Vegetable industry chain distribution management system

Abstract

The invention relates to the technical field of digital management, and discloses a vegetable industry chain distribution management system, which comprises a data acquisition and block chain evidence storage module used for acquiring multi-source data of a vegetable planting end, a transportation end and a sales end in real time, and constructing a distributed data warehouse through a block chain and a lightweight encryption technology; and the industrial chain digital modeling module is used for constructing a loss prediction fitting model of the vegetables at the transportation end and a waste prediction fitting model of the vegetables at the sales end based on the historical data in the distributed data warehouse, and the loss rate of the vegetable transportation end can be dynamically predicted according to the loss prediction fitting model and the waste prediction fitting model. According to the method, the loss and waste prediction model constructed by using historical data can be combined with multi-dimensional factors such as real-time climate, regions and seasons to dynamically estimate the transportation loss rate and the sales end hidden waste rate, so that all parties of an industrial chain can identify a risk region and a high loss link before a problem occurs; and the effective pre-control on the loss and waste of the vegetables is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of digital management, and in particular to a vegetable industry chain distribution management system. BACKGROUND

[0002] In traditional vegetable industry chain management, there are problems of poor information flow and serious data island. There is a lack of effective data sharing and collaboration mechanism among planting, transportation and sales links, resulting in that supply and demand matching relies on manual experience, and loss and waste are difficult to accurately predict and control. At the same time, due to the lack of full-process credible traceability and dynamic supervision means, the quality and safety risks in the vegetable flow process are difficult to discover and intervene in time, and the overall supply chain efficiency is low and resources are wasted seriously.

[0003] An existing patent discloses an agricultural full industry chain operation management system (CN120996712A). The prior art scheme focuses on industry integration and logistics level setting, lacks fine, data-driven prediction and dynamic optimization capability for loss and waste in the vegetable flow process, and fails to effectively solve the problem of asymmetric information among links. The decision relies on manual experience rather than real-time data model, and lacks automatic early warning and compliance supervision mechanism based on smart contract, resulting in limited supply chain efficiency improvement and insufficient loss control. SUMMARY

[0004] The present application provides a vegetable industry chain distribution management system to solve the existing technical problems, which solves the problem of poor information flow in traditional vegetable industry chain management.

[0005] To solve the above technical problems, according to one aspect of the present application, more specifically, a vegetable industry chain distribution management system, comprising: A data acquisition and blockchain storage module is used to acquire multi-source data of vegetable planting end, transportation end and sales end in real time, and to construct a distributed data warehouse through blockchain and lightweight encryption technology; An industry chain digital modeling module is used to construct a loss prediction fitting model of the vegetable transportation end and a waste prediction fitting model of the sales end based on historical data in the distributed data warehouse, and to dynamically predict the loss rate of the vegetable transportation end and the implicit waste rate of the sales end according to the loss prediction fitting model and the waste prediction fitting model; A supply and demand dynamic analysis module is used to analyze the best supply and demand scheme when transporting and selling vegetables in different regions according to the predicted loss rate and implicit waste rate; A safety and compliance supervision module is used to realize vegetable full-process traceability based on blockchain storage data, and to set different loss rate thresholds and implicit waste rate thresholds according to different regions, and to trigger a smart contract to suspend the circulation of related batches when the real-time predicted loss rate or implicit waste rate exceeds the set threshold; The specific steps for solving the information asymmetry between the planting households, the transportation end and the sales end in the vegetable industry chain by the vegetable industry chain distribution management system are as follows: S1, acquiring multi-source data of the vegetable planting end, the transportation end and the sales end in different regions according to the data acquisition and blockchain storage module; S2, inputting the real-time multi-source data acquired in different regions into a loss prediction fitting model or a waste prediction fitting model, and then outputting the loss rate and the implicit waste rate of the sales end in the region under the current month, climate and regional conditions; S3, giving the best supply and demand scheme according to the loss rate and the implicit waste rate, and determining whether to trigger the smart contract to suspend the circulation of the related batches according to whether the output loss rate and the implicit waste rate exceed the set threshold.

