Laying hen breeding early warning analysis system based on early warning index system

Abstract

The invention relates to the technical field of breeding early warning, and discloses a laying hen breeding early warning analysis system based on an early warning index system, which comprises an early warning index system construction module, an early warning index threshold setting module, an early warning index data acquisition module, an early warning index data analysis module, a personalized decision suggestion module and an early warning analysis result display module. The early warning index threshold value setting module comprises an initial early warning threshold value setting module used for providing a corresponding initial early warning threshold value for the target farmer and an optimal early warning threshold value determining module used for carrying out personalized adjustment on the initial early warning threshold value according to the historical performance and the risk bearing capacity of the target farmer. The risk management capability of farmers can be effectively improved, the market adaptability and economic benefits of the farmers are enhanced, the farmers are helped to better cope with market fluctuation and potential risks, and the overall anti-risk capability and economic benefits of the breeding industry are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of breeding early warning, and in particular to an early warning analysis system for laying hen breeding based on an early warning indicator system. Background Art

[0002] In today's egg-laying chicken farming industry, farmers have to deal with a series of complex and ever-changing market and technical challenges while pursuing economic benefits. The fluctuation of market prices, especially the fluctuation of egg prices, poses a severe test to the stability of farmers' income. The rise in feed costs, affected by the fluctuation of global raw material market prices and the instability of the supply chain, has further squeezed the profit margins of farmers. At the same time, the risk of epidemics is always there. Once an outbreak occurs, it may not only cause huge direct economic losses, but also trigger a crisis of market confidence and affect the sales of eggs. In addition, the uncertainty of farming management, including factors such as climate change, feeding technology, and labor costs, has also brought additional pressure to farmers.

[0003] The common characteristics of these challenges are that they are all unpredictable and have potential economic impacts, which may cause farmers to face serious economic losses. Therefore, in order to ensure the healthy development of the aquaculture industry and the vital interests of farmers, an effective early warning system is urgently needed to identify and predict these potential risks in advance, so that farmers can take targeted measures in time to mitigate or avoid possible negative impacts.

[0004] Traditional aquaculture early warning systems mostly rely on experienced experts to conduct risk assessments. Although this approach can provide guidance to a certain extent, it is highly subjective and easily affected by the personal experience and subjective judgment of experts. In addition, as market conditions change at an ever-increasing pace, this approach that relies on expert experience often finds it difficult to keep up with the pace of the market and cannot respond promptly to rapidly changing risks.

[0005] In addition, existing early warning systems often lack personalization. They usually adopt a one-size-fits-all approach and cannot provide customized early warning analysis based on the specific circumstances of each farmer, such as farming scale, regional characteristics, egg-laying chicken breeds, etc. This lack of personalized early warning system may result in the early warning information received by farmers not being consistent with their actual situation, thus affecting the effectiveness of the early warning and the quality of farmers' decision-making.

[0006] Therefore, in order to overcome the limitations of traditional early warning systems, the present invention provides an early warning analysis system for laying hen farming based on an early warning indicator system. Summary of the invention

[0007] 1. Technical issues to be solved

[0008] In view of the shortcomings of the prior art, the present invention provides an early warning analysis system for laying hen farming based on an early warning indicator system, which can set early warning indicator thresholds in an intelligent, personalized, dynamic and comprehensive manner, effectively improve the risk management capabilities of farmers, enhance the market adaptability and economic benefits of farmers, and thus solve the problems in the background technology.

[0009] (II) Technical solution

[0010] In order to achieve the above-mentioned intelligent, personalized, dynamic and comprehensive setting of early warning indicator thresholds, effectively improve the risk management capabilities of farmers, and enhance the market adaptability and economic benefits of farmers, the specific technical solutions adopted by the present invention are as follows:

[0011] An early warning analysis system for laying hen farming based on an early warning indicator system, comprising an early warning indicator system construction module, an early warning indicator threshold setting module, an early warning indicator data collection module, an early warning indicator data analysis module, a personalized decision-making suggestion module and an early warning analysis result display module;

[0012] Among them, the early warning indicator system construction module is used to construct early warning indicators for early warning analysis of laying hen farming. The early warning indicators include the rate of change of egg ex-factory price, the equilibrium point of egg-feed ratio and the net profit per kilogram of laying hen;

[0013] The warning indicator threshold setting module includes an initial warning threshold setting module and an optimal warning threshold determination module;

[0014] The initial warning threshold setting module is used to use the indicator analysis algorithm to provide the initial warning threshold of the corresponding warning indicator for the target farmers according to the farmers' breeding scale, breeding area, breeding entity type and breeding egg-laying chicken breed;

[0015] The optimal warning threshold determination module is used to adjust the initial warning threshold according to the historical performance and risk tolerance of the target farmers to obtain the optimal warning threshold for each warning indicator;

[0016] The early warning indicator data collection module is used to obtain real-time early warning indicator data on the rate of change of egg ex-factory prices, the equilibrium point of the egg-feed ratio, and the net profit per kilogram of laying hens;

[0017] The early warning indicator data analysis module is used to generate corresponding early warning signals based on the comparison and analysis results of the real-time early warning indicator data and the corresponding optimal early warning threshold, and promptly notify the farmers;

[0018] Personalized decision-making suggestion module, used to provide personalized decision-making suggestions to farmers based on early warning signals;

[0019] The early warning analysis result display module is used to provide an intuitive user interface, display real-time data, early warning information and decision-making suggestions, and regularly generate breeding benefit analysis reports and early warning statistical reports.

[0020] Furthermore, the initial warning threshold setting module includes the following steps when using the indicator analysis algorithm to provide the target farmers with the initial warning threshold of the corresponding warning indicators according to the farmers' breeding scale, breeding area, breeding entity type and breeding egg-laying chicken breed:

[0021] Based on the database, the historical breeding data of farmers are obtained and the data is cleaned and preprocessed. The historical breeding data includes breeding scale, breeding area, breeding entity type, breeding egg-laying chicken breeds and corresponding early warning indicator data;

[0022] Construct a scoring matrix for farmers and early warning indicators based on the farmers’ breeding history and results; analyze the scoring trend of farmers on early warning indicators, identify farmers’ scoring habits, and adjust the scoring matrix for farmers and early warning indicators based on the scoring trend;

[0023] Based on the preset indicator weights, the comprehensive similarity between the target farmer and the historical farmers is calculated, and a preset number of similar farmers are selected for the target farmer according to the ranking results of the comprehensive similarity;

[0024] According to the geographical location characteristics between the target farmers and similar farmers, the scores of similar farmers on the early warning indicators are adjusted, and the influence of similar farmers on the target farmers decreases with the increase of distance;

[0025] Based on the comparison results of the adjusted warning indicator scores and the preset score thresholds, a recommended list of warning indicators is generated for the target breeders, and the initial warning thresholds of the warning indicators are set for the target breeders according to the recommended list.

