Product sales prediction system and method based on big data analysis

Abstract

The invention relates to the technical field of product management, and discloses a product sales prediction system and method based on big data analysis, and the system comprises a multi-dimensional collection module and an intelligent prediction module. According to the system, development historical data of all cultures and sales records of all IP culture products are obtained through a multi-dimensional acquisition module and classified to form a data set, an intelligent prediction module analyzes discussion popularity of each IP culture and generates a popularity coefficient, a dynamic basis is provided for prediction, meanwhile, parameter weights are flexibly configured and can be adjusted according to industry characteristics, and the prediction efficiency is improved. According to the method, the multi-dimensional analysis adaptability is high, the intelligent prediction module analyzes the inventory and product sales change at each time point, generates a monitoring data set and provides data support for follow-up dynamic matching of an inventory threshold value, and the intelligent prediction module judges the development stage of an IP culture life cycle, the popularity degree of IP culture products, the market saturation degree and consumer preferences. And the IP stages are automatically divided, differential strategies are formulated, and the sales volume prediction accuracy is high.

Description

Technical Field

[0001] The present invention relates to the field of product management technology, and in particular to a product sales forecasting system and method based on big data analysis. Background Art

[0002] In recent years, the blind box economy, centered around the concept of "unexpected surprises," has swept the global consumer market. Blind boxes, specifically those integrating popular IPs from animation, film, television, and gaming, embody the combined emotional value and scarcity of these IPs. Their unique cultural consumption experience has become a trendsetter among young people. This business model infuses IP cultural DNA into physical goods, successfully amplifying consumers' expectations for both emotional resonance and collectible value through unpredictable design at the time of purchase.

[0003] The logic of the blind box economy encompasses collectible and social attributes, hypermarketing strategies, and IP collaborations. Serialized products stimulate a desire to collect, and the social media "unboxing" and reviewing foster a consumer culture with a sense of ritual. Furthermore, hypermarketing strategies, such as hidden editions with a 1 / 144 chance of appearing, create scarcity to stimulate repeat purchases. Cross-IP collaborations achieve two-way value creation, precisely reaching the IP's original fan base and enhancing product premiums through cultural empowerment. Amidst increasingly competitive markets, some brands are actively promoting technological innovations such as virtual blind boxes and AR interactions to enhance the user experience. Furthermore, they are also entering the high-end collectibles market through limited-edition artist collaborations, demonstrating the potential for growth through the integration of cultural soft power and commercial innovation. Balancing IP content development, technological innovation, and consumer experience has become crucial to the industry's sustainable development.

[0004] At present, traditional product sales forecasting systems have obvious shortcomings when dealing with IP culture-related businesses. Due to the rapid changes in IP cultural trends and the short product design cycle, it is easy to become outdated and difficult to flexibly adapt to market dynamics. This leads to a significant increase in the sales risk of unsold inventory. In addition, the demand for limited products fluctuates extremely sharply, and the explosive demand is strong in the short term. If the inventory is insufficient, it will have an adverse impact on sales. If the inventory is excessive, it will occupy a large amount of cash flow, and the accuracy of product sales forecasts is low. Summary of the Invention

[0005] (1) Technical problems solved

[0006] In response to the shortcomings of the existing technology, the present invention provides a product sales forecasting system and method based on big data analysis, which has the advantages of strong adaptability to multi-dimensional analysis and high sales forecasting accuracy. It solves the problem that traditional product sales forecasting systems are difficult to flexibly adapt to market changes and have low product sales forecasting accuracy.

[0007] (2) Technical solution

[0008] To achieve the above-mentioned object, the present invention provides the following technical solutions: a product sales forecasting system based on big data analysis, comprising a multi-dimensional acquisition module and an intelligent forecasting module;

[0009] The multi-dimensional acquisition module consists of a cultural data unit and a sales data unit. The cultural data unit collects a cultural data set through a network connection to a big data platform. The cultural data set includes the development history data of all IP cultures. The sales data unit collects a product data set through a network connection to a database. The product data set includes the sales records of all IP culture production lines.