[0006] Further, the data acquisition and blockchain storage module specifically includes: A transportation data acquisition unit for acquiring transportation distance, temperature and humidity, vibration data and cargo state of the transportation vehicle in the transportation area; A sales data acquisition unit for acquiring inventory, sales volume and price data of each sales terminal; A blockchain storage unit for encrypting and storing the collected data, and assigning a unique traceability identifier to each batch of vegetables.

[0007] Further, the specific steps for the industry chain digital modeling module to construct a loss prediction fitting model for predicting the loss rate of the vegetables during transportation are as follows: 1) Based on the existing multi-source data in the historical data, the transportation environment deviation index and the transportation time length influence coefficient under the same month, climate and regional conditions, and the actual loss rate of the vegetables caused by transportation are acquired; 2) The relationship between the transportation environment deviation index, the transportation time length influence coefficient and the actual loss rate of the vegetables is determined in sequence by the control variable method; 3) The loss prediction fitting model is generated by fitting the correlation between the transportation environment deviation index and the transportation time length influence coefficient, and the loss prediction fitting model is only used to predict the loss rate caused by transportation under the same month, climate and regional conditions.

[0008] Further, the loss prediction fitting model refers to a mathematical prediction model constructed based on historical transportation data by analyzing the influence of transportation environment conditions and transportation time length on the loss of the vegetables, which is used to quantitatively predict the expected loss rate of the vegetables during transportation under specific climate, regional and seasonal conditions.

[0009] Further, the transportation environment deviation index is used to measure the deviation degree of temperature and humidity from the ideal conditions during transportation; The transportation duration influence coefficient is used to illustrate the influence of transportation time on loss, and the longer the time, the higher the loss risk.

[0010] Further, the specific steps of the industry chain digital modeling module for constructing a waste prediction fitting model for predicting the hidden waste rate at the sales end are: 1) Based on the existing multi-source data in the historical data, the inventory turnover health degree and the sales prediction accuracy under the same month, climate and regional conditions, and the actual hidden waste rate at the sales end are obtained; 2) The relationship between the inventory turnover health degree, the sales prediction accuracy and the actual vegetable hidden waste rate is determined in sequence by the control variable method; 3) Then, according to the correlation between the inventory turnover health degree and the sales prediction accuracy, a waste prediction fitting model is fitted and generated, which is only used to predict the loss rate caused by transportation under the same month, climate and region.

[0011] Further, the waste prediction fitting model refers to a mathematical prediction model constructed based on historical sales data by analyzing the influence relationship of inventory turnover and sales prediction accuracy on vegetable waste, which is used to quantitatively predict the expected hidden waste rate of vegetables in the sales link under specific climate, region and time conditions.

[0012] Further, the inventory turnover health degree is used to reflect whether the inventory turnover at the sales end is reasonable. The sales prediction accuracy is used to reflect the degree of agreement between sales prediction and actual sales.

[0013] Further, the steps of the supply and demand dynamic analysis module for analyzing the best supply and demand scheme when transporting and selling vegetables in different regions according to the loss rate and the hidden waste rate are: 1) For each sales region to be evaluated, the predicted transportation loss rate to the region is obtained from the industry chain digital modeling module And the predicted hidden waste rate of sales in the region , wherein i represents different regions. 2) According to the obtained predicted transportation loss rate And the predicted hidden waste rate , combined with the preset transportation cost weight ɑ and the waste cost weight β (ɑ+β=1), the comprehensive cost coefficient of each region i is calculated , and the calculation formula is: ; Wherein, the comprehensive cost coefficient is used to quantify the overall expected risk and cost level of supplying vegetables to the region; 3) Compare the comprehensive cost coefficients of all regions to be evaluated , filter out the area with a low comprehensive cost coefficient below a preset threshold as a candidate supply area; and generate one or more supply-demand matching schemes in combination with real-time demand of each area, inventory of the supply base and transportation capacity constraints; 4) output the generated optimal supply-demand scheme to a system interaction interface, and dynamically update the predicted loss rate, the predicted waste rate and the corresponding supply-demand scheme according to newly collected real-time data.

