Agricultural product purchase quantity intelligent prediction method and system based on multi-objective optimization

Abstract

The invention relates to the technical field of agricultural intelligent management, in particular to an agricultural product purchase quantity intelligent prediction method and system based on multi-objective optimization, and the method comprises the steps: receiving an agricultural product purchase instruction, and determining a historical inventory node set in a historical inventory record library based on a target type, obtaining a predicted average loss rate, a predicted average decrement value and a target span duration based on the historical inventory node set, and obtaining a predicted purchase moment based on the current inventory, the predicted transportation duration, the predicted average loss rate and the predicted average decrement value, and obtaining a predicted transportation loss rate based on the historical transportation node set, the M transportation indexes and the predicted purchase time, obtaining a predicted purchase quantity based on the predicted transportation time, the target span time, the predicted transportation loss rate and the current inventory, and realizing intelligent prediction of the agricultural product purchase quantity based on the predicted purchase time and the predicted purchase quantity. According to the invention, accuracy and flexibility of intelligent prediction of the agricultural product purchase quantity can be realized.

Description

Technical Field

[0001] This invention relates to the field of agricultural intelligent management technology, and in particular to an intelligent prediction method and system for agricultural product procurement volume based on multi-objective optimization. Background Technology

[0002] In the agricultural product supply chain, accurate procurement volume forecasting is crucial for ensuring market supply and reducing waste. Scientific forecasting allows for the rational planning of procurement, avoiding over- or under-purchasing, thereby optimizing resource allocation, improving resource utilization efficiency, and reducing agricultural product losses and waste. Therefore, intelligent and accurate forecasting of agricultural products is of great significance for optimizing procurement, transportation, and inventory management, and improving the overall efficiency of the supply chain.

[0003] Currently, traditional methods for forecasting agricultural product procurement volume mainly rely on simple statistical analysis and experience-based judgment. However, these methods struggle to handle complex multi-factor influences, such as transportation losses and dynamic inventory depletion, resulting in low forecast accuracy. Furthermore, traditional methods lack the ability to respond quickly to market changes and cannot adjust procurement plans in a timely manner.

[0004] While traditional methods can predict agricultural product purchase volumes, they are inefficient when processing large-scale data and struggle to fully utilize information from historical data. These drawbacks lead to inaccurate purchase volume forecasts, potentially causing inventory buildup or shortages. Therefore, improving the accuracy and flexibility of agricultural product purchase volume forecasts has become an urgent issue to address. Summary of the Invention

[0005] This invention provides a method for intelligent forecasting of agricultural product procurement volume based on multi-objective optimization and a computer-readable storage medium, the main purpose of which is to improve the accuracy and flexibility of agricultural product procurement volume forecasting.

[0006] To achieve the above objectives, this invention provides an intelligent prediction method for agricultural product procurement volume based on multi-objective optimization, comprising:

[0007] Receive agricultural product procurement instructions, confirm the procurement quantity intelligent prediction system based on the agricultural product procurement instructions, obtain historical inventory record database and historical transportation record database based on the procurement quantity intelligent prediction system, and obtain the target type, wherein the target type is the type of agricultural product to be procured;

[0008] Based on the target type, a set of historical inventory nodes is identified in the historical inventory record database. The set of historical inventory nodes contains multiple historical inventory nodes, and each historical inventory node contains the historical span duration, historical initial inventory quantity, historical loss quantity, and historical reduction value.

[0009] Based on the historical inventory node set, the predicted average loss rate, predicted average reduction value, and target span duration are obtained.

[0010] Obtain the current inventory level and predicted transportation time. Based on the current inventory level, predicted transportation time, predicted average loss rate, and predicted average reduction value, obtain the predicted procurement time.

[0011] Get M transportation indicators, and based on the target type and the M transportation indicators, identify a set of historical transportation nodes in the historical transportation record database. The set of historical transportation nodes contains multiple historical transportation nodes, and each historical transportation node contains a transportation loss rate and M transportation indicator values, where M is an integer greater than 1.

[0012] The predicted transportation loss rate is obtained based on the historical transportation node set, M transportation indicators, and the predicted procurement time.

[0013] The predicted purchase quantity is obtained based on the predicted transportation time, target span duration, predicted transportation loss rate, and current inventory. Based on the predicted purchase time and predicted purchase quantity, intelligent prediction of agricultural product purchase quantity is realized.

[0014] Optionally, obtaining the predicted average loss rate, predicted average reduction value, and target span duration based on the historical inventory node set includes:

[0015] The shelf life of agricultural products is obtained by dividing the shelf life of agricultural products into multiple shelf life intervals based on a preset division method. Each shelf life interval includes an interval start value, an interval end value, and an interval mid value.

[0016] Based on multiple shelf-life time intervals, the historical inventory node set is divided into multiple analysis inventory node sets. The shelf-life time intervals correspond one-to-one with the analysis inventory node sets, and the historical span duration corresponding to each analysis inventory node in the analysis inventory node set is located within the shelf-life time interval.

[0017] Based on the multiple sets of analyzed inventory nodes and multiple shelf-life intervals, the predicted average loss rate, predicted average reduction value, and target span duration are obtained.

[0018] Optionally, obtaining the predicted average loss rate, predicted average reduction value, and target span duration based on the multiple sets of analyzed inventory nodes and multiple shelf-life intervals includes:

[0019] An analysis inventory node set is extracted sequentially from multiple analysis inventory node sets, and the corresponding freshness period interval is determined in multiple freshness period intervals based on the analysis inventory node set.