[0010] The intelligent prediction module consists of a heat evaluation unit, a trend analysis unit and a decision management unit. The heat evaluation unit analyzes the discussion heat of each IP culture based on the cultural data set and generates a corresponding heat coefficient Rdx. The trend analysis unit is set with a monitoring period Q of a fixed length, and then combined with the sales data set, analyzes the inventory KL and product sales changes at each time point, and generates a corresponding monitoring data group Jcsj. The decision management unit is set with a fixed minimum inventory threshold DKL, a maximum inventory threshold GKL and a heat threshold RDY, and then combined with the heat coefficient Rdx, inventory KL and monitoring data group Jcsj to judge the development stage of the IP culture life cycle, the popularity of IP culture products, market saturation and consumer preferences, and generate corresponding management suggestions.

[0011] Preferably, the expression of the cultural data set is {W1 d 、W2 d 、W3 d 、...、Wn d}, W1 d To Wn d It represents the development history data of the first nth IP culture, including the number of characters, number of fans, number of online exposures, and amount of online discussions. d represents the birth time of the IP culture.

[0012] Preferably, the expression of the product data set is {C1 s 、C2 s 、C3 s ,..., Cn s}, C1 s To Cn s Represents the sales records of the first to nth IP cultural production lines. The IP cultural production line sales records include the number of versions, the number of models, daily sales volume, and average order value. s represents the time point when the IP cultural production line sales records are obtained. Each IP culture corresponds to an IP cultural production line.

[0013] Preferably, the heat coefficient Rdx calculation process is as follows:

[0014] According to the cultural data set, extract the development history data of the i-th IP culture and mark the number of characters in the i-th IP culture as js i , mark the number of fans of the i-th IP culture as fs i , mark the number of network exposures of IP culture i as bg i , mark the online discussion volume of IP culture i as tl i , mark the birth time of the i-th IP culture as i d ;

[0015]

[0016] In the formula, α1 represents the weight for the number of characters, α2 represents the weight for the number of fans, α3 represents the weight for the ratio of online discussion volume to online exposure times, DT represents the current time point, DT-i d represents the birth time of the i-th IP culture, α4 represents the weight for the birth time, α1, α2, α3 and α4 are all constants, and α1+α2+α3+α4=1, Indicates that the heat coefficient Rdx of the i-th IP culture is calculated according to the weights α1, α2, α3 and α4 i .

[0017] Preferably, the inventory KL calculation process is as follows:

[0018] According to the product data set, the sales records of the i-th IP cultural product line in the monitoring period Q are counted, and the number of versions of the i-th IP cultural product line in the monitoring period Q is marked as {b1 i 、b2 i 、b3 i ,...,bx i}, b1 i to ba i Indicates the number of versions from the first time point to the xth time point. The number of model types of the i-th IP cultural production line in the monitoring period Q is marked as {m1 i 、m2 i 、m3 i ,...,mx i}, m1 i to mx i Indicates the number of model types from the first time point to the xth time point;

[0019]

[0020] In the formula, be iIndicates the number of versions of the i-th IP cultural production line at the e-th time point within the monitoring period Q, me i represents the number of model types of the i-th IP cultural production line at the e-th time point within the monitoring period Q, be i ×me i It represents the number of SKUs of the i-th IP cultural production line at the e-th time point within the monitoring period Q. It represents the number of SKUs at all time points of the i-th IP cultural production line within the comprehensive calculation monitoring period Q, which is the inventory KL of the i-th IP cultural production line. i .

[0021] Preferably, the calculation process of the monitoring data group Jcsj is as follows:

[0022] According to the product data set, the sales records of the i-th IP cultural production line in the monitoring period Q are counted, and the single-day sales volume of the i-th IP cultural production line in the monitoring period Q is marked as {r1 i 、r2 i 、r3 i ,...,rx i}, r1 i to rx i represents the daily sales volume from the first time point to the xth time point, and the customer unit price of the i-th IP cultural production line in the monitoring period Q is marked as {j1 i 、j2 i 、j3 i ,...,jx i}, j1 i To jx i Indicates the average order value from the first time point to the xth time point;

[0023]

[0024] In the formula, It represents the average daily sales volume of the i-th IP cultural production line during the monitoring period Q. represents the average price per customer of the i-th IP cultural production line during the monitoring period Q, rf i represents the daily sales volume at the fth time point within the monitoring period Q, Indicates calculating the difference between the daily sales volume and the average value at each time point in turn, jf i represents the average order value at the fth time point within the monitoring period Q, According to the standard deviation formula, the standard deviation of the customer unit price of the i-th IP cultural production line within the monitoring period Q is calculated, Jcsj i Represents the monitoring data set of the i-th IP culture production line.