[0014] The vegetable industry chain distribution management system provided by the application has the following effects compared with the prior art: 1) The application realizes real-time on-chain and credible notarization of data in each link of planting, transportation and sales by integrating blockchain technology and multi-source data collection, effectively breaking the information island and data barrier existing in the traditional vegetable industry chain. All participants can make decisions based on the same trusted data source, enhancing information synchronization and business collaboration between upstream and downstream, thereby significantly reducing problems such as supply-demand imbalance and resource mismatch caused by information asymmetry.

[0015] 2) The loss and waste prediction model constructed by using historical data can dynamically estimate the transportation loss rate and the hidden waste rate at the sales end in combination with real-time climate, region, season and other multi-dimensional factors. This enables all parties in the industry chain to identify risk areas and high-loss links before problems occur, and actively adjust transportation routes, inventory strategies or sales plans based on the prediction results, achieving effective pre-control of vegetable loss and waste.

[0016] 3) The application converts the predicted loss rate and waste rate into a quantifiable comprehensive cost coefficient through a supply-demand dynamic analysis module, and generates an optimal supply-demand allocation scheme for different areas in combination with real-time demand, inventory and transportation capacity constraints. This process is data-driven, replacing the traditional experience-based decision-making mode, improving the rationality and response speed of supply chain resource allocation, and helping to reduce costs and improve flow efficiency at the overall level.

[0017] 4) The application sets a preset loss and waste threshold and associates it with an intelligent contract, so that the system can automatically perform supervision actions such as suspending circulation for batches that exceed the safe range, realizing the transition from passive spot checks to active early warnings and from manual supervision to intelligent compliance, and improving the safety and compliance management level of vegetable circulation. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 is a system block diagram of the application; Figure 2 is a relationship diagram of the transportation environment deviation index and the actual vegetable loss rate in the application; Figure 3 is a relationship diagram of the transportation time length influence coefficient and the actual vegetable loss rate in the application; Figure 4 Figure for the relationship between the transport environment deviation index and the transport time length influence coefficient in the present application; Figure 5 Figure for the relationship between the transport environment deviation index and the actual vegetable loss rate in the present application; Figure 6 Figure for the relationship between the transport time length influence coefficient and the actual vegetable loss rate in the present application; Figure 7 Figure for the relationship between the transport environment deviation index and the transport time length influence coefficient in the present application; Figure 8 Workflow diagram of the present application. DETAILED DESCRIPTION

[0019] In order to make the technical scheme of the present application clearer, the present application will be further described in detail below in combination with the drawings and specific embodiments.

[0020] Embodiment 1 As shown in the drawings, according to one aspect of the present application, a vegetable industry chain distribution management system is provided, comprising: Figure 1 a data acquisition and blockchain storage module for acquiring multi-source data of the vegetable planting end, the transportation end and the sales end in real time, and constructing a distributed data warehouse through blockchain and lightweight encryption technology; the data acquisition and blockchain storage module specifically comprises: a transportation data acquisition unit for acquiring transportation distance of the transportation vehicle, temperature and humidity of the transportation area, vibration data and cargo state; a sales data acquisition unit for acquiring inventory, sales volume and price data of each sales terminal; a blockchain storage unit for encrypting and storing the acquired data, and assigning a unique traceability identifier for each batch of vegetables. an industry chain digital modeling module for constructing a loss prediction fitting model of the vegetable at the transportation end and a waste prediction fitting model of the sales end based on historical data in the distributed data warehouse, and dynamically predicting the loss rate of the vegetable at the transportation end and the implicit waste rate of the sales end according to the loss prediction fitting model and the waste prediction fitting model;

