in that, by analyzing the execution records of the promotion activities, fuzzy processing is performed on the purchase frequency, price sensitivity, and promotion response data of consumers, the reaction of consumer groups to the promotion strategy is evaluated, and the specific steps of generating the prediction results of the promotion effect are as follows:

[0027] S401: extracting the purchase frequency, price sensitivity and promotion response data of each consumer group based on the promotion activity execution record, and dividing them into multi-level response categories to generate promotion response classification data;

[0028] S402: Based on the promotion response classification data, the group response data is fuzzified, the promotion response data is divided into intervals according to preset fuzzy rules, the response degree is divided into multiple levels of low, medium and high, and fuzzy promotion response information is generated;

[0029] S403: Based on the fuzzy promotion response information, the response strength of the current group is evaluated, the attractiveness of the promotion strategy to each group is analyzed, the response probability of the future promotion strategy is calculated, and the promotion effect prediction result is generated.

[0030] The present invention is improved in that the response intensity of the current group is evaluated using the formula:

[0031]

[0032] Calculate the response probability P of future promotion strategies r ;

[0033] Among them, R i represents the response intensity of group i in historical promotions μ i is the fuzzy membership, V i represents the historical purchase volume of the i-th group, and n is the number of groups participating in the promotion.

[0034] The present invention is improved in that the specific steps of using the promotion effect prediction result to set multiple market scenarios for the promotion strategy, perform random simulation, analyze the fluctuation of the promotion effect under each scenario, and generate the promotion strategy performance data are as follows:

[0035] S501: Based on the promotion effect prediction result, a virtual market scenario is set through historical market data and consumer behavior data, market variables such as demand fluctuations, price sensitivity and seasonal change parameters are combined, a corresponding scenario model is established, and a virtual market scenario parameter set is generated;

[0036] S502: Based on the virtual market scenario parameter set, adjusting the promotion strategy parameters in each scenario model, including setting the discount range, promotion time and target commodity category, simulating the effect of the promotion strategy in multiple scenarios using logic rules, and generating a virtual promotion strategy simulation result;

[0037] S503: Based on the simulation results of the virtual promotion strategy, analyze the promotion strategy performance in each scenario, evaluate the cost-effectiveness and revenue fluctuations, determine the best performing strategy combination by comparing the strategy performance in multiple scenarios, and generate promotion strategy performance data.

[0038] The present invention is improved in that the promotional discount depth, duration and target consumer group are adjusted through the promotional strategy performance data, and promotional resources and budgets are reallocated according to the simulation results. The specific steps of generating an optimized promotional strategy are as follows:

[0039] S601: Analyze the cost-effectiveness ratio and resource utilization efficiency based on the promotion strategy performance data, screen multiple promotion strategies according to the benefit ranking, and generate screened promotion strategy data;

[0040] S602: Based on the filtered promotion strategy data, adjust the discount depth, promotion duration and target consumer group, match resources with strategies, and reallocate budgets to generate adjusted promotion resource configurations;

[0041] S603: Based on the adjusted promotion resource configuration, adjust promotion parameters, combine market data and consumer feedback, optimize promotion discounts, duration and key factors of target groups, match resource allocation with promotion goals, and generate an optimized promotion strategy.

[0042] A retail promotion optimization system based on big data, the system comprising:

[0043] The data collection module collects various promotion and transaction data based on the retail merchant network, performs data standardization and preliminary cleaning, including removing invalid and duplicate data, and generates standardized promotion transaction data;

[0044] The data encryption and synchronization module performs data encryption operations based on the standardized promotion transaction data, synchronizes the encrypted data between nodes through blockchain technology, and generates shared promotion activity data;

[0045] The consumer analysis module groups consumers based on the shared promotion activity data and their purchase history records, and labels the purchase behavior and consumption habits of each group to generate consumer behavior grouping results;

[0046] The smart contract building module builds a smart contract on the blockchain platform based on the consumer behavior grouping results, automatically executes the verification and management of promotion conditions, including the logical judgment and selection of promotion conditions, monitors consumers' responses to promotion activities, and generates promotion activity execution records;

[0047] The strategy optimization module analyzes consumer purchase frequency, price sensitivity and promotion response data based on the promotion activity execution records, processes these data through fuzzy logic, evaluates the consumer group's response to the promotion strategy, conducts market simulation, optimizes promotion strategy parameters, including discount depth and promotion duration, and generates an optimized promotion strategy.