[0025] Preferably, if the birth time of a single IP culture is less than or equal to 1 year, it means that the life cycle of the IP culture is in the growth stage, and the monthly inventory KL in the growth stage should match the minimum inventory threshold DKL. If the birth time of a single IP culture is between 1 and 5 years, it means that the life cycle of the IP culture is in the mature stage, and the monthly inventory KL in the mature stage should be maintained within the range of the minimum inventory threshold DKL and the maximum inventory threshold GKL. The birth time of the single IP culture exceeds or is equal to 5 years, which means that the life cycle of the single IP culture is in the decline stage, and the monthly inventory KL in the decline stage should be lower than the minimum inventory threshold DKL.

[0026] Preferably, if the heat coefficient Rdx of the IP culture in the growth stage of its life cycle exceeds the heat threshold RDY, it means that the IP cultural product is highly popular and the number of fans is growing rapidly, and it is recommended to increase production by 10%. If the heat coefficient Rdx of the IP culture in the mature stage of its life cycle exceeds the heat threshold RDY, it means that the IP cultural product is highly popular and the fan repurchase rate is high, and it is recommended to increase production by 30%. If the heat coefficient Rdx of the IP culture in the decline stage of its life cycle exceeds the heat threshold RDY, it means that the IP cultural product is highly popular and has a high collection value, and it is recommended to increase production by 20%.

[0027] Preferably, in the monitoring data group Jcsj, when the difference between the daily sales volume and the average value shows an inverse relationship over time, it indicates that the market saturation is high and it is recommended to strengthen marketing activities; when the standard deviation of the average price per customer of the IP cultural production line shows a positive relationship over time, it indicates that consumer preferences have changed and it is recommended to update the product version.

[0028] The product sales forecasting method based on big data analysis includes the following steps:

[0029] Step 1: Connect the big data platform and database through the network to obtain the development history data of all cultures and the sales records of all IP cultural products, and classify them into cultural data sets and product data sets;

[0030] Step 2: Based on the cultural data set, analyze the discussion heat of each IP culture and generate the corresponding heat coefficient Rdx;

[0031] Step 3: Set a fixed-length monitoring period Q, and then combine the sales data set to analyze the inventory level KL and product sales changes at each time point, and generate the corresponding monitoring data set Jcsj;

[0032] Step 4: Set fixed values ​​for the minimum inventory threshold DKL, the maximum inventory threshold GKL, and the popularity threshold RDY. Then, combined with the popularity coefficient Rdx, inventory volume KL, and monitoring data group Jcsj, determine the development stage of the IP culture life cycle, the popularity of IP culture products, market saturation, and consumer preferences, and generate corresponding management recommendations.

[0033] Compared with the existing technology, the present invention provides a product sales forecasting system and method based on big data analysis, which has the following beneficial effects:

[0034] 1. The present invention connects the big data platform and database through a multi-dimensional acquisition module network to obtain the development history data of all cultures and the sales records of all IP cultural products, and classifies them into cultural data sets and product data sets. The intelligent prediction module analyzes the discussion heat of each IP culture based on the cultural data set, and generates a corresponding heat coefficient Rdx. It integrates indicators such as character update rate, fan scale, exposure and discussion volume ratio, scientifically quantifies the real-time heat of IP culture, and provides a dynamic basis for prediction. At the same time, the parameter weights can be flexibly configured and adjusted according to industry characteristics. It is suitable for multiple fields such as film and television, games, and cultural and creative industries, and has strong multi-dimensional analysis adaptability.