[0026] Furthermore, the scoring trend of farmers on early warning indicators is analyzed, the scoring habits of farmers are identified, and the scoring matrix of farmers and early warning indicators is adjusted according to the scoring trend, including:

[0027] Obtain the historical scoring data of the early warning indicators by the farmers and perform cleaning and standardization processing, where the scoring data includes timestamp, farmer ID, early warning indicator name and corresponding scoring value;

[0028] Based on time series analysis technology, analyze the changing trend of farmers' scores on early warning indicators over time, and use visualization tools to intuitively display the score trends;

[0029] Statistically analyze the scoring habits of farmers, and classify farmers according to their scoring habits based on the clustering results;

[0030] Adjust the scoring matrix of farmers and early warning indicators based on scoring trends and scoring habits.

[0031] Furthermore, the calculation formula of the comprehensive similarity is:

[0032] S composite (u,v)=ω 1 ·S scale (u,v)+ω 2 ·S location (u,v)+ω 3 ·S type (u,v)+ω 4 ·S breed (u,v)

[0033] In the formula, S composite (u,v) represents the comprehensive similarity between farmers u and v; S scale (u,v) represents the similarity of the breeding scale between farmers u and v; S location (u,v) represents the similarity of the breeding areas between breeders u and v; S type (u,v) represents the similarity of the farming entity types between farmers u and v; S breed (u,v) represents the similarity of laying hen breeds between farmers u and v; ω 1 ,ω 2 ,ω 3 ,ω 4 They respectively represent the weights of breeding scale, breeding area, breeding entity type and laying hen breed.

[0034] Furthermore, according to the geographical location characteristics between the target farmers and similar farmers, the scores of similar farmers on the early warning indicators are adjusted, including:

[0035] Obtain the geographic location data of the target farmer and similar farmers, and calculate the geographic distance between the target farmer and each similar farmer;

[0036] The attenuation function is used to calculate the attenuation factor of the impact of each similar farmer on the score of the target farmer, and the original score of each similar farmer is multiplied by the corresponding attenuation factor to obtain the adjusted early warning indicator score;

[0037] The expression of the attenuation factor is:

[0038]

[0039] In the formula, γ represents the attenuation factor; d represents the distance between the target farmer and similar farmers; σ represents the standard deviation; and e represents the base of the natural logarithm.

[0040] Furthermore, the optimal warning threshold determination module adjusts the initial warning threshold in a personalized manner according to the historical performance and risk tolerance of the target farmers to obtain the optimal warning threshold for each warning indicator, including:

[0041] Determine whether the target farmer is breeding for the first time. If so, use the initial warning threshold as the optimal warning threshold. If not, calculate the statistical indicators of the mean, standard deviation and coefficient of variation of each warning indicator in the historical breeding data of the target farmer respectively;

[0042] Based on the calculation results of statistical indicators, analyze the historical performance of target farmers on different early warning indicators, and calculate the historical performance score of target farmers according to the preset scoring criteria;

[0043] Assess the risk tolerance of target farmers and quantify their risk tolerance, which includes financial status, risk management measures and responses to historical risk events;

[0044] Based on the historical performance and risk tolerance assessment results of the target farmers, the personalized adjustment factor is calculated, and the initial warning threshold is adjusted according to the personalized adjustment factor to obtain the optimal warning threshold for each warning indicator.

[0045] Furthermore, based on the calculation results of statistical indicators, the historical performance of the target farmers on different early warning indicators is analyzed, and the historical performance scores of the target farmers are calculated according to the preset scoring criteria, including:

[0046] Based on the comparison results of the actual values ​​of the statistical indicators with the industry benchmark values, the performance of the target breeders on different early warning indicators is analyzed, and the historical performance scores of the target breeders on each early warning indicator are calculated according to the preset scoring standards. At the same time, the historical performance scores of the target breeders are calculated in combination with the preset early warning indicator weights.

[0047] Furthermore, the calculation formula for the historical performance score is:

[0048]

[0049] Where P l represents the historical performance score of the target farmer; n represents the number of early warning indicators; θ i represents the weight of the i-th early warning indicator; P i Represents the historical performance score of the target farmer on the i-th early warning indicator.

[0050] Furthermore, the target farmers' risk tolerance is assessed and the risk tolerance is quantitatively scored, including:

[0051] Obtain the target farmers’ financial data, risk management measures, and historical records of risk events and responses;

[0052] Analyze financial statements, calculate financial ratios, and assess financial stability and solvency;

[0053] Assess the effectiveness and comprehensiveness of the target farmers’ risk management measures and analyze their ability to identify, assess and respond to potential risks;

[0054] Analyze the risk events that the target farmers have encountered in the past and the effectiveness of their response measures, and assess the target farmers' ability to recover from risk events;

[0055] Construct a quantitative scoring model, incorporate financial status, risk management measures and historical risk event responses into the quantitative scoring model, and set scoring criteria and weights for each dimension;

[0056] A quantitative scoring model is used to score each dimension of the target farmers, and the weighted total score is calculated to obtain the risk tolerance score.

[0057] Furthermore, the expression of the optimal warning threshold is:

[0058] C′=C·λ

[0059] λ=α·P l +(1-α)·P f

[0060] In the formula, C' represents the optimal warning threshold; C represents the initial warning threshold; λ represents the personalized adjustment factor; α represents the scoring weight coefficient; P l , P f They represent the historical performance score and risk tolerance score of the target farmers respectively.

[0061] (III) Beneficial effects

[0062] Compared with the prior art, the present invention provides an early warning analysis system for laying hen farming based on an early warning indicator system, which has the following beneficial effects:

[0063] (1) The present invention can not only intelligently construct an early warning indicator system suitable for each farmer according to factors such as the farmer's farming scale, farming area, farming entity type and laying hen breed, but also can personalize the early warning threshold according to the farmer's historical performance and risk tolerance, thereby effectively improving the accuracy and adaptability of the early warning, helping farmers to better cope with market fluctuations and potential risks, and thus improving the overall risk resistance and economic benefits of the farming industry.

[0064] (2) The present invention firstly constructs a scoring matrix of farmers and early warning indicators by analyzing the farmers' historical breeding data, and analyzes the scoring trend of farmers for early warning indicators, so that the setting of early warning thresholds is more scientific and reliable. Secondly, similar farmers are selected for target farmers by calculating the comprehensive similarity between target farmers and historical farmers, thereby improving the accuracy and applicability of early warning thresholds. At the same time, the scores of similar farmers are adjusted by considering the geographical location characteristics between target farmers and similar farmers, so that the setting of early warning thresholds is closer to the actual situation of farmers. In addition, the initial early warning threshold can be personalized according to the historical performance and risk tolerance of target farmers to obtain the optimal early warning threshold of each early warning indicator, improve the flexibility and adaptability of early warning, and make the setting of early warning thresholds more comprehensive, accurate, objective and quantitative, so that the threshold of early warning indicators can be set intelligently, personalized, dynamically and comprehensively, thereby effectively improving the risk management ability of farmers and enhancing the market adaptability and economic benefits of farmers.