[0020] The average reduction value is obtained based on the analyzed inventory node set, the corresponding shelf life interval of the analyzed inventory node set, and the pre-constructed reduction value calculation formula, wherein the reduction value calculation formula is as follows:

[0021]

[0022] in, The assessed average reduction value, Indicates the first node in the analysis inventory node set Analyze the historical reduction values ​​corresponding to each inventory node. This indicates that the analysis of the inventory node set has a total of One analysis of inventory nodes, Indicates the first node in the analysis inventory node set The analysis focuses on the absolute time difference between the historical span duration corresponding to each inventory node and the median value of the corresponding shelf-life interval. Indicates the first node in the analysis inventory node set The historical span duration corresponding to each inventory node is analyzed;

[0023] The average reduction values ​​are summarized to obtain the set of average reduction values. The average reduction value with the largest value in the set is the predicted average reduction value. The set of analysis inventory nodes corresponding to the predicted average reduction value is the predicted inventory node set. The midpoint of the shelf life interval corresponding to the predicted average reduction value is the target span duration.

[0024] Predicted inventory nodes are extracted sequentially from the predicted inventory node set, and an initial average loss rate is obtained based on the predicted inventory nodes, wherein the initial average loss rate is the ratio of the historical loss amount to the historical span duration in the predicted inventory nodes.

[0025] The initial average loss rates are summarized to obtain multiple initial average loss rates, and the average of the multiple initial average loss rates is used as the predicted average loss rate.

[0026] Optionally, obtaining the predicted procurement time based on the current inventory level, predicted transportation time, predicted average loss rate, and predicted average reduction value includes:

[0027] The predicted duration is obtained based on the current inventory level, the predicted average loss rate, the predicted average reduction value, and a pre-constructed formula for calculating the predicted duration. The formula for calculating the predicted duration is as follows:

[0028]

[0029] in, The predicted duration is... This indicates the current inventory level. This indicates the preset minimum inventory level. This represents the predicted average loss rate. This represents the predicted average reduction value;

[0030] Obtain the current time, and based on the current time, the predicted duration, and the predicted transportation time, obtain the predicted procurement time.

[0031] Optionally, obtaining the predicted transportation loss rate based on the historical transportation node set, M transportation indicators, and the predicted procurement time includes:

[0032] From the M transportation indicators, transportation indicators are extracted sequentially and then removed. The transportation indicators are taken as the primary indicators, and the M-1 transportation indicators after removal are taken as the M-1 secondary indicators.

[0033] Based on the main indicators, M-1 secondary indicators and historical transportation node set, a main node set and a secondary node set are obtained. Each main node in the main node set includes a transportation loss rate and a main indicator value with an indicator type. Each secondary node in the secondary node set includes M-1 secondary indicator values. The main nodes in the main node set and the secondary nodes in the secondary node set correspond one-to-one.

[0034] A pre-defined clustering algorithm is used to perform clustering operations on the secondary node set to obtain multiple node clusters;

[0035] The number of minor nodes in each of the multiple node clusters is counted separately to obtain multiple statistical quantities;

[0036] Multiple statistical ratios are obtained based on multiple statistical quantities, wherein each statistical ratio corresponds one-to-one with a statistical quantity, and the statistical ratio is the ratio of the statistical quantity corresponding to the statistical ratio to the sum of multiple statistical quantities.

[0037] Determine whether all statistical proportions are the same across multiple statistical proportions;

[0038] If they are the same, a first predicted transportation loss rate acquisition scheme is constructed based on multiple node clusters and the main node set, and a first predicted transportation loss rate is obtained based on the first predicted transportation loss rate acquisition scheme. Otherwise, a second predicted transportation loss rate acquisition scheme is constructed based on multiple node clusters and the main node set, and a second predicted transportation loss rate is obtained based on the second predicted transportation loss rate acquisition scheme. The first predicted transportation loss rate or the second predicted transportation loss rate is used as the single indicator corresponding to the main indicator to predict the transportation loss rate.

[0039] After confirming that the corresponding single-indicator predicted transportation loss rate has been obtained for each major indicator, the single-indicator predicted transportation loss rates are summarized to obtain multiple single-indicator predicted transportation loss rates. The average of the multiple single-indicator predicted transportation loss rates is calculated to obtain the predicted transportation loss rate.

[0040] Optionally, the step of constructing a first predicted transportation loss rate acquisition scheme based on multiple node clusters and a main node set, and acquiring the first predicted transportation loss rate based on the first predicted transportation loss rate acquisition scheme, includes:

[0041] A first predicted transportation loss rate acquisition scheme is constructed based on multiple node clusters, wherein the first predicted transportation loss rate acquisition scheme is as follows:

[0042] Perform the following operation on each of the multiple node clusters:

[0043] Calculate the similarity between secondary nodes in a node cluster to obtain the intra-cluster bias.

[0044] The intra-cluster deviations are summarized to obtain an intra-cluster deviation set. The cluster of nodes corresponding to the smallest intra-cluster deviation in the intra-cluster deviation set is the analysis cluster.

[0045] Based on the analysis cluster, multiple main nodes are identified in the main node set, and the main nodes among the multiple main nodes correspond one-to-one with the secondary nodes among the multiple secondary nodes in the analysis cluster.

[0046] Perform the following operation on each of the multiple primary nodes:

[0047] The transportation loss rate and key indicator values ​​are extracted from the main nodes, and the transportation loss rate and key indicator values ​​are correlated to obtain fitted coordinate points, wherein the fitted coordinate points correspond one-to-one with the main nodes.

[0048] By summing the fitted coordinate points, a set of fitted coordinate points is obtained;

[0049] All the fitted coordinate points in the fitted coordinate point set are mapped to a pre-constructed coordinate system to obtain a mapped coordinate point set. The mapped coordinate point set is then fitted to the first predicted curve using a pre-constructed polynomial fitting model.

[0050] Based on the predicted procurement time, a first predicted indicator value with an indicator type is obtained, wherein the indicator type of the first predicted indicator value is the same as the indicator type of the main indicator value.

[0051] Using the first predicted index value, the retrieval index value is identified in the first predicted curve, and the transportation loss rate corresponding to the retrieval index value is the first predicted transportation loss rate.