[0035] 2. The present invention sets a fixed-length monitoring period Q through the intelligent prediction module, and then combines the sales data set to analyze the inventory KL and product sales changes at each time point, and generates a corresponding monitoring data group Jcsj, which provides data support for the subsequent dynamic matching of inventory thresholds. The intelligent prediction module sets a fixed minimum inventory threshold DKL and a maximum inventory threshold GKL and a heat threshold RDY, and then combines the heat coefficient Rdx, inventory KL and monitoring data group Jcsj to judge the development stage of the IP culture life cycle, the popularity of IP cultural products, market saturation and consumer preferences, and generates corresponding management suggestions, automatically divides the IP stage, and formulates differentiation strategies, which significantly improves the accuracy of sales forecasts and decision-making efficiency, and the sales forecast has high accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a flow chart of the system of the present invention;

[0037] Figure 2 It is a step diagram of the method of the present invention. DETAILED DESCRIPTION

[0038] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0039] Traditional product sales forecasting systems have obvious shortcomings when dealing with IP culture-related businesses. Due to the rapid changes in IP culture trends and the short product design cycle, products are easily outdated and difficult to flexibly adapt to market dynamics. This leads to a significant increase in the risk of unsold inventory. In addition, the demand for limited-edition products fluctuates violently, with strong explosive demand in the short term. If there is insufficient inventory, it will have an adverse impact on sales. If there is an overstocking, it will occupy a large amount of cash flow, and the accuracy of product sales forecasts is low. Therefore, a product sales forecasting system and method based on big data analysis is provided. Please refer to Figure 1-Figure 2 , a product sales forecasting system based on big data analysis, including a multi-dimensional acquisition module and an intelligent forecasting module;

[0040] The multi-dimensional collection module consists of cultural data units and sales data units, which collects comprehensive data and avoids the one-sidedness of a single data source;

[0041] The cultural data unit collects cultural data sets through the network connection to the big data platform. The cultural data sets include the development history data of all IP cultures. The expression of the cultural data sets is {W1 d 、W2 d 、W3 d 、...、Wn d}, W1 d To Wn d represents the development history data of the first nth IP culture, including the number of characters, number of fans, number of online exposures, and amount of online discussions. d represents the birth time of the IP culture.

[0042] Specifically, the more characters an IP culture has, the wider its corresponding fan base tends to be. The number of online exposures is calculated based on the number of advertisements, while the number of online discussions is calculated based on the number of posts on related topics on the big data platform.

[0043] The sales data unit collects product data sets through the network connection database. The product data sets include the sales records of all IP cultural product lines. The expression of the product data set is {C1 s 、C2 s 、C3 s ,..., Cn s}, C1 s To Cn s Represents the sales records of the first to nth IP cultural product lines. The IP cultural product line sales records include the number of versions, number of models, daily sales volume, and average order value. s represents the time point at which the IP cultural product line sales records were obtained. Each IP culture corresponds to an IP cultural product line.

[0044] Specifically, the number of versions of IP cultural products, that is, the number of version series of blind box products. The higher the popularity of IP culture, the more versions there are. The number of model types of IP cultural products refers to the number of models included in the blind box version series products. Usually, there are 6 models in a single version series of blind box products.

[0045] The intelligent prediction module consists of a popularity evaluation unit, a trend analysis unit, and a decision management unit. The popularity evaluation unit analyzes the discussion popularity of each IP culture based on the cultural data set and generates the corresponding popularity coefficient Rdx. The calculation process is as follows:

[0046] According to the cultural data set, extract the development history data of the i-th IP culture and mark the number of characters in the i-th IP culture as js i , mark the number of fans of the i-th IP culture as fs i , mark the number of network exposures of IP culture i as bg i , mark the online discussion volume of IP culture i as tl i , mark the birth time of the i-th IP culture as i d ;

[0047]

[0048] In the formula, α1 represents the weight for the number of characters, α2 represents the weight for the number of fans, and α3 represents the weight for the ratio of online discussion volume to online exposure times. This can identify potential IPs with “high discussion and low exposure” and predict the trend of hot products in advance. DT represents the current time point, and DT-i d represents the birth time of the i-th IP culture, α4 represents the weight for the birth time, α1, α2, α3 and α4 are all constants, and α1+α2+α3+α4=1, Indicates that the heat coefficient Rdx of the i-th IP culture is calculated according to the weights α1, α2, α3 and α4 i , comprehensively considers indicators such as character update rate, fan base, exposure and discussion ratio, and scientifically quantifies the real-time popularity of IP culture, providing a dynamic basis for prediction;