[0048] Compared with the prior art, the advantages and positive effects of the present invention are:

[0049] In the present invention, by performing group analysis on consumer behavior and predicting consumer behavior patterns and consumption habits based on purchase history records, the pertinence and personalization of promotional activities are improved. It not only helps retailers to better understand different consumer groups, but also enables a more reasonable allocation of promotional resources, thereby optimizing sales results and improving customer satisfaction. The introduction of smart contracts automates the execution of promotional activities, improving efficiency and accuracy. By fuzzy processing of consumer response data, the evaluation of promotional strategies becomes more detailed and comprehensive, helping retailers to adjust future market strategies more accurately. Finally, by simulating different market scenarios and analyzing the performance of promotional strategies, retailers can make more information-supported decisions in an uncertain market environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 A flowchart of a retail promotion optimization method based on big data is proposed for the present invention;

[0051] Figure 2 This is a detailed flow chart of step S1 of the present invention;

[0052] Figure 3 This is a detailed flow chart of step S2 of the present invention;

[0053] Figure 4 This is a detailed flow chart of step S3 of the present invention;

[0054] Figure 5 This is a detailed flow chart of step S4 of the present invention;

[0055] Figure 6 This is a detailed flow chart of step S5 of the present invention;

[0056] Figure 7 This is a detailed flow chart of step S6 of the present invention;

[0057] Figure 8 A module diagram of a retail promotion optimization system based on big data is proposed for the present invention. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0059] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating positions or positional relationships, are based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0060] See also Figure 1 The present invention provides a technical solution: a retail promotion optimization method based on big data, comprising the following steps:

[0061] S1: Collect promotion and transaction data from the retail merchant network, perform data processing and encryption operations, synchronize and update it on the blockchain network, and generate shared promotion activity data;

[0062] S2: Use shared promotional activity data and consumer purchase history records to group consumers, annotate each group's purchase behavior patterns and consumption habits, and generate consumer behavior grouping results;

[0063] S3: Based on the results of consumer behavior grouping, a smart contract is built on the blockchain platform through merchant nodes to set promotion conditions, automatically perform verification and management of promotion activities, and generate promotion execution records;

[0064] S4: By analyzing the execution records of promotion activities, consumers’ purchase frequency, price sensitivity, and promotion response data are fuzzified to evaluate the consumer group’s response to the promotion strategy and generate promotion effect prediction results;

[0065] S5: Using the promotion effect prediction results, set multiple market scenarios for the promotion strategy, conduct random simulations, analyze the fluctuations of the promotion effect under each scenario, and generate promotion strategy performance data;

[0066] S6: Based on the promotion strategy performance data, adjust the promotion discount depth, duration and target consumer group, reallocate promotion resources and budget according to the simulation results, and generate an optimized promotion strategy.

[0067] Shared promotion activity data includes the timestamp of the promotion activity, the amount of the promotion transaction, and the transaction verification information of the merchant node. The consumer behavior segmentation results include the average purchase frequency of the consumer group, the preferred product type, and the number of historical promotion participation times. The promotion activity execution records include the number of participants in each promotion activity, the number of completed transactions, and the promotion response speed. The promotion effect prediction results include the response level of the consumer group, the sensitivity range of price discounts, and the expected promotion conversion rate. The promotion strategy performance data includes the promotion cost, revenue fluctuation range, market response rate, and stability of the strategy effect under various market scenarios. The optimized promotion strategy includes the optimized promotion discount percentage, adjusted promotion duration, re-positioned target consumer group, and resource allocation results.

[0068] See also Figure 2 , by collecting promotion and transaction data from the retail merchant network, performing data processing and encryption operations, and synchronizing and updating it on the blockchain network, the specific steps for generating shared promotion activity data are as follows:

[0069] S101: Based on the promotion and transaction data collected from the retail merchant network, standardize the data in multiple formats, handle missing values ​​and inconsistent records, integrate information from all sources, and generate formatted promotion transaction data;

[0070] First, these data are classified into multiple categories such as transaction data, promotional activity data, and consumer behavior data. Then, data standardization is performed to ensure compatibility of data formats between different systems by converting data from different sources and formats into a consistent format. In this process, missing values ​​are filled with adjacent values ​​or historical averages, and inconsistent records are corrected or deleted by comparing the verification mechanism of the original data source. When merging data, the consistency of key fields such as the data timestamp and transaction amount is ensured to avoid the overlap or omission of duplicate data. Finally, promotion and transaction information from all sources are integrated, and all processed data are stored in a unified format.

[0071] S102: Based on the formatted promotion transaction data, perform data encryption operations, archive and classify all encrypted data, and generate encrypted promotion transaction data;

[0072] First, a unique encryption key is generated for each type of data. During the encryption process, different encryption keys are used for transaction data and promotional data to ensure data privacy and security. The specific operation process includes segmented encryption of data, especially separate encryption of consumers' sensitive information (such as transaction amounts and consumption habits). The encrypted data is classified and stored according to different types. Transaction data, promotional activity data, and consumer feedback data are stored in different encrypted folders so that they can be quickly retrieved and classified during subsequent processing. After the encryption operation, all data is archived and managed by category.