[0052] Optionally, the step of constructing a second predicted transportation loss rate acquisition scheme based on multiple node clusters and a main node set, and acquiring the second predicted transportation loss rate based on the second predicted transportation loss rate acquisition scheme, includes:

[0053] A second scheme for obtaining the predicted transportation loss rate is constructed based on multiple node clusters, as shown below:

[0054] Based on a preset statistical ratio threshold, one or more target node clusters are identified from multiple node clusters, wherein the statistical ratio corresponding to the target node cluster is greater than the statistical ratio threshold.

[0055] Perform the following operation on each of one or more target node clusters:

[0056] The second prediction curve is obtained based on the target node cluster and the main node set;

[0057] The second prediction curves are summarized to obtain the second prediction curve set, wherein the second prediction curves correspond one-to-one with the statistical proportions;

[0058] For each second prediction curve in the second prediction curve set, the following operation is performed:

[0059] The second forecast indicator value is obtained based on the predicted procurement time.

[0060] The initial second predicted transportation loss rate is obtained based on the second predicted index value and the second predicted curve, wherein the initial second predicted transportation loss rate corresponds one-to-one with the statistical proportion.

[0061] By associating the initial second predicted transportation loss rate with the statistical ratio, an initial node is obtained;

[0062] By summing up the initial nodes, we obtain the initial node set;

[0063] The second predicted transportation loss rate is obtained based on the initial node set.

[0064] Optionally, obtaining the second predicted transportation loss rate based on the initial node set includes:

[0065] The second predicted transportation loss rate is obtained based on the initial node set and the pre-constructed second predicted transportation loss rate calculation formula, wherein the second predicted transportation loss rate calculation formula is as follows:

[0066]

[0067] in, This represents the second predicted transportation loss rate. Indicates that the initial node set has a total of An initial node, Represents the first node in the initial node set. The initial second predicted transportation loss rate corresponding to each initial node. Represents the first node in the initial node set. The statistical proportions corresponding to each initial node.

[0068] Optionally, obtaining the predicted purchase quantity based on the predicted transportation time, target span duration, predicted transportation loss rate, and current inventory includes:

[0069] The predicted purchase quantity is obtained based on the predicted transportation time, target span duration, predicted transportation loss rate, current inventory level, and a pre-constructed purchase quantity prediction calculation formula, wherein the purchase quantity prediction calculation formula is as follows:

[0070]

[0071] in, This indicates the predicted purchase quantity. This indicates the predicted transportation time. This represents the predicted transportation loss rate. This indicates the duration of the target span.

[0072] To achieve the above objectives, the present invention also provides an intelligent forecasting system for agricultural product procurement volume based on multi-objective optimization, comprising:

[0073] The historical inventory acquisition module is used to receive agricultural product purchase instructions, confirm the purchase quantity intelligent prediction system based on the agricultural product purchase instructions, acquire the historical inventory record database and the historical transportation record database based on the purchase quantity intelligent prediction system, and acquire the target type, wherein the target type is the type of agricultural product to be purchased;

[0074] Based on the target type, a set of historical inventory nodes is identified in the historical inventory record database. The set of historical inventory nodes contains multiple historical inventory nodes, and each historical inventory node contains the historical span duration, historical initial inventory quantity, historical loss quantity, and historical reduction value.

[0075] The predicted reduction acquisition module is used to acquire the predicted average loss rate, the predicted average reduction value, and the target span duration based on the historical inventory node set.

[0076] Obtain the current inventory level and predicted transportation time. Based on the current inventory level, predicted transportation time, predicted average loss rate, and predicted average reduction value, obtain the predicted procurement time.

[0077] The transportation loss prediction module is used to obtain M transportation indicators and, based on the target type and the M transportation indicators, identify a set of historical transportation nodes in the historical transportation record database. The set of historical transportation nodes contains multiple historical transportation nodes, and each historical transportation node contains a transportation loss rate and M transportation indicator values, where M is an integer greater than 1.

[0078] The predicted transportation loss rate is obtained based on the historical transportation node set, M transportation indicators, and the predicted procurement time.

[0079] The predicted purchase volume acquisition module is used to obtain the predicted purchase volume based on the predicted transportation time, target span duration, predicted transportation loss rate and current inventory, and realize intelligent prediction of agricultural product purchase volume based on the predicted purchase time and predicted purchase volume.

[0080] To address the above problems, the present invention also provides an electronic device, the electronic device comprising:

[0081] A memory for storing at least one instruction; and a processor for executing the instructions stored in the memory to implement the above-described intelligent prediction method for agricultural product procurement based on multi-objective optimization.

[0082] To address the aforementioned problems, the present invention also provides a computer-readable storage medium storing at least one instruction, which is executed by a processor in an electronic device to implement the aforementioned intelligent prediction method for agricultural product procurement based on multi-objective optimization.