[0049] The trend analysis unit is set with a fixed-length monitoring period Q. It then combines the sales data set to analyze the inventory level KL and product sales changes at each time point and generates the corresponding monitoring data set Jcsj;

[0050] The calculation process of inventory KL is as follows:

[0051] According to the product data set, the sales records of the i-th IP cultural product line in the monitoring period Q are counted, and the number of versions of the i-th IP cultural product line in the monitoring period Q is marked as {b1i 、b2 i 、b3 i ,...,bx i}, b1 i to ba i Indicates the number of versions from the first time point to the xth time point. The number of model types of the i-th IP cultural production line in the monitoring period Q is marked as {m1 i 、m2 i 、m3 i ,...,mx i}, m1 i to mx i Indicates the number of model types from the first time point to the xth time point;

[0052]

[0053] In the formula, be i Indicates the number of versions of the i-th IP cultural production line at the e-th time point within the monitoring period Q, me i represents the number of model types of the i-th IP cultural production line at the e-th time point within the monitoring period Q, be i ×me i It represents the number of SKUs of the i-th IP cultural production line at the e-th time point within the monitoring period Q. It represents the number of SKUs at all time points of the i-th IP cultural production line within the comprehensive calculation monitoring period Q, which is the inventory KL of the i-th IP cultural production line. i , assisting enterprises to accurately plan storage space and production plans, and providing data support for dynamic matching of inventory thresholds according to the growth, maturity, and decline stages of IP culture;

[0054] The calculation process of Jcsj for the monitoring data group is as follows:

[0055] According to the product data set, the sales records of the i-th IP cultural production line in the monitoring period Q are counted, and the single-day sales volume of the i-th IP cultural production line in the monitoring period Q is marked as {r1 i 、r2 i 、r3 i ,...,rx i}, r1 i to rx i represents the daily sales volume from the first time point to the xth time point, and the customer unit price of the i-th IP cultural production line in the monitoring period Q is marked as {j1 i 、j2 i 、j3 i ,...,jx i}, j1 i To jx iIndicates the average order value from the first time point to the xth time point;

[0056]

[0057] In the formula, It represents the average daily sales volume of the i-th IP cultural production line during the monitoring period Q. represents the average price per customer of the i-th IP cultural production line during the monitoring period Q, rf i represents the daily sales volume at the fth time point within the monitoring period Q, Indicates calculating the difference between the daily sales volume and the average value at each time point. If the daily sales volume is continuously lower than the average, it is recommended to strengthen marketing. i represents the average order value at the fth time point within the monitoring period Q, According to the standard deviation formula, the standard deviation of the customer unit price of the i-th IP cultural production line within the monitoring period Q is calculated, Jcsj i The monitoring data set for the i-th IP cultural production line indicates that when the standard deviation increases, it prompts you to update the product version and recommends launching a limited edition, successfully attracting collectors and increasing profit margins.

[0058] The decision-making management unit sets fixed values ​​for the minimum inventory threshold DKL, the maximum inventory threshold GKL, and the popularity threshold RDY. Combining the popularity coefficient Rdx, the inventory level KL, and the monitoring data set Jcsj, it determines the development stage of the IP culture life cycle, the popularity of IP culture products, market saturation, and consumer preferences, and generates corresponding management recommendations.

[0059] If a single IP culture is less than or equal to 1 year old, it means that the life cycle of the IP culture is in the growth stage. The monthly inventory volume KL in the growth stage should match the minimum inventory threshold DKL. If a single IP culture is between 1 and 5 years old, it means that the life cycle of the IP culture is in the mature stage. The monthly inventory volume KL in the mature stage should be maintained within the range of the minimum inventory threshold DKL and the maximum inventory threshold GKL. If a single IP culture is more than or equal to 5 years old, it means that the life cycle of the IP culture is in the decline stage. The monthly inventory volume KL in the decline stage should be lower than the minimum inventory threshold DKL to reduce the probability of inventory backlog or missed sales opportunities.