[0021] a supply and demand dynamic analysis module for analyzing the best supply and demand scheme when transporting and selling vegetables in different regions according to the predicted loss rate and implicit waste rate; a safety and compliance supervision module for realizing vegetable whole-process traceability based on blockchain storage data, and setting different loss rate thresholds and implicit waste rate thresholds according to different regions, and triggering an intelligent contract to suspend the circulation of related batches when the real-time predicted loss rate or implicit waste rate exceeds the set threshold; Embodiment 2 As shown in the drawings, according to one aspect of the present application, a vegetable industry chain distribution management system is provided, comprising: a data acquisition and blockchain storage module for acquiring multi-source data of the vegetable planting end, the transportation end and the sales end in real time, and constructing a distributed data warehouse through blockchain and lightweight encryption technology; the data acquisition and blockchain storage module specifically comprises: Figure 1As shown, the specific steps of the vegetable industry chain distribution management system for solving the information asymmetry between the planting households, transportation end and sales end in the vegetable industry chain are as follows: S1, acquiring multi-source data of the vegetable planting end, transportation end and sales end in different regions according to the data acquisition and blockchain storage module; S2, inputting the real-time multi-source data acquired in different regions into the loss prediction fitting model, and then outputting the loss rate of the sales end in the region under the current month, climate and regional conditions; Wherein, the specific steps of constructing the loss prediction fitting model are as follows: 1) Based on the existing multi-source data in the historical data, the transportation environment deviation index and transportation time length influence coefficient under the same month, climate and regional conditions, and the actual vegetable loss rate caused by transportation are obtained; In this embodiment, the transportation environment deviation index is used to measure the deviation degree of temperature and humidity from ideal conditions in the transportation process; the specific calculation formula is as follows: ; In the above formula, if , then ; T and H represent the actual temperature and humidity, respectively; , represent the optimal transportation temperature and humidity of the vegetable, respectively; , represent the allowable temperature and humidity fluctuation range, respectively; , represent the weight of temperature and humidity, respectively, and .

[0022] In this embodiment, the transportation time length influence coefficient is used to explain the influence of transportation time on loss, and the longer the time, the higher the loss risk; the specific calculation formula is as follows: ; In the above formula, if , then ; represents the actual transportation time length; represents the maximum safe transportation time length of the vegetable.

[0023] 2) The relationship between the transportation environment deviation index, the transportation time length influence coefficient and the actual vegetable loss rate is determined in turn by the control variable method; I. Determine the relationship between the transportation environment deviation index and the actual vegetable loss rate For example, in the spring 20-25℃, humidity 40-60% climate, will be in broccoli production base A to B city area of supermarket sales. Then, from the 1000 sample data, screening transport time impact coefficient of 0.5 of the sample 100, and the actual vegetable loss rate generated by the transport of the 100 sample data is recorded as loss X.

[0024] Then, the mathematical relationship between the loss prediction fitting model and the loss X is: (Formula 1); In the above formula 1, indicates the predicted loss rate output by the loss prediction fitting model in the process of transporting broccoli from production base A to the supermarket in the B city area in the spring 20-25℃, humidity 40-60% climate; , for controlling tends to be approximately the constant. And by Figure 2 data can determine , , the in formula 1 tends to be approximately the loss X ( indicates the curve in Figure 2 ).

[0025] II. Determine the relationship between the transport time impact coefficient and the actual vegetable loss rate For example, in the spring 20-25℃, humidity 40-60% climate, will be in broccoli production base A to B city area of supermarket sales. Then, from the 1000 sample data, screening transport environment deviation index of 0.5 of the sample 100, and the actual vegetable loss rate generated by the transport of the 100 sample data is recorded as loss Y.

[0026] Then, the mathematical relationship between the loss prediction fitting model and the loss Y is: (Formula 2); In the above formula 1, indicates the predicted loss rate output by the loss prediction fitting model in the process of transporting broccoli from production base A to the supermarket in the B city area in the spring 20-25℃, humidity 40-60% climate; , for controlling tends to be approximately the constant. And by Figure 3 data can determine , , the in formula 2 tends to be approximately the loss Y ( indicates the curve in Figure 3the curve in FIG. 6).