[0083] To address the problems described in the background art, this invention receives agricultural product procurement instructions, establishes an intelligent procurement quantity prediction system based on these instructions, acquires historical inventory records and historical transportation records based on this system, and obtains a target type, wherein the target type is the type of agricultural product to be procured. Based on the target type, a historical inventory node set is identified in the historical inventory records, wherein the historical inventory node set contains multiple historical inventory nodes, and each historical inventory node includes historical span duration, historical initial inventory quantity, historical loss quantity, and historical reduction value. Based on the historical inventory node set, a predicted average loss rate, a predicted average reduction value, and a target span duration are obtained. Thus, this invention improves the inventory turnover rate of agricultural products per unit time by dividing the shelf life of agricultural products into intervals to obtain the predicted average reduction value, and by maximizing the predicted average reduction value, thereby optimizing the inventory management of agricultural products and reducing unsold losses. The invention obtains the current inventory level and predicted transportation time. Based on the current inventory level, predicted transportation time, predicted average loss rate, and predicted average reduction value, the predicted procurement time is obtained. It is evident that this invention determines the predicted procurement time by comprehensively considering the current inventory level, predicted transportation time, predicted average loss rate, and predicted average reduction value. By combining the current time with the predicted transportation time, procurement can be arranged in advance to avoid stockouts, thereby improving the efficiency of the agricultural product supply chain and ensuring a stable supply of agricultural products. M transportation indicators are obtained. Based on the target type and the M transportation indicators, a set of historical transportation nodes is identified in the historical transportation record database. This set of historical transportation nodes contains multiple historical transportation nodes, and each historical transportation node contains a transportation loss rate and M transportation indicator values, where M is an integer greater than 1. Based on the set of historical transportation nodes, the M transportation indicators, and the predicted procurement time, the predicted transportation loss rate is obtained. It is evident that this invention constructs a primary / secondary node set by rotating the primary indicators, quantifying the independent impact of each transportation indicator on the transportation loss rate. When node clusters are evenly distributed (with the same statistical proportion), the first predicted transportation loss rate acquisition scheme is adopted. The node cluster with the most consistent secondary indicators is selected to fit the curve, highlighting the influence of the primary indicators, thus obtaining the first predicted transportation loss rate. When node clusters are unevenly distributed (with different statistical proportions), the proportional weighted fusion method (the second predicted transportation loss rate acquisition scheme) is adopted. The prediction results of high-proportion target node clusters are integrated and dynamically weighted according to the statistical proportions to obtain the second predicted transportation loss rate. This method can accurately predict the transportation loss rate and improve the accuracy of procurement quantity forecasting using the predicted transportation loss rate. Based on the predicted transportation time, target span duration, predicted transportation loss rate, and current inventory, the predicted procurement quantity is obtained. Based on the predicted procurement time and predicted procurement quantity, intelligent prediction of agricultural product procurement quantity is achieved. It is evident that this embodiment accurately predicts procurement quantity by comprehensively considering transportation time, loss rate, and inventory dynamics. By predicting the procurement time and predicted procurement quantity, a dynamic balance between supply and demand is achieved, avoiding supply and demand imbalances, improving supply chain efficiency, and realizing intelligent procurement forecasting for agricultural products.Therefore, this invention can improve the accuracy and flexibility of intelligent forecasting of agricultural product procurement volume. Attached Figure Description

[0084] Figure 1 This is a flowchart illustrating an intelligent prediction method for agricultural product procurement volume based on multi-objective optimization, provided in an embodiment of the present invention.

[0085] Figure 2 A functional block diagram of an intelligent agricultural product procurement quantity prediction system based on multi-objective optimization provided in an embodiment of the present invention;

[0086] Figure 3 This is a schematic diagram of the structure of an electronic device that implements the intelligent prediction method for agricultural product procurement volume based on multi-objective optimization, according to an embodiment of the present invention.

[0087] Figure 4 This is a software interface diagram of an intelligent agricultural product procurement quantity prediction system provided in an embodiment of the present invention.

[0088] Explanation of reference numerals in the attached figures:

[0089] 10. Electronic device; 11. Processor; 12. Memory; 13. Bus.

[0090] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0091] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0092] This application provides a method for intelligent forecasting of agricultural product procurement volume based on multi-objective optimization. The executing entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0093] Reference Figure 1 The diagram shown is a flowchart illustrating an intelligent forecasting method for agricultural product procurement volume based on multi-objective optimization, according to an embodiment of the present invention. In this embodiment, the intelligent forecasting method for agricultural product procurement volume based on multi-objective optimization includes:

[0094] S1. Receive agricultural product procurement instructions, confirm the procurement quantity intelligent prediction system based on the agricultural product procurement instructions, obtain the historical inventory record database and historical transportation record database based on the procurement quantity intelligent prediction system, and obtain the target type, wherein the target type is the type of agricultural product to be procured.

[0095] It should be explained that the agricultural product procurement order is an instruction issued by a person who wants to intelligently predict the procurement quantity of agricultural products. The target type is the specific type of agricultural product to be procured, explicitly specified in the agricultural product procurement order. The intelligent procurement quantity prediction system is software used to predict the procurement quantity of agricultural products. This embodiment of the invention utilizes the agricultural product procurement quantity prediction system to intelligently predict the procurement quantity. The intelligent procurement quantity prediction system uses the target type as an index, retrieves the historical inventory record database corresponding to the agricultural product using the target type, predicts the average loss rate and average reduction value of the agricultural product using the historical inventory record database, and combines this with the current inventory quantity of the agricultural product and the predicted transportation time to obtain the predicted procurement time. It also retrieves the historical transportation record database corresponding to the agricultural product using the target type, and intelligently predicts the transportation loss rate and procurement quantity based on the historical transportation record database and the predicted procurement time. The intelligent prediction of the agricultural product procurement quantity is achieved by predicting the procurement time and procurement quantity. Taking into account the historical records of the "inventory-transportation" dual-link system, the loss rate and quantity reduction are first dynamically calculated from the historical inventory record database. Then, the procurement time is locked by combining the transportation duration. Subsequently, the historical transportation record database is searched in reverse using the same target type, and the transportation loss rate is generated according to real-time transportation indicators. Finally, a closed loop is achieved to realize two-dimensional prediction of "procurement time + procurement quantity," realizing a leap from experience-based estimation to data intelligence in agricultural product procurement. For details, please refer to [link to relevant documentation]. Figure 4 As shown, Figure 4 This is a screenshot of the software interface for an intelligent procurement quantity prediction system.

[0096] For example, Zhang, as the purchasing manager of a supermarket, issues the agricultural product purchase order in order to ensure that the supply of corn in the supermarket meets consumer demand. Here, corn is the target type.

[0097] S2. Based on the target type, identify a set of historical inventory nodes in the historical inventory record database. The set of historical inventory nodes contains multiple historical inventory nodes, and each historical inventory node contains the historical span duration, historical initial inventory quantity, historical loss quantity, and historical reduction value.