[0060] If the heat coefficient Rdx of an IP culture in the growth stage of its life cycle exceeds the heat threshold RDY, it indicates that the IP culture product is highly popular and the number of fans is growing rapidly. It is recommended to increase production by 10%. If the heat coefficient Rdx of an IP culture in the mature stage of its life cycle exceeds the heat threshold RDY, it indicates that the IP culture product is highly popular and has a high fan repurchase rate. It is recommended to increase production by 30%. If the heat coefficient Rdx of an IP culture in the decline stage of its life cycle exceeds the heat threshold RDY, it indicates that the IP culture product is highly popular and has a high collection value. It is recommended to increase production by 20%.

[0061] In the monitoring data group Jcsj, when the difference between daily sales volume and the average value shows an inverse relationship over time, it indicates that the market saturation is high and it is recommended to strengthen marketing activities. When the standard deviation of the average customer price of the IP cultural production line shows a positive relationship over time, it indicates that consumer preferences have changed and it is recommended to update the product version.

[0062] The product sales forecasting method based on big data analysis includes the following steps:

[0063] Step 1: Connect the big data platform and database through the network to obtain the development history data of all cultures and the sales records of all IP cultural products, and classify them into cultural data sets and product data sets;

[0064] Step 2: Based on the cultural data set, analyze the discussion heat of each IP culture and generate the corresponding heat coefficient Rdx. Flexible parameter weights can be configured and adjusted according to industry characteristics. It is applicable to multiple fields such as film and television, games, and cultural and creative industries, and has strong adaptability to multi-dimensional analysis.

[0065] Step 3: Set a fixed-length monitoring period Q, and then combine the sales data set to analyze the inventory level KL and product sales changes at each time point, and generate the corresponding monitoring data set Jcsj;

[0066] Step 4: Set fixed values ​​for the minimum inventory threshold DKL, the maximum inventory threshold GKL, and the popularity threshold RDY. Then, combined with the popularity coefficient Rdx, inventory level KL, and monitoring data group Jcsj, determine the development stage of the IP culture life cycle, the popularity of IP culture products, market saturation, and consumer preferences, and generate corresponding management recommendations. This automatically divides the IP stage and formulates a differentiation strategy, significantly improving the accuracy of sales forecasts and decision-making efficiency, with high sales forecast accuracy.

[0067] Example 1:

[0068] In this experiment, the original IP "Little Bear Diary" was selected as the experimental object. According to statistics, the number of characters in this IP culture is 10, the number of fans is 10,000 users, the number of online exposure is 50,000 times, and the number of online discussions is 2,000 related posts. The creation time point is 2012, and the current time point is 2025. The calculation process of the heat coefficient Rdx of this IP culture is as follows:

[0069]

[0070] In the formula, α1=0.2 represents the weight for the number of characters, α2=0.2 represents the weight for the number of fans, α3=0.3 represents the weight for the ratio of online discussion volume to online exposure times, DT=2025 represents the current time point, DT-i d =2025-2012 represents the birth time of the IP culture, α4=0.3 represents the weight for the birth time, α1, α2, α3 and α4 are all constants, and 0.2+0.2+0.3+0.3=1. According to the weights of α1, α2, α3 and α4, the heat coefficient Rdx of the IP culture is calculated. i The minimum inventory threshold DKL is set to 10,000 pieces, and the popularity threshold RDY is set to 2,000. It is judged that the birth time of the IP culture has exceeded 5 years, and the life cycle of the IP culture is in the decline stage. The monthly inventory KL in the decline stage should be less than 10,000 pieces. In addition, the popularity coefficient Rdx of the IP culture in the decline stage of the life cycle i It has exceeded the popularity threshold RDY, indicating that the IP cultural product is highly popular and has high collection value. It is recommended to increase production by 20%.

[0071] Example 2:

[0072] In this experiment, the original IP Love Cat was selected as the experimental object, and the monitoring period Q was set to 3 days. According to statistics, within 3 days, the number of versions of this IP culture were 3, 5, and 5 respectively, and the number of model types were 6, 6, and 6 respectively. The inventory KL calculation process of this IP culture is as follows:

[0073]

[0074] KL i =(3×6)+(5×6)+(5×6)=78

[0075] In the formula, be i =3 means that within 3 days, the number of versions of this IP cultural production line on the first day is 3, me i Indicates that within 3 days, the number of model types of the IP cultural production line on the first day is 6, be i ×mei =3×6 represents the number of SKUs of the IP cultural production line on the first day within 3 days. It indicates the comprehensive calculation of the number of SKUs of the IP cultural production line at all time points within 3 days. 78 is the inventory KL of the IP cultural production line. i .