[0073] S103: Based on the encrypted promotion transaction data, synchronize nodes in the blockchain network, verify the promotion and transaction data of each node with the main network, and synchronously update all records to generate shared promotion activity data;

[0074] First, the encrypted promotion and transaction data is transmitted through multiple nodes on the chain. The data hash value of each node is used to verify the consistency of the data of each node to ensure that the data has not been tampered with. After completing the verification, each node will be synchronously updated to the latest promotion and transaction records. In this process, the data synchronization of each node is achieved through the network broadcast communication mode. The system compares the core parameters such as the timestamp and transaction amount of each transaction to ensure that the data of all nodes is consistent. Through this mechanism, a complete and tamper-proof promotion transaction record is finally formed in the blockchain network.

[0075] See also Figure 3 , using shared promotional activity data and combining consumer purchase history records to group consumers, annotate each group's purchase behavior patterns and consumption habits, and generate consumer behavior grouping results in the following specific steps:

[0076] S201: extracting consumer purchase behavior data based on shared promotion activity data and combined with consumer purchase history records, and generating consumer purchase behavior data based on multiple dimensions such as consumption frequency, product preference, and consumption amount;

[0077] First, the transaction records of all consumers are screened and sorted. By extracting key indicators in the transaction data, such as product category, transaction amount, purchase time, etc., consumers' behavior can be deeply analyzed. Then these data are classified according to dimensions such as consumption frequency, product preference, and consumption amount. Consumption frequency is the number of times a consumer makes a purchase within a specific time period. Product preference is obtained based on the category weight distribution of the purchased products. The consumption amount is summarized by accumulating the transaction amount in each time period. Finally, detailed consumer purchasing behavior data is generated to ensure that it can accurately reflect the purchasing habits of each consumer.

[0078] S202: Based on the consumer purchase behavior data, label and stratify the consumption habits of each group to generate consumer classification data;

[0079] First, we mark consumers' consumption habits and extract indicators such as consumption frequency, average consumption amount, and product preferences. Then, we stratify these data and divide them into different consumer groups. The specific stratification standard is to compare the average consumption frequency and consumption amount of each consumer. According to the consumption frequency, they are divided into high-frequency, medium-frequency and low-frequency consumers. The group with high consumption frequency and large consumption amount is marked as the high-consumption group, the group with medium frequency and relatively stable consumption is marked as the medium-consumption group, and the group with low consumption frequency and diversified purchase of goods is divided into the low-consumption group.

[0080] S203: Based on the consumer classification data, the promotion feedback data of each group is associated with its purchase behavior data, and the promotion response and purchase habits of each group are sorted and analyzed to generate consumer behavior grouping results;

[0081] Associate each group’s promotion feedback data with their purchase behavior data, extract the number of promotion participations and each purchase record, according to the formula:

[0082]

[0083] Calculate the weighted average of purchase behavior in promotion response for each group;

[0084] Among them, B i represents the weighted average of the group’s purchasing behavior after the i-th promotion, which is used to measure the effectiveness of the promotion. i Represents the average purchase behavior value of the group before the i-th promotion. This data is usually calculated based on historical purchase records or purchase behavior data from the previous cycle, using the group's average transaction amount or purchase quantity as a measurement standard. i represents the average purchasing behavior value of the group after the i-th promotion, similar to A i , which is obtained by collecting purchase data after the promotion ends, and generated by monitoring transaction behavior after the promotion. f The pre-promotion weight factor is used to measure the importance of pre-promotion purchase behavior. This value can be set based on the historical consumption habits of the group and the level of activity before the promotion. It is usually extracted and set from historical data. a The post-promotion weight factor is used to measure the impact of post-promotion purchasing behavior. The post-promotion weight is usually higher to reflect the effect of promotion on group behavior. This factor is set after analyzing the behavioral data of the group after the promotion. i represents the current promotion activity number, which is used to distinguish data comparison and cumulative analysis during different promotion activities.

[0085] Take the purchasing behavior data of the group before and after the promotion as A i and C i For example, in a certain promotion, the purchase value of a certain group before the promotion is A i is 100, the purchase value after promotion is C i is 150. Determine the weight factor W before and after the promotion f and W a , according to the historical data analysis before the promotion, set W f =0.4, the data after promotion shows that W a =0.6, based on the different responses of the groups before and after the promotion.

[0086] Substitute the parameters into the formula:

[0087]

[0088] For example, the historical benchmark value shows that the weighted average of the group's purchase behavior after the promotion under similar conditions is 120, and the weighted average of the purchase behavior after the current promotion is calculated to be 130, indicating that the promotion has increased by 8.33% compared to the historical benchmark value. This result shows that this promotion has a good driving effect on the group's purchase behavior, and the purchase behavior after the promotion has increased significantly. Compared with the historical benchmark value, this promotion has brought a higher purchase conversion rate, indicating that the optimization of the promotion strategy has increased the group's participation and consumption enthusiasm, and obtained the consumer behavior grouping results.