[0098] Understandably, the historical inventory record database stores information on the inflow and outflow of various agricultural products. Optionally, these agricultural products include, but are not limited to, vegetables, corn, and watermelons. This inflow and outflow information includes the historical duration, historical initial inventory level, historical losses, and historical reduction values.

[0099] For example, at 9:00 AM on June 10th, the supermarket had 10kg of corn remaining and received 90kg of newly purchased corn. By 9:00 AM on June 12th, 80kg of corn had been sold, 5kg had been lost, and 15kg remained. Here, the time span (48 hours) between 9:00 AM on June 10th and 9:00 AM on June 12th is the historical time span. The sum of the remaining 10kg of corn and the newly purchased 90kg of corn (100kg) is the historical initial inventory. The 80kg of corn sold is the historical reduction value, and the 5kg loss is the historical loss amount. For example, historical inventory nodes can be represented as {48h, 100kg, 5kg, 80kg}.

[0100] S3. Based on the historical inventory node set, obtain the predicted average loss rate, the predicted average reduction value, and the target span duration.

[0101] It should be explained that obtaining the predicted average loss rate, predicted average reduction value, and target span duration based on the historical inventory node set includes:

[0102] The shelf life of agricultural products is obtained by dividing the shelf life of agricultural products into multiple shelf life intervals based on a preset division method. Each shelf life interval includes an interval start value, an interval end value, and an interval mid value.

[0103] Based on multiple shelf-life time intervals, the historical inventory node set is divided into multiple analysis inventory node sets. The shelf-life time intervals correspond one-to-one with the analysis inventory node sets, and the historical span duration corresponding to each analysis inventory node in the analysis inventory node set is located within the shelf-life time interval.

[0104] Based on the multiple sets of analyzed inventory nodes and multiple shelf-life intervals, the predicted average loss rate, predicted average reduction value, and target span duration are obtained.

[0105] It is understood that the shelf life of agricultural products refers to the maximum time range during which agricultural products maintain their marketable or usable quality under specific storage conditions. The shelf life of agricultural products is divided into multiple shelf life intervals using a preset method, as follows: For example, assuming the shelf life of agricultural products is 60 hours, meaning that products exceeding 60 hours are considered unsellable, the division method yields multiple shelf life intervals of [0h, 20h], [20h, 40h], and [40h, 60h]. The division method involves dividing the shelf life of agricultural products into three equal shelf life intervals. Taking the shelf life interval [0h, 20h] as an example: the interval starts at 0h, ends at 20h, and has a midpoint of (20h-0h) / 2 = 10h. Other shelf life intervals achieve the same effect as the [0h, 20h] interval, and will not be elaborated further here.

[0106] For example, suppose the historical inventory node set is {(48h, 100kg, 5kg, 80kg), (42h, 100kg, 4kg, 78kg), (32h, 60kg, 3kg, 55kg), (26h, 60kg, 2kg, 50kg), (16h, 30kg, 2kg, 22kg), (12h, 30kg, 1kg, 20kg)}. Here, we only take the historical inventory nodes (48h, 100kg, 5kg, 80kg) and (42h, 100kg, 4kg, 78kg) as examples: the historical span of 48h and 42h both fall within the freshness preservation period [40h, Within the 60h period, the analysis inventory node set corresponding to the shelf life interval [40h, 60h] is {(48h, 100kg, 5kg, 80kg), (42h, 100kg, 4kg, 78kg)}. Similarly, the analysis inventory node set corresponding to the shelf life interval [20h, 40h] is {(32h, 60kg, 3kg, 55kg), (26h, 60kg, 2kg, 50kg)}, and the analysis inventory node set corresponding to the shelf life interval [0h, 20h] is {(16h, 30kg, 2kg, 22kg), (12h, 30kg, 1kg, 20kg)}.

[0107] Furthermore, the step of obtaining the predicted average loss rate, predicted average reduction value, and target span duration based on the multiple sets of analyzed inventory nodes and multiple shelf-life intervals includes:

[0108] An analysis inventory node set is extracted sequentially from multiple analysis inventory node sets, and the corresponding freshness period interval is determined in multiple freshness period intervals based on the analysis inventory node set.

[0109] The average reduction value is obtained based on the analyzed inventory node set, the corresponding shelf life interval of the analyzed inventory node set, and the pre-constructed reduction value calculation formula, wherein the reduction value calculation formula is as follows:

[0110]

[0111] in, The assessed average reduction value, Indicates the first node in the analysis inventory node set Analyze the historical reduction values ​​corresponding to each inventory node. This indicates that the analysis of the inventory node set has a total of One analysis of inventory nodes, Indicates the first node in the analysis inventory node set The analysis focuses on the absolute time difference between the historical span duration corresponding to each inventory node and the median value of the corresponding shelf-life interval. Indicates the first node in the analysis inventory node set The historical span duration corresponding to each inventory node is analyzed;

[0112] The average reduction values ​​are summarized to obtain the set of average reduction values. The average reduction value with the largest value in the set is the predicted average reduction value. The set of analysis inventory nodes corresponding to the predicted average reduction value is the predicted inventory node set. The midpoint of the shelf life interval corresponding to the predicted average reduction value is the target span duration.

[0113] Predicted inventory nodes are extracted sequentially from the predicted inventory node set, and an initial average loss rate is obtained based on the predicted inventory nodes, wherein the initial average loss rate is the ratio of the historical loss amount to the historical span duration in the predicted inventory nodes.

[0114] The initial average loss rates are summarized to obtain multiple initial average loss rates, and the average of the multiple initial average loss rates is used as the predicted average loss rate.