[0065] (3) The present invention can not only obtain the data of key early warning indicators in real time and generate timely early warning signals through the data analysis module to help farmers respond quickly to market changes, but also provide personalized decision-making suggestions based on the early warning signals to enhance the market competitiveness of farmers.

[0066] (4) The present invention enables farmers to easily view real-time data, early warning information and decision-making suggestions through an intuitive user interface. At the same time, regular breeding benefit analysis reports and early warning statistical reports provide farmers with detailed data support, thereby helping farmers to better cope with market changes and potential risks, and improve breeding benefits and risk resistance. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0068] Figure 1 It is a structural block diagram of an early warning analysis system for laying hen farming based on an early warning indicator system according to an embodiment of the present invention.

[0069] In the figure:

[0070] 1. Early warning indicator system construction module; 2. Early warning indicator threshold setting module; 3. Early warning indicator data collection module; 4. Early warning indicator data analysis module; 5. Personalized decision-making recommendation module; 6. Early warning analysis result display module. DETAILED DESCRIPTION

[0071] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in the field should be able to understand other possible implementations and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0072] According to an embodiment of the present invention, a laying hen breeding early warning analysis system based on an early warning indicator system is provided.

[0073] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, the laying hen farming early warning analysis system based on the early warning indicator system according to the embodiment of the present invention includes an early warning indicator system construction module 1, an early warning indicator threshold setting module 2, an early warning indicator data collection module 3, an early warning indicator data analysis module 4, a personalized decision suggestion module 5 and an early warning analysis result display module 6;

[0074] Among them, the early warning indicator system construction module 1 is used to construct early warning indicators for early warning analysis of laying hen farming. The early warning indicators include the change rate of egg ex-factory price, the balance point of egg-feed ratio and the net profit per kilogram of laying hen;

[0075] Specifically, the rate of change of egg ex-factory price is used to reflect the impact of changes in market supply and demand on egg prices; the egg-feed ratio equilibrium point is used to measure the ratio of feed cost to egg selling price, which is crucial to the profitability of farmers; and the net profit per kilogram of laying hens is directly related to the economic benefits of farmers.

[0076] The warning indicator threshold setting module 2 includes an initial warning threshold setting module and an optimal warning threshold determination module;

[0077] The initial warning threshold setting module is used to use the indicator analysis algorithm to provide the initial warning threshold of the corresponding warning indicator for the target farmers according to the farmers' breeding scale, breeding area, breeding entity type and breeding egg-laying chicken breed;

[0078] Specifically, the initial warning threshold setting module includes:

[0079] 1) Based on the database, obtain the historical breeding data of farmers and perform data cleaning and preprocessing to ensure the quality and integrity of the data. The historical breeding data includes breeding scale, breeding area, breeding entity type (such as enterprises, cooperatives, individual households, etc.), breeding egg-laying chicken varieties and corresponding early warning indicator data;

[0080] In this embodiment, historical breeding data is a crucial part of the laying hen breeding early warning analysis system, which provides a solid foundation for the intelligent analysis and decision-making of the system. These data cover multiple dimensions, ensuring the comprehensiveness and depth of the analysis.

[0081] First, the scale of breeding is a key indicator in historical breeding data. It not only includes the number of laying hens raised, but also involves the area of ​​the farm, the scale of facilities, etc. This information helps the system understand the operating characteristics of farmers of different sizes, so as to take into account the scale effect in early warning analysis.

[0082] Secondly, the breeding area is also an important data point. Factors such as climate conditions, market demand, and feed supply in different regions may affect the risk status of farmers.

[0083] The types of breeding entities, such as enterprises, cooperatives, and individual households, reflect the organizational form and business model of the breeders. This information helps the system identify the response capabilities and resource mobilization of different types of breeders when facing risks.

[0084] The breed of laying hens is also part of the historical breeding data. Different breeds of laying hens have different production performance, disease resistance and market preferences, which will affect the risk assessment and early warning indicator setting of farmers.

[0085] By comprehensively analyzing these historical breeding data, the laying hen breeding early warning analysis system can provide farmers with more accurate and personalized early warning services. The system can build a scoring matrix based on these data, analyze the scoring trends of farmers, identify farmers' operating habits, and adjust early warning indicators and thresholds based on this information, thereby improving farmers' risk prevention capabilities and decision-making quality.

[0086] 2) Construct a scoring matrix for farmers and early warning indicators based on the farmers’ breeding history and results; analyze the scoring trend of farmers on early warning indicators, identify farmers’ scoring habits (such as a tendency to give high or low ratings), and adjust the scoring matrix for farmers and early warning indicators based on the scoring trend to eliminate the deviation caused by inconsistent scoring habits;

[0087] Specifically, the breeding effect of farmers refers to the performance or results of farmers in different early warning indicators during the breeding process. These effects reflect the ability and achievements of farmers in preventing and managing various risk factors. Specifically, "effect" includes the following aspects:

[0088] Health status: such as the incidence of laying hen diseases, mortality rate, treatment effectiveness, etc.

[0089] Production efficiency: including egg production rate, feed conversion rate, growth rate, etc.

[0090] Environmental factors: such as the degree of pollution in the breeding environment, the effectiveness of waste treatment, etc.

[0091] Economic indicators: such as cost control, profit margin, market adaptability, etc.

[0092] When constructing the user-item rating matrix, these effects are quantified into scores of the following form:

[0093] Positive score: When a farmer performs well on a certain early warning indicator, such as low incidence of laying hen diseases and high egg production rate, he will be given a higher score.

[0094] Negative score: If the farmer has deficiencies in a certain early warning indicator, such as high laying hen mortality rate and poor cost control, he will be given a lower score.

[0095] These scores reflect the farmers’ ability to respond to different early warning indicators and their management effects, and are an important basis for constructing a scoring matrix. In this way, a scoring file can be established for each farmer that reflects his or her farming history and effects, which can then be used for subsequent early warning indicator recommendations and threshold setting.