[0089] See also Figure 4 Based on the results of consumer behavior grouping, the specific steps of building smart contracts on the blockchain platform through merchant nodes, setting promotion conditions, and automatically executing the verification and management of promotion activities and generating promotion activity execution records are as follows:

[0090] S301: Based on the consumer behavior grouping results, extract the characteristic data of the group, configure the promotion conditions, match the promotion conditions with the group characteristics by setting the discount ratio, participating product categories and promotion period parameters, and generate the set promotion condition information;

[0091] First, extract the characteristic data of each group, including consumption frequency, product preference and average consumption amount. By classifying and organizing these data, obtain the consumption behavior patterns of each group in different dimensions. The consumption frequency reflects the activity of the group, the product preference determines the selection of promotional product categories, and the average consumption amount is used to set the discount ratio. Subsequently, configure targeted promotion conditions. The discount ratio will be adjusted according to the consumption frequency and consumption amount. The length of the promotion period is determined by the consumption behavior data, and the product category is obtained from the product preference data to ensure that the promotion conditions can fully match the characteristics of the group. By matching and combining these parameters, set the promotion condition information.

[0092] S302: Based on the set promotion condition information, deploy the promotion contract through the blockchain node, set the automatic execution rules of the promotion activities, and generate the promotion contract execution data according to the progress of the promotion activities and the participation information of each consumer group;

[0093] The promotion contract is deployed through blockchain nodes. Automatic execution rules are set in the promotion contract, including the management of promotion time periods, adjustment of discount ratios, and selection of product categories. These rules are determined by the parameters in the promotion condition information. The system collects real-time sales data through each node. The promotion contract can trigger preset rules based on these real-time data and dynamically adjust the promotion conditions. For example, when the sales volume of a certain type of product reaches the set threshold, the system will automatically adjust the discount of the product or extend the promotion time. All operations are automatically completed by blockchain nodes to ensure that promotional activities can be flexibly adjusted according to sales progress, and the execution data of promotional activities are synchronized to the blockchain network in real time.

[0094] S303: Based on the promotion contract execution data, record the participation of consumer groups in the promotion activities, as well as the purchase behavior and response frequency of each group, analyze the transaction amount and promotion participation according to the promotion conditions, and generate a promotion activity execution record;

[0095] Analyze the participation of each group in promotional activities, extract the number of promotional participation and time period information of each group, further analyze the transaction amount data of each group, and judge the participation of each group in promotional activities by combining the transaction amount with the promotional participation frequency. By associating promotional conditions with actual transaction behaviors, classify and analyze the execution effect of promotional activities. For example, the matching degree between the transaction amount of different groups and promotional activities will be used as the basis for adjusting subsequent promotional strategies. Ultimately, the participation behavior and transaction data of all groups will be integrated into the execution record of promotional activities.

[0096] See also Figure 5 , by analyzing the records of promotion activities, fuzzifying the consumer's purchase frequency, price sensitivity, and promotion response data, evaluating the consumer group's response to the promotion strategy, and generating the promotion effect prediction results, the specific steps are as follows:

[0097] S401: extracting the purchase frequency, price sensitivity and promotion response data of each consumer group based on the promotion activity execution records, and dividing them into multi-level response categories to generate promotion response classification data;

[0098] First, we extract the purchase frequency, price sensitivity and promotion response data of each consumer group. The consumption activity of the consumer group is judged by the purchase frequency. The price sensitivity is obtained by analyzing the purchase changes in different price ranges before and after the promotion. The promotion response data reflects the direct reaction of consumers to the promotion activities. These data are further divided into multi-level response categories. According to the different response intensities, the promotion response of the group is divided into multiple categories, from low response group to high response group. Each category has a different promotion participation mode to ensure the accurate classification of the response data of each group.

[0099] S402: Based on the promotion response classification data, the group response data is fuzzified, and the promotion response data is divided into intervals according to preset fuzzy rules, and the response degree is divided into multiple levels of low, medium and high, and fuzzy promotion response information is generated;

[0100] First, the group response data is fuzzy processed. Fuzzy processing includes dividing the response data of the consumer group into intervals according to preset fuzzy rules, usually divided into three levels: low response, medium response and high response. The data interval of each level is divided according to the response degree of the group. The membership degree of each group data is calculated according to the set fuzzy rules. Based on these data and preset fuzzy rules, the system further marks the groups at different response levels. This interval division can help better identify the promotion response level of the group.