[0115] For example, taking the freshness period interval [0h, 20h] and the corresponding analysis inventory node set {(16h, 30kg, 2kg, 22kg), (12h, 30kg, 1kg, 20kg)} as an example: the absolute time difference between the historical span duration of 16h corresponding to the analysis inventory node (16h, 30kg, 2kg, 22kg) and the interval median of 10h corresponding to the freshness period interval is 6h (the absolute difference between 16h and 10h). Similarly, the absolute time difference between the historical span duration of 12h corresponding to the analysis inventory node (12h, 30kg, 1kg, 20kg) and the interval median of 10h corresponding to the freshness period interval is 2h (the absolute difference between 12h and 10h).

[0116] It should be understood that the assessed average reduction value is a weighted average of the rate of consumption of agricultural product inventory per unit time, calculated based on historical inventory node sets and corresponding shelf-life intervals. It reflects the average rate of inventory reduction corresponding to different shelf-life intervals in historical periods. For example, using the reduction value calculation formula, the assessed average reduction value for the analyzed inventory node sets {(16h, 30kg, 2kg, 22kg), (12h, 30kg, 1kg, 20kg)} is 1.45kg ( / h), for the analyzed inventory node sets {(32h, 60kg, 3kg, 55kg), (26h, 60kg, 2kg, 50kg)} is 1.85kg ( / h), and for the analyzed inventory node sets {(48h, 100kg, 5kg, 80kg), (42h, 100kg, 4kg, 78kg)} is 1.82kg ( / h). A larger assessed average reduction value indicates higher sales volume of agricultural products per unit time. Using an assessed average reduction value of 1.85 kg / h as the predicted average reduction value, and analyzing the inventory node sets {(32h, 60kg, 3kg, 55kg), (26h, 60kg, 2kg, 50kg)} as the predicted inventory node set, the predicted average loss rate is the average of the initial average loss rate (3 / 32) and the initial average loss rate (2 / 26) (8.53%). The predicted average reduction value characterizes the average rate of inventory reduction corresponding to a historical sales cycle for agricultural products. The predicted average loss rate represents the loss rate of agricultural products per unit time within a sales cycle. The target sales cycle is used to characterize the future predicted sales cycle. This embodiment of the invention obtains the predicted average reduction value by dividing the shelf life of agricultural products into intervals, and maximizes the predicted average reduction value to improve the inventory turnover rate of agricultural products per unit time, optimize inventory management, and reduce unsold losses.

[0117] S4. Obtain the current inventory level and predicted transportation time. Based on the current inventory level, predicted transportation time, predicted average loss rate, and predicted average reduction value, obtain the predicted procurement time.

[0118] In detail, the process of obtaining the predicted procurement time based on current inventory levels, predicted transportation time, predicted average loss rate, and predicted average reduction value includes:

[0119] The predicted duration is obtained based on the current inventory level, the predicted average loss rate, the predicted average reduction value, and a pre-constructed formula for calculating the predicted duration. The formula for calculating the predicted duration is as follows:

[0120]

[0121] in, The predicted duration is... This indicates the current inventory level. This indicates the preset minimum inventory level. This represents the predicted average loss rate. This represents the predicted average reduction value;

[0122] Obtain the current time, and based on the current time, the predicted duration, and the predicted transportation time, obtain the predicted procurement time.

[0123] Understandably, the current inventory level represents the remaining inventory of agricultural products as of now, and the predicted transportation time represents the estimated time span from the issuance of a purchase order to the actual warehousing of agricultural products, which can be obtained based on historical transportation data. The predicted duration refers to the maximum safe time span during which the current inventory level can support the warehousing of new purchases, taking into account losses and consumption. The preset minimum inventory level is a manually set inventory safety threshold to avoid stockouts of agricultural products.

[0124] For example, assuming the current time is 9:00 AM on June 15th, the predicted duration is 48 hours, and the predicted transportation time is 12 hours, to ensure uninterrupted supply of agricultural products before 9:00 AM on June 17th, procurement of agricultural products will begin 12 hours before 9:00 AM on June 17th (i.e., 9:00 AM on June 16th). This embodiment of the invention determines the predicted procurement time by comprehensively considering current inventory, predicted transportation time, predicted average loss rate, and predicted average reduction value. By combining the current time with the predicted transportation time, procurement can be arranged in advance to avoid stockouts, improve the efficiency of the agricultural product supply chain, and ensure a stable supply of agricultural products.

[0125] S5. Obtain M transportation indicators. Based on the target type and the M transportation indicators, identify a set of historical transportation nodes in the historical transportation record database. The set of historical transportation nodes contains multiple historical transportation nodes, and each historical transportation node contains a transportation loss rate and M transportation indicator values, where M is an integer greater than 1.

[0126] Understandably, the transportation indicators represent the influencing factors that determine the extent of agricultural product loss during transportation. Optionally, the transportation indicators include, but are not limited to, transportation time, number of loading and unloading operations, and temperature control compliance rate. The historical transportation record database stores the values ​​of various transportation indicators and transportation loss rates for various agricultural products from the issuance of the purchase order to safe storage. The transportation indicator values ​​represent the quantitative values ​​of the transportation indicators.

[0127] For example, taking the transportation time, number of loading and unloading operations, and temperature control compliance rate in the transportation indicators as examples: Assume the historical transportation node is {3.2%-24h-5 times-90%}, indicating that a certain transportation lasted 24 hours, with 5 loading and unloading operations en route, and the temperature control compliance rate was 90%, resulting in a 3.2% loss of agricultural products. Here, 3.2% is the transportation loss rate, 24h is the transportation indicator value corresponding to the transportation time, 5 times is the transportation indicator value corresponding to the number of loading and unloading operations, and 90% is the transportation indicator value corresponding to the temperature control compliance rate.

[0128] S6. Based on the historical transportation node set, M transportation indicators and the predicted procurement time, obtain the predicted transportation loss rate.