[0096] Specifically, the scoring trend of farmers on early warning indicators is analyzed, the scoring habits of farmers are identified, and the scoring matrix of farmers and early warning indicators is adjusted according to the scoring trend, including:

[0097] Obtain the historical scoring data of the early warning indicators by the farmers and perform cleaning and standardization processing, where the scoring data includes timestamp, farmer ID, early warning indicator name and corresponding scoring value;

[0098] Based on time series analysis techniques (such as ARIMA model, time series decomposition, etc.), analyze the changing trend of farmers' scores on early warning indicators over time, and use visualization tools (such as line charts and curve charts) to intuitively display the score trends;

[0099] Statistical analysis (such as calculating the mean, median, and variance of the scores) of farmers’ scoring habits, and classifying farmers according to their scoring habits based on the clustering results;

[0100] In this embodiment, statistical analysis is an important means to understand the scoring habits of farmers. It analyzes historical scoring data to reveal the behavior patterns and tendencies of farmers in early warning indicators. The following are steps for conducting statistical analysis and classifying farmers according to their scoring habits:

[0101] Data collection and preprocessing: First, the historical scoring data of farmers on early warning indicators are collected, which includes timestamp, farmer ID, name of early warning indicator and corresponding scoring value. Then, the data is cleaned and standardized to ensure the quality and consistency of the data.

[0102] Statistical analysis:

[0103] I) Calculate the average value: Calculate the average value of the farmers' scores for each early warning indicator to understand the overall evaluation level of the farmers.

[0104] II) Calculate the median: The median can reflect the middle level of farmers’ scores. It has less impact on outliers and can better reflect the farmers’ real scoring habits.

[0105] III) Calculate variance: Variance and standard deviation can measure the volatility of farmers’ scores and understand the stability and consistency of farmers in evaluating early warning indicators.

[0106] IV) Other statistics: Other statistics such as mode, maximum, minimum, etc. can also be calculated to obtain more comprehensive information on rating habits.

[0107] Cluster analysis: Use clustering algorithms, such as K-means, hierarchical clustering, etc. to classify farmers into different categories based on their scoring habits. Cluster analysis can help identify groups of farmers with similar scoring behaviors.

[0108] Cluster result interpretation: Analyze the characteristics of each cluster group, such as one group may be more sensitive to certain early warning indicators, while another group may pay more attention to other indicators. This information helps to understand the risk preferences and early warning needs of different farmer groups.

[0109] Classification application: Based on the clustering results, farmers are classified according to their scoring habits. In this way, the early warning system can provide customized early warning services for farmers of different categories, such as:

[0110] I) For farmers whose scores are more volatile, the system can provide more frequent early warning notifications.

[0111] II) For farmers who are particularly sensitive to specific indicators, the system can focus on monitoring the changes in these indicators and provide more detailed early warning analysis.

[0112] Through this statistical analysis method, the scoring habits of farmers have been effectively explored and utilized, providing data support for the personalized services of the early warning system, thereby improving the accuracy and practicality of the early warning.

[0113] Adjust the scoring matrix of farmers and early warning indicators based on scoring trends and scoring habits.

[0114] Specifically, the farmer's score for a specific early warning indicator is adjusted based on the scoring trend. If the score shows an upward trend, the weight of the indicator is increased; otherwise, the weight is reduced; different adjustment coefficients are applied to different categories of farmers based on scoring habits. For example, for farmers who are particularly sensitive to a certain indicator, the weight of the indicator is increased.

[0115] 3) Based on the preset indicator weights, the comprehensive similarity between the target farmer and the historical farmers is calculated, and a preset number of similar farmers are selected for the target farmer according to the ranking results of the comprehensive similarity;

[0116] Specifically, the calculation formula of the comprehensive similarity is:

[0117] S composite (u,v)=ω 1 ·S scale (u,v)+ω 2 ·S location (u,v)+ω 3 ·S type (u,v)+ω 4 ·S breed (u,v)

[0118] In the formula, S composite (u,v) represents the comprehensive similarity between farmers u and v; S scale (u,v) represents the similarity of the breeding scale between farmers u and v; S location (u,v) represents the similarity of the breeding areas between breeders u and v; S type (u,v) represents the similarity of the farming entity types between farmers u and v; S breed (u,v) represents the similarity of laying hen breeds between farmers u and v; ω 1 ,ω 2 ,ω 3 ,ω 4 They respectively represent the weights of breeding scale, breeding area, breeding entity type and laying hen breed.

[0119] Specifically, the similarity of each dimension can be calculated as follows:

[0120] Breeding scale similarity: calculated by comparing the breeding scales of breeders (such as measured by the number of breeding), and the normalized distance can be used to represent the similarity.

[0121] Among them, the similarity of breeding scale = 1-(|scale of breeder A-scale of breeder B|) / max(scale of breeder A, scale of breeder B);

[0122] Farming area similarity: Based on the geographical location information of the farmers, the similarity is calculated using geographical distance, climate conditions and soil type.

[0123] Among them, the similarity of breeding areas = a 1 ×[1-(geographic distance / maximum geographical distance)]+a 2 ×(number of common climate features / total number of climate features)+a 3 × (number of common soil type characteristics / number of total soil type characteristics), where a 1 、a 2 、a 3 The weight of each factor respectively.

[0124] Similarity of breeding entity type: calculated based on the type of breeder (such as enterprise, cooperative, individual business, etc.). Category matching can be used. If the match is successful, the similarity is 1, otherwise it is 0 (or other values ​​between 0 and 1).

[0125] Similarity of laying hen breeds: The similarity is calculated based on the laying hen breeds farmed, and can also be measured using category matching.

[0126] 4) According to the geographical location characteristics between the target farmers and similar farmers, the scores of similar farmers on the early warning indicators are adjusted, and the influence of similar farmers on the target farmers decreases with the increase of distance;

[0127] Specifically, according to the geographical location characteristics between the target farmers and similar farmers, the scores of similar farmers on early warning indicators are adjusted as follows:

[0128] Obtaining geographic location data (including longitude and latitude information) of the target farmer and similar farmers, and calculating the geographic distance between the target farmer and each similar farmer, for example, using an appropriate geographic distance calculation formula, such as the Haversine formula, to calculate the distance between two points;

[0129] The attenuation function is used to calculate the attenuation factor of the impact of each similar farmer on the score of the target farmer, and the original score of each similar farmer is multiplied by the corresponding attenuation factor to obtain the adjusted early warning indicator score;

[0130] The expression of the attenuation factor is:

[0131]

[0132] In the formula, γ represents the attenuation factor; d represents the distance between the target farmer and similar farmers; σ represents the standard deviation; and e represents the base of the natural logarithm.

[0133] 5) Based on the comparison results of the adjusted warning indicator scores and the preset score thresholds, a recommended list of warning indicators is generated for the target breeders, and the initial warning thresholds of the warning indicators are set for the target breeders according to the recommended list.

[0134] In this embodiment, in order to generate a recommended list of early warning indicators for the target farmers based on the comparison results of the adjusted early warning indicator scores with the preset score thresholds, and set the initial early warning thresholds, the following steps can be followed:

[0135] Preset scoring thresholds: A series of scoring thresholds are preset based on industry standards and historical data analysis.