[0076] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. Product sales forecasting system based on big data analysis, characterized by: Including multi-dimensional acquisition module and intelligent prediction module; The multi-dimensional acquisition module consists of a cultural data unit and a sales data unit. The cultural data unit collects a cultural data set through a network connection to a big data platform. The cultural data set includes the development history data of all IP cultures. The sales data unit collects a product data set through a network connection to a database. The product data set includes the sales records of all IP culture production lines. The intelligent prediction module consists of a heat evaluation unit, a trend analysis unit and a decision management unit. The heat evaluation unit analyzes the discussion heat of each IP culture based on the cultural data set and generates a corresponding heat coefficient Rdx. The trend analysis unit is set with a monitoring period Q of a fixed length, and then combined with the sales data set, analyzes the inventory KL and product sales changes at each time point, and generates a corresponding monitoring data group Jcsj. The decision management unit is set with a fixed minimum inventory threshold DKL, a maximum inventory threshold GKL and a heat threshold RDY, and then combined with the heat coefficient Rdx, inventory KL and monitoring data group Jcsj to judge the development stage of the IP culture life cycle, the popularity of IP culture products, market saturation and consumer preferences, and generate corresponding management suggestions.

2. The product sales forecasting system based on big data analysis according to claim 1, characterized in that: The expression of the cultural data set is {W1 d 、W2 d 、W3 d 、...、Wn d }, W1 d To Wn d It represents the development history data of the first nth IP culture, including the number of characters, number of fans, number of online exposures, and amount of online discussions. d represents the birth time of the IP culture.

3. The product sales forecasting system based on big data analysis according to claim 2, characterized in that: The expression of the product data set is {C1 s 、C2 s 、C3 s ,..., Cn s }, C1 s To Cn s Represents the sales records of the first to nth IP cultural production lines. The IP cultural production line sales records include the number of versions, the number of models, daily sales volume, and average order value. s represents the time point when the IP cultural production line sales records are obtained. Each IP culture corresponds to an IP cultural production line.

4. The product sales forecasting system based on big data analysis according to claim 3, characterized in that: The heat coefficient Rdx calculation process is as follows: According to the cultural data set, extract the development history data of the i-th IP culture and mark the number of characters in the i-th IP culture as js i , mark the number of fans of the i-th IP culture as fs i , mark the number of network exposures of IP culture i as bg i , mark the online discussion volume of IP culture i as tl i , mark the birth time of the i-th IP culture as i d ; In the formula, α1 represents the weight for the number of characters, α2 represents the weight for the number of fans, α3 represents the weight for the ratio of online discussion volume to online exposure times, DT represents the current time point, DT-i d represents the birth time of the i-th IP culture, α4 represents the weight for the birth time, α1, α2, α3 and α4 are all constants, and α1+α2+α3+α4=1, Indicates that the heat coefficient Rdx of the i-th IP culture is calculated according to the weights α1, α2, α3 and α4 i .

5. The product sales forecasting system based on big data analysis according to claim 4, characterized in that: The inventory KL calculation process is as follows: According to the product data set, the sales records of the i-th IP cultural product line in the monitoring period Q are counted, and the number of versions of the i-th IP cultural product line in the monitoring period Q is marked as {b1 i 、b2 i 、b3 i ,...,bx i }, b1 i to ba i Indicates the number of versions from the first time point to the xth time point. The number of model types of the i-th IP cultural production line in the monitoring period Q is marked as {m1 i 、m2 i 、m3 i ,...,mx i }, m1 i to mx i Indicates the number of model types from the first time point to the xth time point; In the formula, be i Indicates the number of versions of the i-th IP cultural production line at the e-th time point within the monitoring period Q, me i represents the number of model types of the i-th IP cultural production line at the e-th time point within the monitoring period Q, be i ×me i It represents the number of SKUs of the i-th IP cultural production line at the e-th time point within the monitoring period Q. It represents the number of SKUs at all time points of the i-th IP cultural production line within the comprehensive calculation monitoring period Q, which is the inventory KL of the i-th IP cultural production line. i .