[0027] 3) According to the correlation between the transportation environment deviation index and the transportation time length influence coefficient, a loss prediction fitting model is generated, which is only used to predict the loss rate generated by transportation under the same month, climate and regional conditions.

[0028] III. Determine the correlation between the transportation environment deviation index and the transportation time length influence coefficient For example, 100 continuous sample data are selected from 1000 sample data, including the transportation environment deviation index, the transportation time length influence coefficient, and the actual vegetable loss rate generated by transportation of the 100 sample data, denoted as loss Z.

[0029] Then, according to the relationship between the transportation environment deviation index, the transportation time length influence coefficient and the loss Z, the relationship between the above formula 1 and formula 2 is determined, that is: × (formula 1) × (formula 2) + In the above, , is a constant used to control the approximation of the loss prediction fitting model output value L and the loss Z. And by Figure 4 data in the above , , the loss prediction fitting model output value L and the loss Z tend to be approximate.

[0030] Therefore, from the 1000 samples, the loss prediction fitting model mathematical expression of the loss rate during transportation when the broccoli produced in base A is transported to the supermarket in region B for sale is: In the above formula, L represents the predicted loss rate output by the loss prediction fitting model. When L exceeds the set threshold, the smart contract will be triggered to suspend the circulation of the related batch.

[0031] S3, according to the loss rate, give the best supply and demand scheme, and according to whether the output loss rate exceeds the set threshold to judge whether to trigger the smart contract to suspend the circulation of the related batch.

[0032] Embodiment 3 As shown in Figure 1 , the specific steps of the vegetable industry chain distribution management system to solve the information asymmetry between the growers, transportation end and sales end in the vegetable industry chain are: S1, according to the data acquisition and block chain storage module, obtain multi-source data of vegetable planting end, transportation end and sales end in different regions; ​​S2, input the real-time multi-source data of different areas into the waste prediction fitting model, and then output the implicit waste rate of the sales end in the current month, climate and regional conditions; Wherein, the specific steps of constructing the waste prediction fitting model are: 1) Based on the existing multi-source data in the historical data, the inventory turnover health degree and the sales prediction accuracy under the same month, climate and regional conditions, and the actual implicit waste rate of the sales end are obtained; In this embodiment, the inventory turnover health degree is used to reflect whether the inventory turnover of the sales terminal is reasonable; the specific calculation formula is: ; In the above formula, if s>1, then s=1, indicating no extrusion. Wherein, the closer the value of s is to 1, the better the inventory turnover is, and the lower the waste risk is.

[0033] In this embodiment, the sales prediction accuracy is used to reflect the degree of fit between the sales prediction and the actual sales; the specific calculation formula is: ; In the above formula, if p is negative, then p=0. Wherein, the closer the value of p is to 1, the more accurate the prediction is, and the lower the waste risk is.

[0034] 2) The relationship between the inventory turnover health degree, the sales prediction accuracy and the actual vegetable implicit waste rate is determined by the control variable method; I. Determine the relationship between the transportation environment deviation index and the actual vegetable loss rate For example, in the climate of 20-25℃ and humidity of 40-60% in spring, broccoli produced in production base A is transported to supermarket sales in B city area. Then, 100 samples with a sales prediction accuracy of 0.5 are selected from the 1000 sample data, and the actual implicit waste rate of the sales end of the 100 sample data is recorded as waste x.

[0035] Then, the mathematical relationship between the inventory turnover health degree and waste x is: (Formula 3); In the above formula 1, represents the predicted implicit waste rate output by the waste prediction fitting model in the process of transporting broccoli produced in production base A to supermarket sales in B city area in the climate of 20-25℃ and humidity of 40-60% in spring; , is used to control and waste x tends to be similar. And Figure 5 data in , , the constant in formula 3 is tends to approximate (waste x) represents Figure 5 the curve in (waste x).