[0136] Compare scores with thresholds: Compare the adjusted warning indicator scores with the preset score thresholds.

[0137] Generate a recommendation list: Select warning indicators with scores higher than the preset score threshold and generate a recommendation list of warning indicators.

[0138] Set the initial warning threshold: obtain the warning threshold of each warning indicator in the recommended list from the historical database. For each warning indicator of the target farmer, select the warning threshold of the warning indicator with the highest score in the recommended list as the initial warning threshold.

[0139] Dynamic adjustment and optimization: Establish a dynamic adjustment mechanism to re-evaluate the scores and thresholds of early warning indicators regularly (such as monthly or quarterly). Adjust the weights and thresholds of early warning indicators based on new data and market conditions.

[0140] The optimal warning threshold determination module is used to adjust the initial warning threshold according to the historical performance and risk tolerance of the target farmers to obtain the optimal warning threshold for each warning indicator;

[0141] Specifically, the optimal warning threshold determination module adjusts the initial warning threshold according to the historical performance and risk tolerance of the target farmers to obtain the optimal warning threshold of each warning indicator, including:

[0142] 1) Determine whether the target farmer is breeding for the first time. If so, use the initial warning threshold as the optimal warning threshold. If not, calculate the statistical indicators such as the mean, standard deviation and coefficient of variation of each warning indicator in the historical breeding data of the target farmer;

[0143] In this embodiment, when setting the warning threshold, it is very important to distinguish whether the target farmer is a first-time farmer. Farmers who are raising livestock for the first time lack historical data, so different methods need to be used to determine the warning threshold. The following are the steps to determine whether the target farmer is a first-time farmer and calculate the statistical indicators accordingly:

[0144] Determine whether it is the first time to breed: query the historical breeding records of the target breeder. If there is no historical breeding data, or the start time of the breeding record is close to the current time, it can be determined that the breeder is breeding for the first time.

[0145] Treatment of first-time breeding: If the target farmer is breeding for the first time, due to the lack of historical data, the preset initial warning threshold can be used as the optimal warning threshold.

[0146] Treatment of non-first-time breeding: If the target farmer is not a first-time breeder, a more accurate warning threshold can be calculated based on his historical breeding data.

[0147] For each early warning indicator, the following statistical indicators are calculated:

[0148] Mean: Calculate the arithmetic mean of each early warning indicator in the historical breeding data of the target farmers to understand their average level.

[0149] Standard Deviation: Calculate the standard deviation of each early warning indicator to measure the volatility or dispersion in the historical data.

[0150] Coefficient of variation: Calculate the coefficient of variation (standard deviation divided by the mean) for each early warning indicator to understand relative volatility, which is particularly useful for comparisons between farmers of different sizes or levels.

[0151] 2) Based on the calculation results of statistical indicators, analyze the historical performance of target farmers on different early warning indicators, and calculate the historical performance score of target farmers according to the preset scoring criteria;

[0152] Specifically, based on the calculation results of statistical indicators, the historical performance of target farmers on different early warning indicators is analyzed, and the historical performance scores of target farmers are calculated according to the preset scoring criteria, including:

[0153] According to the comparison results of the actual value of the statistical indicator with the industry benchmark value, the performance of the target breeders on different early warning indicators is analyzed. If the actual value of the statistical indicator is better than the industry benchmark value, the historical performance of the target breeders on the early warning indicator is advantageous; if the actual value of the statistical indicator is lower than the industry benchmark value, the historical performance of the target breeders on the early warning indicator is disadvantageous; and the historical performance score of the target breeders on each early warning indicator is calculated according to the preset scoring criteria, and the historical performance score of the target breeders is calculated in combination with the preset early warning indicator weights.

[0154] Among them, the scoring criteria of this embodiment are as follows:

[0155] I) If the actual value of the target farmer is equal to or lower than 80% of the benchmark value, the score is 0;

[0156] II) If the actual value of the target farmer is higher than 120% of the benchmark value, the score is 100;

[0157] III) Between 80% and 120%, the score increases linearly;

[0158] Then, the calculation formula for the historical performance score is:

[0159]

[0160] Where P l represents the historical performance score of the target farmer; n represents the number of early warning indicators; θ i represents the weight of the i-th early warning indicator; P i Represents the historical performance score of the target farmer on the i-th early warning indicator.

[0161] 3) Assess the risk tolerance of target farmers and quantify their risk tolerance, where risk tolerance includes financial status, risk management measures and responses to historical risk events;

[0162] Specifically, the target farmers’ risk tolerance is assessed and the risk tolerance is quantitatively scored, including:

[0163] Obtain the target farmers’ financial data, such as financial statements, balance sheets, cash flow statements, risk management measures (including insurance coverage, disease prevention and control measures, emergency plans, etc.), and records of historical risk events and response measures;

[0164] Analyze financial statements, calculate financial ratios (such as current ratio, quick ratio, debt ratio, net profit margin, etc.), and evaluate financial stability and debt repayment ability;

[0165] Assess the effectiveness and comprehensiveness of the target farmers’ risk management measures and analyze their ability to identify, assess and respond to potential risks;

[0166] Analyze the risk events that the target farmers have encountered in the past and the effectiveness of their response measures, and assess the target farmers' ability to recover from risk events;

[0167] Construct a quantitative scoring model, incorporate financial status, risk management measures and historical risk event responses into the quantitative scoring model, and set scoring criteria and weights for each dimension;

[0168] A quantitative scoring model is used to score each dimension of the target farmers, and the weighted total score is calculated to obtain the risk tolerance score.

[0169] In this embodiment, to build a quantitative scoring model, it is necessary to convert qualitative factors such as financial status, risk management measures, and historical risk event responses into quantitative data for mathematical modeling and analysis. The following are the steps to build such a model:

[0170] Identify scoring dimensions: Identify dimensions that need to be included in the quantitative scoring model, such as financial status, risk management measures, and responses to historical risk events.

[0171] Develop scoring criteria and weights: Develop scoring criteria and weights for each dimension. The scoring criteria should be quantitative and can be derived based on industry standards and historical data analysis. The weight setting should reflect the impact of each dimension on the overall risk of the farmer.

[0172] Data collection and preprocessing: Collect relevant data of target farmers, including financial statements, risk management records and historical risk event archives. Preprocess the data, including cleaning, standardization and conversion, to make it suitable for quantitative analysis.