[0129] It should be explained that the step of obtaining the predicted transportation loss rate based on the historical transportation node set, M transportation indicators, and the predicted procurement time includes:

[0130] From the M transportation indicators, transportation indicators are extracted sequentially and then removed. The transportation indicators are taken as the primary indicators, and the M-1 transportation indicators after removal are taken as the M-1 secondary indicators.

[0131] Based on the main indicators, M-1 secondary indicators and historical transportation node set, a main node set and a secondary node set are obtained. Each main node in the main node set includes a transportation loss rate and a main indicator value with an indicator type. Each secondary node in the secondary node set includes M-1 secondary indicator values. The main nodes in the main node set and the secondary nodes in the secondary node set correspond one-to-one.

[0132] A pre-defined clustering algorithm is used to perform clustering operations on the secondary node set to obtain multiple node clusters;

[0133] The number of minor nodes in each of the multiple node clusters is counted separately to obtain multiple statistical quantities;

[0134] Multiple statistical ratios are obtained based on multiple statistical quantities, wherein each statistical ratio corresponds one-to-one with a statistical quantity, and the statistical ratio is the ratio of the statistical quantity corresponding to the statistical ratio to the sum of multiple statistical quantities.

[0135] Determine whether all statistical proportions are the same across multiple statistical proportions;

[0136] If they are the same, a first predicted transportation loss rate acquisition scheme is constructed based on multiple node clusters and the main node set, and a first predicted transportation loss rate is obtained based on the first predicted transportation loss rate acquisition scheme. Otherwise, a second predicted transportation loss rate acquisition scheme is constructed based on multiple node clusters and the main node set, and a second predicted transportation loss rate is obtained based on the second predicted transportation loss rate acquisition scheme. The first predicted transportation loss rate or the second predicted transportation loss rate is used as the single indicator corresponding to the main indicator to predict the transportation loss rate.

[0137] After confirming that the corresponding single-indicator predicted transportation loss rate has been obtained for each major indicator, the single-indicator predicted transportation loss rates are summarized to obtain multiple single-indicator predicted transportation loss rates. The average of the multiple single-indicator predicted transportation loss rates is calculated to obtain the predicted transportation loss rate.

[0138] For example, assuming M is 3, the corresponding transportation indicators are transportation time, number of loading and unloading operations, and temperature control compliance rate. Generally, agricultural products are affected by various transportation indicators during actual transportation; here, only three transportation indicators are used as examples. Assuming that transportation time is the primary indicator, and the number of loading / unloading operations and temperature control compliance rate are secondary indicators, if the historical transportation indicator node set is {(3.2%-24h-5 times-90%), (3.4%-20h-5 times-95%), (2.5%-16h-3 times-80%), (2.8%-18h-3 times-83%), (4.5%-40h-7 times-92%), (4.8%-42h-7 times-90%)}, then the primary node set is {(3.2%-24h-transportation time), (3.4%-20h-transportation time), (2.5%-16h-transportation time), (2.8%-18h-transportation time)}. -Transportation time), (4.5%-40h-transportation time), (4.8%-42h-transportation time)}, the secondary node set is {(5 times-90%), (5 times-95%), (3 times-80%), (3 times-83%), (7 times-92%), (7 times-90%)}, which can yield multiple node clusters as {(5 times-90%), (5 times-95%)}, {(3 times-80%), (3 times-83%)}, {(7 times-92%), (7 times-90%)}. Optionally, K-Means can be used as the clustering algorithm to obtain multiple node clusters, which is an existing technology and will not be elaborated here.

[0139] Understandably, from the above example, the statistical quantity corresponding to each of the three node clusters is 2, and the corresponding statistical proportions are all 2 / (2+2+2)=1 / 3. Since each statistical proportion is the same in the example, a first predicted transportation loss rate acquisition scheme is constructed. If each statistical proportion is different, a second predicted transportation loss rate acquisition scheme is constructed. The single-indicator predicted transportation loss rate is the expected loss rate of agricultural products during transportation, predicted by cluster analysis of historical transportation nodes when a certain transportation indicator is the main influencing factor. The predicted transportation loss rate is an expected loss rate that comprehensively considers the impact of multiple transportation indicators on the transportation process of agricultural products. For specific implementation processes of the first and second predicted transportation loss rate acquisition schemes, please refer to the following embodiments.

[0140] In detail, the step of constructing a first predicted transportation loss rate acquisition scheme based on multiple node clusters and a main node set, and obtaining the first predicted transportation loss rate based on the first predicted transportation loss rate acquisition scheme, includes:

[0141] A first predicted transportation loss rate acquisition scheme is constructed based on multiple node clusters, wherein the first predicted transportation loss rate acquisition scheme is as follows:

[0142] Perform the following operation on each of the multiple node clusters:

[0143] Calculate the similarity between secondary nodes in a node cluster to obtain the intra-cluster bias.

[0144] The intra-cluster deviations are summarized to obtain an intra-cluster deviation set. The cluster of nodes corresponding to the smallest intra-cluster deviation in the intra-cluster deviation set is the analysis cluster.

[0145] Based on the analysis cluster, multiple main nodes are identified in the main node set, and the main nodes among the multiple main nodes correspond one-to-one with the secondary nodes among the multiple secondary nodes in the analysis cluster.

[0146] Perform the following operation on each of the multiple primary nodes:

[0147] The transportation loss rate and key indicator values ​​are extracted from the main nodes, and the transportation loss rate and key indicator values ​​are correlated to obtain fitted coordinate points, wherein the fitted coordinate points correspond one-to-one with the main nodes.

[0148] By summing the fitted coordinate points, a set of fitted coordinate points is obtained;

[0149] All the fitted coordinate points in the fitted coordinate point set are mapped to a pre-constructed coordinate system to obtain a mapped coordinate point set. The mapped coordinate point set is then fitted to the first predicted curve using a pre-constructed polynomial fitting model.

[0150] Based on the predicted procurement time, a first predicted indicator value with an indicator type is obtained, wherein the indicator type of the first predicted indicator value is the same as the indicator type of the main indicator value.