[0101] In addition, the response data of the group is fuzzified according to the formula:

[0102]

[0103] Calculate the degree of membership of each group;

[0104] Among them, x i represents the response data of the ith group, μ ijk is the membership of the ith group to the response level k, L jk ,M jk ,U jk They represent the lower limit, center and upper limit parameters of the response level k respectively, which are determined based on historical data analysis to reflect the actual distribution of different response levels.

[0105] If the response data x of a group i is 20, for the middle response level, it is known that L jk =15,M jk =25,U jk =35:

[0106]

[0107] The calculation results show that the group's membership to the middle response level is 0.5, which is a medium level.

[0108] S403: Based on the fuzzy promotion response information, evaluate the response strength of the current group, analyze the attractiveness of the promotion strategy to each group, calculate the response probability of the future promotion strategy, and generate the promotion effect prediction result;

[0109] The response intensity of the current group is evaluated using the formula:

[0110]

[0111] Calculate the response probability P of future promotion strategies r ;

[0112] Among them, R i represents the response intensity of the i-th group in historical promotions, extracted from the promotion execution records, μ i is the fuzzy membership, which indicates the response degree membership value of group i in the fuzzy processing, V i represents the historical purchase volume of the i-th group, extracted from the sales data, and n is the number of groups participating in the promotion.

[0113] If the response strengths of the three groups are 0.6, 0.8, and 0.7, the memberships are 0.5, 0.6, and 0.7, the historical purchase volumes are 300, 400, and 350, and the weights are 0.3, 0.4, and 0.3:

[0114]

[0115] The results show that after considering the membership, historical purchase volume and weight, the predicted response probability of the future promotion strategy is 0.492, which provides a quantitative prediction of the future promotion effect. If the historical benchmark value is 0.45 compared with the historical benchmark value of 0.45, it shows that the improvement of the promotion strategy has increased the response rate.

[0116] By evaluating the response probability of each group, we can analyze the attractiveness of the promotion strategy to each group. Groups with high response probability show a strong response to promotion activities, indicating that these groups are very sensitive and receptive to the current promotion strategy. Groups with low response probability may need more incentives or adjustments to promotion activities to increase their participation.

[0117] See also Figure 6 , using the promotion effect prediction results, set multiple market scenarios for the promotion strategy, conduct random simulations, analyze the fluctuations of the promotion effect under each scenario, and generate the promotion strategy performance data in the following specific steps:

[0118] S501: Based on the promotion effect prediction results, a virtual market scenario is set through historical market data and consumer behavior data, market variables such as demand fluctuations, price sensitivity and seasonal change parameters are combined, a corresponding scenario model is established, and a virtual market scenario parameter set is generated;

[0119] The system constructs virtual market scenarios by integrating historical market data and consumer behavior data. The construction process first ensures the integrity and consistency of input variables by cleaning and standardizing market data, and then uses big data analysis methods to extract key market variables including demand fluctuations, price sensitivity, seasonal changes, etc. These variables are based on time series and trend analysis as input parameters of the model. Next, through machine learning algorithms and statistical models (such as ARIMA models, regression analysis, etc.), different combinations of these market variables are simulated and a set of virtual scenario parameters are generated, defining possible dynamics under different market conditions.

[0120] S502: Based on the virtual market scenario parameter set, adjust the promotion strategy parameters under each scenario model, including setting the discount range, promotion time and target commodity category, and use logical rules to simulate the effect of the promotion strategy under multiple scenarios to generate a virtual promotion strategy simulation result;

[0121] Under each scenario model, promotion strategy parameters were adjusted, including setting discount range, promotion time, and target product categories. The adjustment of strategy parameters was based on logical rules that simulated promotion effects under various market conditions. The logical rules included: if market demand forecasts increased, increase discount range to attract more consumers; if competitor activities increased, extend promotion time to maintain market share; if a product category showed high demand in a specific season, the setting of promotion strategy parameters was optimized based on actual market demand, consumer purchase history, and expected market changes. The AnyLogic tool was used to analyze the performance of promotion activities in different market dynamics, ensuring the accuracy and applicability of simulation results.

[0122] S503: Based on the simulation results of the virtual promotion strategy, analyze the promotion strategy performance in each scenario, evaluate the cost-effectiveness and revenue fluctuations, determine the best strategy combination by comparing the strategy performance in multiple scenarios, and generate promotion strategy performance data;

[0123] By comparing the strategy costs and benefit results under different scenarios, the financial impact of each strategy is quantified using cost-benefit analysis model tools. The evaluation model includes but is not limited to rate of return and risk-adjusted cost. By comparing the strategy performance under multiple scenarios, it is possible to determine which promotion strategy is most effective under specific market conditions.