[0173] Build a scoring model: Build a scoring model using an appropriate statistical or machine learning algorithm; the following is a simplified example:

[0174] import numpy as np

[0175] from sklearn.linear_model import LinearRegression

[0176] #Assume that data is a DataFrame containing all farmers' data, and columns contains all dimensions

[0177] #For example: columns = ['Financial status', 'Risk management measures', 'Historical risk event response', 'Risk score']

[0178] #Separate features and target variables

[0179] X=data[['Financial status', 'Risk management measures', 'Historical risk event response']]

[0180] y = data['risk score'] #Assume that there is already a risk score as the target variable

[0181] #Build a linear regression model

[0182] model=LinearRegression()

[0183] model.fit(X,y)

[0184] #Get the coefficient, that is, the weight

[0185] coefficients = model.coef_

[0186] Set scoring criteria and weights: Set scoring criteria and weights for each dimension based on expert opinions, industry standards, and historical data analysis. For example:

[0187] Financial status (weight 0.4): The scoring criteria can be based on financial indicators such as current ratio and debt ratio.

[0188] Risk management measures (weight 0.3): The scoring criteria can be based on whether the farmers have adopted insurance, diversification strategies, etc.

[0189] Historical response to risk events (weight 0.3): The scoring criteria can be based on the success rate of farmers in responding to risk events in the past.

[0190] When building a quantitative scoring model, each dimension needs to set scoring criteria and weights in order to quantitatively evaluate the performance of the farmer. The scoring criteria define how to give scores based on the specific situation of the farmer, while the weights reflect the importance of each dimension in the overall assessment. For example, if a farmer's financial situation is very solid, but there are deficiencies in risk management, then in the quantitative scoring model, the financial situation will be given a higher weight, while the risk management measures will be given a lower weight. In this way, the overall risk tolerance of the farmer can be more accurately assessed.

[0191] 4) Based on the historical performance and risk tolerance assessment results of the target farmers, the personalized adjustment factor is calculated, and the initial warning threshold is adjusted according to the personalized adjustment factor to obtain the optimal warning threshold for each warning indicator.

[0192] Specifically, the expression of the optimal warning threshold is:

[0193] C′=C·λ

[0194] λ=α·P l +(1-α)·P f

[0195] In the formula, C' represents the optimal warning threshold; C represents the initial warning threshold; λ represents the personalized adjustment factor; α represents the scoring weight coefficient; P l , P f They represent the historical performance score and risk tolerance score of the target farmers respectively.

[0196] The early warning indicator data collection module 3 is used to obtain the real-time early warning indicator data of the egg ex-factory price change rate, the egg-feed ratio equilibrium point and the net profit per kilogram of laying hens, including:

[0197] Egg ex-factory price change rate: Through market monitoring and data analysis, collect and track fluctuations in egg ex-factory prices.

[0198] Egg-feed ratio balance point: Monitor the ratio of feed cost to egg selling price to determine the profit balance point.

[0199] Net profit per kilogram of laying hens: Calculate and track the net profit per kilogram of eggs to reflect the economic benefits of breeding.

[0200] The early warning indicator data analysis module 4 is used to generate a corresponding early warning signal according to the comparison and analysis results of the real-time early warning indicator data and the corresponding optimal early warning threshold, and promptly notify the farmers;

[0201] Specifically, generating corresponding warning signals based on the comparison and analysis results of real-time warning indicator data and the corresponding optimal warning threshold, and notifying farmers in time, is a key link in the laying hen farming early warning analysis system. The following are the steps to achieve this function:

[0202] Data preprocessing: Clean and standardize real-time data to ensure data quality and consistency. Process missing values ​​and outliers to avoid interference with early warning analysis.

[0203] Comparative analysis: Compare the real-time warning indicator data with the preset optimal warning threshold. Determine whether the data of each indicator exceeds, falls below, or approaches the threshold to determine whether there is a potential risk.

[0204] Generate early warning signals: Generate corresponding early warning signals based on the results of the comparison analysis. For example, if a certain indicator exceeds the threshold, a "high risk" early warning signal can be generated. The generation of early warning signals can be based on rule engines or machine learning models to achieve automated and intelligent early warning.

[0205] Notify farmers: Once an early warning signal is generated, the system should immediately notify farmers through appropriate channels. These channels may include email, SMS, mobile app push or phone calls. The notification content should be clear and concise, including the type of early warning signal, data on relevant indicators and possible risk impacts.

[0206] Recording and reporting: The system should record all generated warning signals and notification activities for subsequent analysis and audit. Generate reports regularly to summarize the trends and effects of warning activities for reference by farmers and managers.

[0207] Continuous optimization: Based on farmers’ feedback and the actual effects of early warning signals, we continuously optimize early warning thresholds and notification mechanisms. We regularly update the optimal early warning thresholds to adapt to market changes and farmers’ actual conditions.

[0208] Through these steps, the laying hen farming early warning analysis system can timely and effectively transmit potential risk information to farmers, help farmers take appropriate measures to reduce risks, and improve the stability and economic benefits of the farming industry.

[0209] Personalized decision-making suggestion module 5, used to provide personalized decision-making suggestions for farmers according to early warning signals;

[0210] Specifically, providing farmers with personalized decision-making suggestions based on early warning signals is an important part of the early warning analysis system for laying hen farming. The following are the steps to achieve this function:

[0211] Early warning signal analysis: Conduct in-depth analysis of the generated early warning signals to understand the reasons and potential risks behind them. Distinguish different types of early warning signals, such as market risk, epidemic risk, financial risk, etc.

[0212] Farmer characteristics analysis: Analyze the specific characteristics of the target farmers, such as farming scale, geographical location, business history, financial status, etc. Understand the risk tolerance and preferences of farmers.

[0213] Decision suggestion generation: Generate personalized decision suggestions based on the warning signal type and farmer characteristics. For example, if it is a market risk warning, farmers are advised to consider adjusting sales strategies or inventory management; if it is an epidemic risk warning, farmers are advised to strengthen health management and vaccination; if it is a financial risk warning, farmers are advised to optimize cost structure or seek financial advice.

[0214] Prioritization and feasibility of recommendations: Prioritize the recommendations for decision making to ensure that farmers focus on the most urgent and important recommendations first. Consider the feasibility and difficulty of implementation of the recommendations to ensure that the recommendations are practical for farmers.

[0215] Presentation of recommendations: Present decision recommendations in a user-friendly way, such as through a mobile app, email or web interface. Use charts, graphs and clear text to help farmers understand the recommendations and the logic behind them.

[0216] Dynamic adjustment and optimization: Establish a dynamic adjustment mechanism to re-evaluate the characteristics of farmers and market conditions on a regular basis (e.g. monthly or quarterly). Adjust the content and priority of decision-making recommendations based on new data and market conditions.

[0217] User feedback and improvement: Collect farmers’ feedback on decision-making suggestions to understand their satisfaction and implementation effects. Based on the feedback information, continuously improve the suggestion generation algorithm and presentation method.