[0036] II. Determine the relationship between the transport duration impact coefficient and the actual vegetable loss rate For example, in the spring, 20-25℃, humidity 40-60% climate, broccoli production base A is transported to B city area supermarket sales. Then, from the 1000 sample data, 100 samples with a stock turnover health degree of 0.5 are selected, and the actual implicit waste rate of the sales end of the 100 sample data is recorded as waste y.

[0037] Then, the mathematical relationship between sales forecast accuracy and waste y is: (Formula 4); In the above formula 1, represents the predicted implicit waste rate output by the waste prediction fitting model in the process of transporting broccoli production base A to B city area supermarket sales in the spring, 20-25℃, humidity 40-60% climate; , for controlling tends to approximate (waste y). And from the data in (waste x) Figure 6 , , , formula 4 tends to approximate (waste y) (waste x) represents Figure 6 the curve in (waste x).

[0038] 3) According to the correlation between inventory turnover health degree and sales forecast accuracy, a waste prediction fitting model is generated, which is only used to predict the loss rate generated by transportation in the same month, climate and region.

[0039] III. Determine the correlation between the transport environment deviation index and the transport duration impact coefficient For example, 100 continuous sample data are selected from 1000 sample data, including inventory turnover health degree, sales forecast accuracy, and the actual implicit waste rate of the sales end of the 100 sample data is recorded as waste z.

[0040] Then, according to the relationship between inventory turnover health degree, sales forecast accuracy and waste z, the relationship between the above formula 3 and formula 4 is determined, that is: × (formula 3) × (formula 4) + ; In the above, , A constant for controlling the waste prediction fitting model output value W to approximate the waste z. And by Figure 7 the data in , , the waste prediction fitting model output value W approximates the waste z.

[0041] Therefore, from the 1000 samples that can be obtained from the broccoli production base A to the supermarket sales in the B city area, the waste prediction fitting model mathematical expression for predicting the implicit waste rate is: ; In the above formula, W represents the predicted implicit waste rate of the waste prediction fitting model output, and when W exceeds the set threshold, the smart contract will be triggered to suspend the circulation of the related batch.

[0042] S3, according to the implicit waste rate, the best supply and demand scheme is given, and whether the output implicit waste rate exceeds the set threshold is judged to trigger the smart contract to suspend the circulation of the related batch.

[0043] Embodiment 4 As Figure 1 shown, the specific steps of the vegetable industry chain distribution management system to solve the information asymmetry between the growers, transportation end, and sales end in the vegetable industry chain are: S1, according to the data acquisition and blockchain storage module, obtain the multi-source data of the vegetable planting end, transportation end, and sales end in different regions; S2, input the real-time multi-source data obtained in different regions into the loss prediction fitting model or the waste prediction fitting model, and then output the loss rate and implicit waste rate of the sales end in the current month, climate and regional conditions in the region; S3, according to the loss rate and implicit waste rate, the best supply and demand scheme is given, and whether the output loss rate and implicit waste rate exceed the set threshold is judged to trigger the smart contract to suspend the circulation of the related batch.

[0044] Among them, the steps of the supply and demand dynamic analysis module to analyze the best supply and demand scheme when transporting and selling vegetables in different regions according to the loss rate and implicit waste rate are: 1), for each sales region to be evaluated, obtain the predicted transportation loss rate to the region from the industry chain digital modeling module and the predicted implicit waste rate of the sales in the region , where i represents different regions; 2), according to the obtained predicted transportation loss rate and the predicted implicit waste rate , combined with the preset transportation cost weight ɑ and waste cost weight β (where ɑ+β=1), calculate the comprehensive cost coefficient of each region i , the calculation formula is: ; Among them, the comprehensive cost coefficient is used to quantify the overall expected risk and cost level of supplying vegetables to the region; 3) Compare the comprehensive cost coefficients of all regions to be evaluated , and select the regions with a comprehensive cost coefficient below the preset threshold as candidate supply regions; and combine the real-time demand of each region, the inventory of the supply base and the transportation capacity constraints to generate one or more supply-demand matching schemes; 4) Output the generated optimal supply-demand scheme to the system interaction interface, and dynamically update the predicted loss rate, predicted waste rate and corresponding supply-demand scheme according to the real-time collected new data.