[0124] See also Figure 7, through the promotion strategy performance data, adjust the promotion discount depth, duration and target consumer group, reallocate promotion resources and budget according to the simulation results, and generate the optimized promotion strategy. The specific steps are as follows:

[0125] S601: Based on the promotion strategy performance data, analyze the cost-effectiveness ratio and resource utilization efficiency, screen multiple promotion strategies according to the benefit ranking, and generate screened promotion strategy data;

[0126] By considering the cost-effectiveness ratio and resource utilization efficiency, various promotional strategies are ranked and screened. This process involves the collection and processing of a large amount of data. First, the performance data of different promotional strategies in the market are collected, including but not limited to consumer responses, changes in sales volume and input-output ratio. Then, data analysis techniques such as regression analysis are used to identify which strategies have the best performance in terms of cost-effectiveness ratio and resource utilization efficiency. By setting a benefit evaluation model, the benefit indicators of each strategy are calculated, and then the strategies are ranked according to these indicators. Finally, the promotional strategy with the highest benefit is selected.

[0127] To calculate the benefit index of each strategy, follow the formula:

[0128]

[0129] Among them, E represents the efficiency rate, which is obtained by comparing the net value of income with the cost, R' represents the total revenue brought by the strategy, which is obtained through sales data and market feedback statistics, and C' represents the total cost of the strategy, which is determined through financial records and cost analysis.

[0130] If the total revenue R' generated by a certain promotional strategy is 500,000 yuan and the total cost C' is 300,000 yuan, according to the formula:

[0131]

[0132] It can be calculated that the benefit rate of this strategy is 0.67. This means that for every 1 yuan of cost invested, 0.67 yuan of additional income can be generated. Compared with the historical data of the same period, if the historical benefit rate averages 0.50, it shows that the effect of this strategy has been significantly improved, which helps companies evaluate the return on investment and make further market decisions.

[0133] S602: Based on the filtered promotion strategy data, adjust the discount depth, promotion duration and target consumer group, match resources with strategies, and reallocate budgets to generate adjusted promotion resource configurations;

[0134] First, evaluate the resources required for each promotional strategy, including financial resources, human resources, and material resources. Then, adjust resource allocation based on the specific resource requirements of each strategy to ensure that resources can effectively support the adjusted promotional strategy. In addition, through market research and consumer feedback, analyze the sensitivity of different consumer groups to promotional discounts and duration, and optimize discount settings and duration arrangements accordingly.

[0135] S603: Based on the adjusted promotion resource configuration, adjust the promotion parameters, combine market data and consumer feedback, optimize the key factors of promotion discount, duration and target group, match resource allocation with promotion goals, and generate an optimized promotion strategy.

[0136] Adjust promotion parameters and combine market data and consumer feedback to optimize key factors such as promotion discount, duration and target group according to the formula:

[0137] G=P d ×T×N g

[0138] Calculate the effectiveness of optimized promotional strategies;

[0139] Among them, P d represents the discount rate, which is obtained based on historical discount data and competitor analysis. T represents the promotion duration, which is determined through market demand analysis and evaluation of past promotion effects. g represents the number of target groups, obtained through consumer market research and demographic data, and G represents the number of consumer behaviors potentially influenced by the promotional activities.

[0140] If the discount rate P of a promotional activity d is 0.20 (i.e. 20% discount), the promotion duration T is 30 days, and the number of target groups N g For 10,000 people, according to the formula:

[0141] G = 0.20 × 30 × 10,000

[0142] It can be calculated that the potential reach of this promotion is 60,000. This value represents the number of consumer behaviors that the promotion can influence under ideal circumstances. Compared with the same promotion in the previous quarter, if the reach was 40,000 in the previous quarter, this adjustment shows a clear improvement and enhanced effect.

[0143] According to the target group N g The scale and distribution of marketing budget and personnel are prioritized to areas with high population density. At the same time, considering the discount rate P dThe optimal combination of promotion time T can maximize resource utilization, such as offering deeper discounts during peak consumption periods or extending the duration to cover more potential customers. In addition, resource allocation is adjusted based on market feedback and sales data to ensure that resource investment in each link can correspond to the expected sales increase, thereby maximizing promotion goals.

[0144] See also Figure 8 , a retail promotion optimization system based on big data, the system includes:

[0145] The data collection module collects various promotion and transaction data based on the retail merchant network, performs data standardization and preliminary cleaning, including removing invalid and duplicate data, and generates standardized promotion transaction data;

[0146] The data encryption and synchronization module performs data encryption operations based on standardized promotion transaction data, synchronizes encrypted data between nodes through blockchain technology, and generates shared promotion activity data;

[0147] The consumer analysis module groups consumers based on shared promotional activity data and their purchase history, and labels the purchase behavior and consumption habits of each group to generate consumer behavior grouping results;

[0148] The smart contract building module builds smart contracts on the blockchain platform based on the results of consumer behavior grouping, automatically performs the verification and management of promotion conditions, including the logical judgment and selection of promotion conditions, monitors consumers' responses to promotion activities, and generates promotion execution records;

[0149] The strategy optimization module analyzes consumer purchase frequency, price sensitivity and promotion response data based on promotion activity execution records, processes these data through fuzzy logic, evaluates consumer groups' reactions to promotion strategies, conducts market simulations, optimizes promotion strategy parameters, including discount depth and promotion duration, and generates optimized promotion strategies.