[0218] Through these steps, the laying hen farming early warning analysis system can provide farmers with targeted decision-making support, helping them better respond to market changes and potential risks, thereby improving the overall competitiveness and sustainability of the farming industry.

[0219] The early warning analysis result display module 6 is used to provide an intuitive user interface, display real-time data, early warning information and decision-making suggestions, and regularly generate aquaculture benefit analysis reports and early warning statistical reports.

[0220] Specifically, in order to provide an intuitive user interface, display real-time data, warning information and decision-making suggestions, and regularly generate aquaculture benefit analysis reports and warning statistical reports, the following steps can be followed:

[0221] User interface design: Design an intuitive and easy-to-use user interface to ensure that farmers can easily access real-time data, early warning information and decision-making suggestions. The interface includes a clear navigation bar to facilitate users to quickly switch to different functional modules.

[0222] Real-time data display: Display key early warning indicator data in real time on the user interface, such as market prices, feed costs, disease incidence, etc. Use charts, dashboards and real-time data streams to intuitively display data to help farmers quickly understand the current situation.

[0223] Warning information display: When the system detects a potential risk, the warning information is displayed through the user interface, including the type of warning signal, data of related indicators and possible risk impact. Use color coding, icons and alarm sounds to enhance the readability and perceptibility of warning information.

[0224] Decision suggestion display: Display personalized decision suggestions on the user interface, including the type of suggestion, priority, and implementation steps. Provide interactive tools such as sliders, drop-down menus, and checkboxes to enable farmers to adjust the suggestions according to their own circumstances.

[0225] Farming benefit analysis report: Generate farming benefit analysis reports regularly (such as monthly or quarterly), including analysis of key financial indicators such as revenue, cost, profit, etc. Use charts and graphs to display time series analysis, comparative analysis and trend forecasting.

[0226] Early warning statistical reports: Generate early warning statistical reports regularly, including the distribution, frequency and effect analysis of early warning signals. Use tables, pie charts and bar charts to display statistical data to help farmers understand the performance of the early warning system and room for improvement.

[0227] User feedback and improvement: Provide feedback mechanisms on the user interface, such as online surveys, suggestion boxes, and customer support, to collect feedback and improvement suggestions from farmers. Based on user feedback, continuously optimize the user interface and report content to improve user experience and satisfaction.

[0228] These steps can ensure that farmers can effectively obtain and manage real-time data, early warning information and decision-making recommendations, while regularly obtaining detailed reports on farming benefits and early warning effects, thereby improving farmers' risk management capabilities and decision-making quality.

[0229] In summary, with the help of the above-mentioned technical scheme of the present invention, the present invention can not only intelligently construct an early warning indicator system suitable for each farmer according to factors such as the farmer's farming scale, farming area, farming entity type and laying hen variety, but also can personalize the early warning threshold according to the farmer's historical performance and risk tolerance, thereby effectively improving the accuracy and adaptability of the early warning, helping farmers to better cope with market fluctuations and potential risks, thereby improving the overall risk resistance and economic benefits of the farming industry.

[0230] In addition, the present invention first constructs a scoring matrix of farmers and early warning indicators by analyzing the historical breeding data of farmers, and analyzes the scoring trend of farmers on early warning indicators, so that the setting of early warning thresholds is more scientific and reliable. Secondly, similar farmers are selected for target farmers by calculating the comprehensive similarity between target farmers and historical farmers, so as to improve the accuracy and applicability of early warning thresholds. At the same time, the scoring of similar farmers is adjusted by considering the geographical location characteristics between target farmers and similar farmers, so that the setting of early warning thresholds is closer to the actual situation of farmers. In addition, the initial early warning threshold can also be personalized according to the historical performance and risk tolerance of target farmers, and the optimal early warning threshold of each early warning indicator is obtained, so as to improve the flexibility and adaptability of early warning, so that the setting of early warning thresholds is more comprehensive, accurate, objective and quantitative, so that the threshold of early warning indicators can be set intelligently, personalized, dynamically and comprehensively, and then the risk management ability of farmers can be effectively improved, and the market adaptability and economic benefits of farmers can be enhanced.

[0231] In addition, the present invention can not only obtain the data of key early warning indicators in real time and generate timely early warning signals through the data analysis module to help farmers respond quickly to market changes, but also provide personalized decision-making suggestions based on the early warning signals to enhance the market competitiveness of farmers.

[0232] In addition, the present invention enables farmers to easily view real-time data, early warning information and decision-making suggestions through an intuitive user interface. At the same time, regular breeding benefit analysis reports and early warning statistical reports provide farmers with detailed data support, thereby helping farmers to better cope with market changes and potential risks, and improve breeding benefits and risk resistance.

[0233] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A laying hen breeding early warning analysis system based on an early warning indicator system, characterized in that: It includes an early warning indicator system construction module (1), an early warning indicator threshold setting module (2), an early warning indicator data collection module (3), an early warning indicator data analysis module (4), a personalized decision-making suggestion module (5) and an early warning analysis result display module (6); The early warning indicator system construction module (1) is used to construct early warning indicators for early warning analysis of laying hen breeding, wherein the early warning indicators include the rate of change of ex-factory egg prices, the equilibrium point of the egg-feed ratio and the net profit per kilogram of laying hens; The warning indicator threshold setting module (2) includes an initial warning threshold setting module and an optimal warning threshold determination module; The initial warning threshold setting module is used to use the indicator analysis algorithm to provide the target farmers with the initial warning threshold of the corresponding warning indicators according to the farmers' breeding scale, breeding area, breeding entity type and breeding egg-laying chicken breeds; The optimal warning threshold determination module is used to adjust the initial warning threshold in a personalized manner according to the historical performance and risk tolerance of the target farmers to obtain the optimal warning threshold for each warning indicator; The early warning indicator data collection module (3) is used to obtain real-time early warning indicator data of the rate of change of the ex-factory price of eggs, the equilibrium point of the egg-to-feed ratio and the net profit per kilogram of laying hens; The early warning indicator data analysis module (4) is used to generate a corresponding early warning signal based on the comparison and analysis results of the real-time early warning indicator data and the corresponding optimal early warning threshold, and to notify the breeder in a timely manner; The personalized decision suggestion module (5) is used to provide personalized decision suggestions for farmers according to the early warning signal; The early warning analysis result display module (6) is used to provide an intuitive user interface, display real-time data, early warning information and decision-making suggestions, and regularly generate aquaculture benefit analysis reports and early warning statistical reports.