6. The product sales forecasting system based on big data analysis according to claim 5, characterized in that: The calculation process of the monitoring data set Jcsj is as follows: According to the product data set, the sales records of the i-th IP cultural production line in the monitoring period Q are counted, and the single-day sales volume of the i-th IP cultural production line in the monitoring period Q is marked as {r1 i 、r2 i 、r3 i ,...,rx i }, r1 i to rx i represents the daily sales volume from the first time point to the xth time point, and the customer unit price of the i-th IP cultural production line in the monitoring period Q is marked as {j1 i 、j2 i 、j3 i ,...,jx i }, j1 i To jx i Indicates the average order value from the first time point to the xth time point; In the formula, It represents the average daily sales volume of the i-th IP cultural production line during the monitoring period Q. represents the average price per customer of the i-th IP cultural production line during the monitoring period Q, rf i represents the daily sales volume at the fth time point within the monitoring period Q, Indicates calculating the difference between the daily sales volume and the average value at each time point in turn, jf i represents the average order value at the fth time point within the monitoring period Q, According to the standard deviation formula, the standard deviation of the customer unit price of the i-th IP cultural production line within the monitoring period Q is calculated, Jcsj i Represents the monitoring data set of the i-th IP culture production line.

7. The product sales forecasting system based on big data analysis according to claim 6, characterized in that: If the birth time of a single IP culture is less than or equal to 1 year, it means that the life cycle of the IP culture is in the growth stage. The monthly inventory KL in the growth stage should match the minimum inventory threshold DKL. If the birth time of a single IP culture is between 1 and 5 years, it means that the life cycle of the IP culture is in the mature stage. The monthly inventory KL in the mature stage should be maintained within the range of the minimum inventory threshold DKL and the maximum inventory threshold GKL. The birth time of the single IP culture exceeds or is equal to 5 years, which means that the life cycle of the single IP culture is in the decline stage. The monthly inventory KL in the decline stage should be lower than the minimum inventory threshold DKL.

8. The product sales forecasting system based on big data analysis according to claim 7, characterized in that: If the heat coefficient Rdx of IP culture in the growth stage of its life cycle exceeds the heat threshold RDY, it means that the IP cultural product is highly popular and the number of fans is growing rapidly. It is recommended to increase production by 10%. If the heat coefficient Rdx of IP culture in the mature stage of its life cycle exceeds the heat threshold RDY, it means that the IP cultural product is highly popular and the fan repurchase rate is high. It is recommended to increase production by 30%. If the heat coefficient Rdx of IP culture in the decline stage of its life cycle exceeds the heat threshold RDY, it means that the IP cultural product is highly popular and has a high collection value. It is recommended to increase production by 20%.

9. The product sales forecasting system based on big data analysis according to claim 8, characterized in that: In the monitoring data group Jcsj, when the difference between the daily sales volume and the average value shows an inverse relationship over time, it indicates that the market saturation is high and it is recommended to strengthen marketing activities. When the standard deviation of the average customer price of the IP cultural production line shows a positive relationship over time, it indicates that consumer preferences have changed and it is recommended to update the product version.

10. A product sales forecasting method based on big data analysis, applied to a product sales forecasting system based on big data analysis as claimed in any one of claims 1 to 9, characterized in that: The following steps are involved: Step 1: Connect the big data platform and database through the network to obtain the development history data of all cultures and the sales records of all IP cultural products, and classify them into cultural data sets and product data sets; Step 2: Based on the cultural data set, analyze the discussion heat of each IP culture and generate the corresponding heat coefficient Rdx; Step 3: Set a fixed-length monitoring period Q, and then combine the sales data set to analyze the inventory level KL and product sales changes at each time point, and generate the corresponding monitoring data set Jcsj; Step 4: Set fixed values ​​for the minimum inventory threshold DKL, the maximum inventory threshold GKL, and the popularity threshold RDY. Then, combined with the popularity coefficient Rdx, inventory volume KL, and monitoring data group Jcsj, determine the development stage of the IP culture life cycle, the popularity of IP culture products, market saturation, and consumer preferences, and generate corresponding management recommendations.

Images