[0045] Then, under the climate conditions of spring 20-25℃, humidity 40-60%, the dynamic analysis results of transporting broccoli from production base A to different regions for sale. Assuming that the transportation cost weight α=0.6, the waste cost weight β=0.4, and the comprehensive cost coefficient calculation formula is: ; Among them, Regions with a comprehensive cost coefficient below 0.15 are considered low-risk supply candidate regions; then there are: Table 1 Relationship between predicted loss rate, predicted implicit waste rate and comprehensive cost coefficient of some vegetables ; No. 1 Both loss and waste are low, and the comprehensive cost coefficient is the lowest, so the system preferentially recommends allocating resources to this region. No. 4 has high loss and waste, and the comprehensive cost coefficient exceeds 0.15, so the system will prompt the risk and may trigger a threshold alarm. No. 5 has low loss rate but high waste rate, and the comprehensive cost coefficient is still controllable, which belongs to a medium-priority region.

[0046] The above-described embodiments only express several embodiments of the present application, and the description is more specific and detailed, but it cannot be understood as limiting the scope of the present patent. It should be noted that for ordinary skilled persons in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of protection of the present application. Therefore, the protection scope of the present patent should be subject to the appended claims.

Claims

1. A vegetable industry chain distribution management system characterized by comprising: The application relates to a vegetable industry chain distribution management system. The system comprises: a data acquisition and blockchain storage module for acquiring multi-source data of a vegetable planting end, a transportation end and a sales end in real time, and constructing a distributed data warehouse through a blockchain and a lightweight encryption technology; an industry chain digital modeling module for constructing a loss prediction fitting model of the vegetable transportation end and a waste prediction fitting model of the sales end based on historical data in the distributed data warehouse, and dynamically predicting a loss rate of the vegetable transportation end and an implicit waste rate of the sales end regionally according to the loss prediction fitting model and the waste prediction fitting model; a supply and demand dynamic analysis module for analyzing a best supply and demand scheme when the vegetable is transported and sold in different regions according to the predicted loss rate and the implicit waste rate; a safety and compliance supervision module for realizing vegetable whole-process traceability based on the blockchain storage data, and setting different loss rate thresholds and implicit waste rate thresholds according to different regions, and triggering an intelligent contract to suspend the circulation of a related batch when the real-time predicted loss rate or the implicit waste rate exceeds the set threshold. The specific steps of the vegetable industry chain distribution management system for solving the information asymmetry among a grower, a transportation end and a sales end in a vegetable industry chain are as follows: S1, acquiring multi-source data of a vegetable planting end, a transportation end and a sales end in different regions according to the data acquisition and blockchain storage module; S2, inputting the acquired real-time multi-source data of different regions into the loss prediction fitting model or the waste prediction fitting model, and then outputting the loss rate and the implicit waste rate of the sales end in the current month, climate and regional conditions; 2. The vegetable industry chain distribution management system according to claim 1, characterized in that: S3, giving a best supply and demand scheme according to the loss rate and the implicit waste rate, and judging whether an intelligent contract is triggered to suspend the circulation of a related batch according to whether the output loss rate and the implicit waste rate exceed the set threshold. The data acquisition and blockchain storage module specifically comprises: a transportation data acquisition unit for acquiring transportation distance of a transportation vehicle, temperature and humidity of a transportation region, vibration data and a cargo state; a sales data acquisition unit for acquiring inventory, sales volume and price data of each sales terminal; 3.The vegetable industry chain distribution management system according to claim 1, characterized in that: a blockchain storage unit for encrypting and storing the acquired data, and allocating a unique traceability identifier for each batch of vegetables. The specific steps of the industry chain digital modeling module for constructing a loss prediction fitting model for predicting a vegetable transportation loss rate are as follows: 1) acquiring a transportation environment deviation index and a transportation time length