[0150] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A retail promotion optimization method based on big data, characterized in that: The following steps are involved: By collecting promotion and transaction data from the retail merchant network, performing data processing and encryption operations, and synchronizing and updating it on the blockchain network, shared promotion activity data is generated; The shared promotion activity data is used in combination with the consumer purchase history records to group consumers, and the purchase behavior patterns and consumption habits of each group are marked to generate consumer behavior grouping results; Based on the consumer behavior grouping results, a smart contract is constructed on the blockchain platform through the merchant node, promotion conditions are set, the verification and management of promotion activities are automatically executed, and promotion activity execution records are generated; By analyzing the execution records of the promotion activities, the purchase frequency, price sensitivity and promotion response data of consumers are fuzzy processed to evaluate the response of consumer groups to the promotion strategy and generate promotion effect prediction results; Using the promotion effect prediction results, set multiple market scenarios for the promotion strategy, perform random simulations, analyze the fluctuations of the promotion effect under each scenario, and generate promotion strategy performance data; Through the promotion strategy performance data, the promotion discount depth, duration and target consumer group are adjusted, and promotion resources and budgets are reallocated according to the simulation results to generate an optimized promotion strategy.

2. The retail promotion optimization method based on big data according to claim 1, characterized in that: The shared promotion activity data includes the timestamp of the promotion activity, the amount of the promotion transaction, and the transaction verification information of the merchant node. The consumer behavior grouping results include the average purchase frequency, preferred product type and historical promotion participation times of the consumer group. The promotion activity execution record includes the number of participants in each promotion activity, the number of completed transactions, and the promotion response speed. The promotion effect prediction results include the response level of the consumer group, the sensitivity range of price discounts and the expected promotion conversion rate. The promotion strategy performance data includes the promotion cost, revenue fluctuation range, market response rate and stability of strategy effect under various market scenarios. The optimized promotion strategy includes the optimized promotion discount percentage, the adjusted promotion duration, the re-positioned target consumer group and resource allocation results.

3. The retail promotion optimization method based on big data according to claim 1, characterized in that: The specific steps of collecting promotion and transaction data from the retail merchant network, performing data processing and encryption operations, and synchronizing and updating it on the blockchain network to generate shared promotion activity data are as follows: Based on the promotion and transaction data collected from the retail merchant network, standardize the data in multiple formats, handle missing values ​​and inconsistent records, integrate information from all sources, and generate formatted promotion transaction data; Based on the formatted promotion transaction data, perform data encryption operations, archive and classify all encrypted data, and generate encrypted promotion transaction data; Based on the encrypted promotional transaction data, node synchronization is performed in the blockchain network, the promotion and transaction data of each node is verified with the main network, and all records are updated synchronously to generate shared promotional activity data.

4. The retail promotion optimization method based on big data according to claim 1, characterized in that: The specific steps of using the shared promotion activity data and combining the consumer purchase history records to group consumers, annotating the purchase behavior patterns and consumption habits of each group, and generating the consumer behavior grouping results are as follows: Based on the shared promotion activity data, combined with the consumer's purchase history, the consumer's purchase behavior data is extracted, and the consumer's purchase behavior data is generated according to multiple dimensions such as consumption frequency, product preference, and consumption amount; Based on the consumer purchase behavior data, the consumption habits of each group are labeled and layered to generate consumer classification data; Based on the consumer classification data, the promotion feedback data of each group is associated with its purchase behavior data, and the promotion response and purchase habits of each group are sorted and analyzed to generate consumer behavior grouping results.

5. The retail promotion optimization method based on big data according to claim 1, characterized in that: Based on the consumer behavior grouping results, the specific steps of building a smart contract on the blockchain platform through the merchant node, setting promotion conditions, automatically executing the verification and management of promotion activities, and generating promotion activity execution records are as follows: Based on the consumer behavior grouping results, extract the characteristic data of the group, configure the promotion conditions, match the promotion conditions with the group characteristics by setting the discount ratio, participating product categories and promotion period parameters, and generate the set promotion condition information; Based on the set promotion condition information, the promotion contract is deployed through the blockchain node, the automatic execution rules of the promotion activity are set, and the promotion contract execution data is generated according to the progress of the promotion activity and the participation information of each consumer group; Based on the promotion contract execution data, the participation of consumer groups in promotional activities, as well as the purchasing behavior and response frequency of each group are recorded, the transaction amount and promotion participation are analyzed according to the promotion conditions, and the promotion activity execution record is generated.