2. The laying hen breeding early warning analysis system based on the early warning indicator system according to claim 1, characterized in that: The initial warning threshold setting module includes the following steps when providing the target farmers with the initial warning threshold of the corresponding warning indicators according to the farmers' breeding scale, breeding area, breeding subject type and breeding egg-laying chicken breeds by using the indicator analysis algorithm: Based on the database, the historical breeding data of farmers are obtained and the data is cleaned and preprocessed. The historical breeding data includes breeding scale, breeding area, breeding entity type, breeding egg-laying chicken breeds and corresponding early warning indicator data; Construct a scoring matrix for farmers and early warning indicators based on the farmers’ breeding history and results; analyze the scoring trend of farmers on early warning indicators, identify farmers’ scoring habits, and adjust the scoring matrix for farmers and early warning indicators based on the scoring trend; Based on the preset indicator weights, the comprehensive similarity between the target farmer and the historical farmers is calculated, and a preset number of similar farmers are selected for the target farmer according to the ranking results of the comprehensive similarity; According to the geographical location characteristics between the target farmers and similar farmers, the scores of similar farmers on the early warning indicators are adjusted, and the influence of similar farmers on the target farmers decreases with the increase of distance; Based on the comparison results of the adjusted warning indicator scores and the preset score thresholds, a recommended list of warning indicators is generated for the target breeders, and the initial warning thresholds of the warning indicators are set for the target breeders according to the recommended list.

3. The laying hen breeding early warning analysis system based on the early warning indicator system according to claim 2 is characterized in that: The analysis of the scoring trend of the farmers on the early warning indicators, identifying the scoring habits of the farmers, and adjusting the scoring matrix of the farmers and the early warning indicators according to the scoring trend includes: Obtain the historical scoring data of the early warning indicators by the farmers and perform cleaning and standardization processing, where the scoring data includes timestamp, farmer ID, early warning indicator name and corresponding scoring value; Based on time series analysis technology, analyze the changing trend of farmers' scores on early warning indicators over time, and use visualization tools to intuitively display the score trends; Statistically analyze the scoring habits of farmers, and classify farmers according to their scoring habits based on the clustering results; Adjust the scoring matrix of farmers and early warning indicators based on scoring trends and scoring habits.

4. The laying hen breeding early warning analysis system based on the early warning indicator system according to claim 2, characterized in that: The calculation formula of the comprehensive similarity is: S composite (u,v)=ω1·S scale (u,v)+ω2·S location (u,v)+ω3·S type (u,v)+ω4·S breed (u,v) In the formula, S composite (u,v) represents the comprehensive similarity between farmers u and v; S scale (u,v) represents the similarity of farming scale between farmers u and v; S location (u,v) represents the similarity of the breeding areas between breeders u and v; S type (u,v) represents the similarity of the farming entity types between farmers u and v; S breed (u,v) represents the similarity of laying hen breeds between farmers u and v; ω1, ω2, ω3, and ω4 represent the weights of breeding scale, breeding area, breeding entity type, and laying hen breed, respectively.

5. The laying hen breeding early warning analysis system based on the early warning indicator system according to claim 2, characterized in that: The adjustment of the scores of similar farmers on early warning indicators based on the geographical location characteristics between the target farmers and similar farmers includes: Obtain the geographic location data of the target farmer and similar farmers, and calculate the geographic distance between the target farmer and each similar farmer; The attenuation function is used to calculate the attenuation factor of the impact of each similar farmer on the score of the target farmer, and the original score of each similar farmer is multiplied by the corresponding attenuation factor to obtain the adjusted early warning indicator score; The expression of the attenuation factor is: In the formula, γ represents the attenuation factor; d represents the distance between the target farmer and similar farmers; σ represents the standard deviation; e represents the base of natural logarithms.

6. The laying hen breeding early warning analysis system based on the early warning indicator system according to claim 1, characterized in that: The optimal warning threshold determination module adjusts the initial warning threshold individually according to the historical performance and risk tolerance of the target farmers to obtain the optimal warning threshold of each warning indicator, including: Determine whether the target farmer is breeding for the first time. If so, use the initial warning threshold as the optimal warning threshold. If not, calculate the statistical indicators of the mean, standard deviation and coefficient of variation of each warning indicator in the historical breeding data of the target farmer respectively; Based on the calculation results of statistical indicators, analyze the historical performance of target farmers on different early warning indicators, and calculate the historical performance score of target farmers according to the preset scoring criteria; Assess the risk tolerance of target farmers and quantify their risk tolerance, which includes financial status, risk management measures and responses to historical risk events; Based on the historical performance and risk tolerance assessment results of the target farmers, the personalized adjustment factor is calculated, and the initial warning threshold is adjusted according to the personalized adjustment factor to obtain the optimal warning threshold for each warning indicator.

7. The laying hen breeding early warning analysis system based on the early warning indicator system according to claim 6, characterized in that: The calculation results based on the statistical indicators, analyzing the historical performance of the target farmers on different early warning indicators, and calculating the historical performance scores of the target farmers according to the preset scoring criteria include: Based on the comparison results of the actual values ​​of the statistical indicators with the industry benchmark values, the performance of the target breeders on different early warning indicators is analyzed, and the historical performance scores of the target breeders on each early warning indicator are calculated according to the preset scoring standards. At the same time, the historical performance scores of the target breeders are calculated in combination with the preset early warning indicator weights.

8. The laying hen breeding early warning analysis system based on the early warning indicator system according to claim 7, characterized in that: The calculation formula for the historical performance score is: Where P l Indicates the historical performance score of the target farmer; n represents the number of early warning indicators; θ i represents the weight of the i-th early warning indicator; P i Represents the historical performance score of the target farmer on the i-th early warning indicator.

9. The laying hen breeding early warning analysis system based on the early warning indicator system according to claim 6, characterized in that: The assessment of the target farmers' risk tolerance and the quantitative scoring of the risk tolerance include: Obtain the target farmers’ financial data, risk management measures, and historical records of risk events and responses; Analyze financial statements, calculate financial ratios, and assess financial stability and solvency; Assess the effectiveness and comprehensiveness of the target farmers’ risk management measures and analyze their ability to identify, assess and respond to potential risks; Analyze the risk events that the target farmers have encountered in the past and the effectiveness of their response measures, and assess the target farmers' ability to recover from risk events; Construct a quantitative scoring model, incorporate financial status, risk management measures and historical risk event responses into the quantitative scoring model, and set scoring criteria and weights for each dimension; A quantitative scoring model is used to score each dimension of the target farmers, and the weighted total score is calculated to obtain the risk tolerance score.

10. The laying hen breeding early warning analysis system based on the early warning indicator system according to claim 6, characterized in that: The expression of the optimal warning threshold is: C′=C·λ λ=α·P l +(1-a)·P f In the formula, C' represents the optimal warning threshold; C represents the initial warning threshold; λ represents the personalized adjustment factor; α represents the scoring weight coefficient; P l , P f They represent the historical performance score and risk tolerance score of the target farmers respectively.

Images