influence coefficient under the same month, climate and regional conditions, and an actual vegetable loss rate generated by transportation based on the existing multi-source data in the historical data; 2) sequentially determining the relationship among the transportation environment deviation index, the transportation time length influence coefficient and the actual vegetable loss rate through a control variable method; 3) fitting and generating the loss prediction fitting model according to the correlation between the transportation environment deviation index and the transportation time length influence coefficient, and the loss prediction fitting model is only used for predicting the loss rate generated by transportation under the same month, climate and regional conditions. 4.The vegetable industry chain distribution management system according to claim 3, characterized in that: The loss prediction fitting model refers to a mathematical prediction model constructed based on historical transportation data by analyzing the influence relationship between transportation environmental conditions and transportation time length on the loss of vegetables, for quantitatively predicting the expected loss rate of vegetables in the transportation process under specific climate, region and time conditions. 5.The vegetable industry chain distribution management system according to claim 3, characterized in that: The transportation environmental deviation index is used to measure the deviation degree of temperature and humidity from ideal conditions in the transportation process. The transportation time length influence coefficient is used to explain the influence of transportation time on loss, and the longer the time, the higher the loss risk. 6.The vegetable industry chain distribution management system according to claim 1, characterized in that: The specific steps of the waste prediction fitting model constructed by the industry chain digital modeling module for predicting the implicit waste rate of the sales end are as follows: 1) Based on the existing multi-source data in the historical data, the inventory turnover health degree and sales prediction accuracy under the same month, climate and regional conditions, and the actual implicit waste rate of the sales end are obtained. 2) The relationship between the inventory turnover health degree, the sales prediction accuracy and the actual vegetable implicit waste rate is determined in turn by the control variable method. 3) Then, the waste prediction fitting model is generated according to the correlation between the inventory turnover health degree and the sales prediction accuracy, which is only used to predict the loss rate generated by transportation under the same month, climate and region. 7.The vegetable industry chain distribution management system according to claim 6, characterized in that: The waste prediction fitting model refers to a mathematical prediction model constructed based on historical sales data by analyzing the influence relationship between inventory turnover and sales prediction accuracy on the loss of vegetables, for quantitatively predicting the expected implicit waste rate of vegetables in the sales link under specific climate, region and time conditions. 8.The vegetable industry chain distribution management system according to claim 6, characterized in that: The inventory turnover health degree is used to reflect whether the inventory turnover of the sales terminal is reasonable. The sales prediction accuracy is used to reflect the degree of agreement between sales prediction and actual sales. 9.The vegetable industry chain distribution management system according to claim 1, characterized in that: The steps of the supply and demand dynamic analysis module for analyzing the best supply and demand scheme when transporting and selling vegetables in different regions according to the loss rate and the implicit waste rate are as follows: 1) For each sales region to be evaluated, obtain from the industry chain digital modeling module the predicted transportation loss rate to that region and the predicted hidden waste rate for sales in that region where i represents different regions; 2) According to the obtained predicted transportation loss rate And the predicted implicit waste rate , combined with the preset transportation cost weight a and waste cost weight β, the comprehensive cost coefficient of each region i is calculated The calculation formula is: ； wherein the overall cost coefficient to quantify the overall expected risk and cost level of supplying vegetables to the area; 3) compare the comprehensive cost coefficients of all the regions to be evaluated , and screen out the regions with comprehensive cost coefficients lower than a preset threshold as candidate supply regions; 4) The generated best supply and demand scheme is output to the system interaction interface, and the prediction loss rate, prediction waste rate and the corresponding supply and demand scheme are dynamically updated according to the real-time collected new data.

Images