6. The retail promotion optimization method based on big data according to claim 1, characterized in that: By analyzing the execution records of the promotion activities, the purchase frequency, price sensitivity and promotion response data of consumers are fuzzy processed, the reaction of consumer groups to the promotion strategy is evaluated, and the specific steps of generating the prediction results of the promotion effect are as follows: Based on the promotion activity execution records, extract the purchase frequency, price sensitivity and promotion response data of each consumer group, and divide them into multi-level response categories to generate promotion response classification data; Based on the promotion response classification data, the group response data is fuzzified, the promotion response data is divided into intervals according to preset fuzzy rules, the response degree is divided into low, medium and high levels, and fuzzy promotion response information is generated; Based on the fuzzy promotion response information, the response strength of the current group is evaluated, the attractiveness of the promotion strategy to each group is analyzed, the response probability of the future promotion strategy is calculated, and the promotion effect prediction result is generated.

7. The retail promotion optimization method based on big data according to claim 6 is characterized by: To evaluate the response intensity of the current group, the formula is used: Calculate the response probability P of future promotion strategies r ; Among them, R i represents the response intensity of the i-th group in the historical promotion μ i is the fuzzy membership, V i represents the historical purchase volume of the i-th group, and n is the number of groups participating in the promotion.

8. The retail promotion optimization method based on big data according to claim 1, characterized in that: Using the promotion effect prediction results, multiple market scenarios are set for the promotion strategy, random simulation is performed, and the fluctuation of the promotion effect under each scenario is analyzed. The specific steps for generating promotion strategy performance data are as follows: Based on the promotion effect prediction results, a virtual market scenario is set through historical market data and consumer behavior data, market variables such as demand fluctuations, price sensitivity and seasonal change parameters are combined, a corresponding scenario model is established, and a virtual market scenario parameter set is generated; Based on the virtual market scenario parameter set, adjusting the promotion strategy parameters under each scenario model, including setting the discount range, promotion time and target commodity category, simulating the effect of the promotion strategy under multiple scenarios using logic rules, and generating virtual promotion strategy simulation results; Based on the simulation results of the virtual promotion strategy, the promotion strategy performance in each scenario is analyzed, the cost-effectiveness and revenue fluctuations are evaluated, and by comparing the strategy performance in multiple scenarios, the best performing strategy combination is determined to generate promotion strategy performance data.

9. The retail promotion optimization method based on big data according to claim 1, characterized in that: The specific steps to adjust the promotion discount depth, duration and target consumer group through the promotion strategy performance data, and reallocate promotion resources and budget according to the simulation results to generate an optimized promotion strategy are as follows: Based on the promotion strategy performance data, analyzing the cost-effectiveness ratio and resource utilization efficiency, screening multiple promotion strategies according to the benefit ranking, and generating screened promotion strategy data; Based on the filtered promotion strategy data, adjust the discount depth, promotion duration and target consumer group, match resources with strategies, and reallocate budgets to generate adjusted promotion resource configurations; Based on the adjusted promotion resource configuration, the promotion parameters are adjusted, and the key factors of promotion discounts, duration and target groups are optimized in combination with market data and consumer feedback. The resource allocation is matched with the promotion goals to generate an optimized promotion strategy.

10. A retail promotion optimization system based on big data, characterized in that: According to any one of claims 1 to 9, the retail promotion optimization method based on big data is implemented, and the system comprises: The data collection module collects various promotion and transaction data based on the retail merchant network, performs data standardization and preliminary cleaning, including removing invalid and duplicate data, and generates standardized promotion transaction data; The data encryption and synchronization module performs data encryption operations based on the standardized promotion transaction data, synchronizes the encrypted data between nodes through blockchain technology, and generates shared promotion activity data; The consumer analysis module groups consumers based on the shared promotion activity data and their purchase history records, and labels the purchase behavior and consumption habits of each group to generate consumer behavior grouping results; The smart contract building module builds a smart contract on the blockchain platform based on the consumer behavior grouping results, automatically executes the verification and management of promotion conditions, including the logical judgment and selection of promotion conditions, monitors consumers' responses to promotion activities, and generates promotion activity execution records; The strategy optimization module analyzes consumer purchase frequency, price sensitivity and promotion response data based on the promotion activity execution records, processes these data through fuzzy logic, evaluates the consumer group's response to the promotion strategy, conducts market simulation, optimizes promotion strategy parameters, including discount depth and promotion duration, and generates an optimized promotion strategy.

Images