[0151] Using the first predicted index value, the retrieval index value is identified in the first predicted curve, and the transportation loss rate corresponding to the retrieval index value is the first predicted transportation loss rate.

[0152] It should be understood that the intra-cluster deviation reflects the dispersion among secondary nodes within the same node cluster. The smaller the intra-cluster deviation, the higher the consistency of the secondary nodes within the same node cluster. Optionally, the intra-cluster deviation can be obtained by calculating the average Euclidean distance value corresponding to the node cluster. The process is as follows: calculate the Euclidean distance value corresponding to every two secondary nodes in the node cluster, obtain multiple Euclidean distance values, take the average of the multiple Euclidean distance values, and use the average Euclidean distance value as the intra-cluster deviation. The process of obtaining the Euclidean distance value is feasible with existing technology and will not be elaborated here. Assuming that transportation time is the primary indicator and the number of loading and unloading operations and temperature control compliance rate are secondary indicators, the smaller the intra-cluster deviation, the smaller the impact of the secondary indicators, such as the number of loading and unloading operations and temperature control compliance rate, on the transportation loss rate, thus highlighting the impact of the primary indicator, transportation time, on the transportation loss rate. For example, assuming the analysis cluster is {(7th - 92%), (7th - 90%)}, multiple primary nodes corresponding to each secondary node in the analysis cluster are identified in the primary node set as (4.5% - 40h) and (4.8% - 42h). The process of obtaining the fitting coordinate points by associating the transportation loss rate and the primary index values ​​is as follows: using transportation time as the abscissa and transportation loss rate as the ordinate, the set of fitting coordinate points is obtained as {(40h, 4.5%), (42h, 4.8%)}, resulting in the first prediction curve. Optionally, the polynomial fitting model includes, but is not limited to, standard polynomial regression model and orthogonal polynomial regression model. Generally, multiple fitting coordinate points are usually obtained for fitting the curve; here, only two fitting coordinate points are used as an example.

[0153] It is understood that the first prediction index value refers to the predicted transportation time (when transportation time is the main indicator). The acquisition process is as follows: the transportation time is estimated based on the transportation distance and the average speed of historical transportation. For example, assuming the first prediction index value is (45h - transportation time), the corresponding transportation loss rate is identified in the first prediction curve using the first prediction index value (45h - transportation time), which is the first predicted transportation loss rate.

[0154] Furthermore, the step of constructing a second predicted transportation loss rate acquisition scheme based on multiple node clusters and a main node set, and acquiring the second predicted transportation loss rate based on the second predicted transportation loss rate acquisition scheme, includes:

[0155] A second scheme for obtaining the predicted transportation loss rate is constructed based on multiple node clusters, as shown below:

[0156] Based on a preset statistical ratio threshold, one or more target node clusters are identified from multiple node clusters, wherein the statistical ratio corresponding to the target node cluster is greater than the statistical ratio threshold.

[0157] Perform the following operation on each of one or more target node clusters:

[0158] The second prediction curve is obtained based on the target node cluster and the main node set;

[0159] The second prediction curves are summarized to obtain the second prediction curve set, wherein the second prediction curves correspond one-to-one with the statistical proportions;

[0160] For each second prediction curve in the second prediction curve set, the following operation is performed:

[0161] The second forecast indicator value is obtained based on the predicted procurement time.

[0162] The initial second predicted transportation loss rate is obtained based on the second predicted index value and the second predicted curve, wherein the initial second predicted transportation loss rate corresponds one-to-one with the statistical proportion.

[0163] By associating the initial second predicted transportation loss rate with the statistical ratio, an initial node is obtained;

[0164] By summing up the initial nodes, we obtain the initial node set;

[0165] The second predicted transportation loss rate is obtained based on the initial node set.

[0166] Understandably, if multiple statistical ratios are 1 / 6, 2 / 6, and 3 / 6, and the statistical ratio threshold is 1 / 6, then multiple node clusters corresponding to statistical ratios 2 / 6 and 3 / 6 are considered as multiple target node clusters. Each target node cluster can obtain a corresponding second prediction curve. The process of obtaining the second prediction curve is the same as that of obtaining the first prediction curve, and will not be repeated here. Similarly, the process of obtaining the initial second predicted transportation loss rate is the same as that of obtaining the first predicted transportation loss rate, and will not be repeated here. The process of obtaining the initial node by associating the initial second predicted transportation loss rate with the statistical ratio is as follows: assuming the initial second predicted transportation loss rate corresponding to statistical ratio 3 / 6 is 3%, the initial node can be represented as (3 / 6, 3%). Similarly, the initial node corresponding to statistical ratio 2 / 6 can be obtained.

[0167] It should be explained that obtaining the second predicted transportation loss rate based on the initial node set includes:

[0168] The second predicted transportation loss rate is obtained based on the initial node set and the pre-constructed second predicted transportation loss rate calculation formula, wherein the second predicted transportation loss rate calculation formula is as follows:

[0169]

[0170] in, This represents the second predicted transportation loss rate. Indicates that the initial node set has a total of An initial node, Represents the first node in the initial node set. The initial second predicted transportation loss rate corresponding to each initial node. Represents the first node in the initial node set. The statistical proportions corresponding to each initial node.

[0171] For example, the second predicted transportation loss rate is a weighted average loss rate obtained by integrating the prediction results of multiple target node clusters. This embodiment of the invention constructs primary / secondary node sets by rotating primary indicators, quantifying the independent impact of each transportation indicator on the transportation loss rate. When the node clusters are evenly distributed (with the same statistical proportion), the first predicted transportation loss rate acquisition scheme is adopted, selecting the node cluster with the most consistent secondary indicators to fit the curve, highlighting the influence of the primary indicators, and obtaining the first predicted transportation loss rate. When the node clusters are unevenly distributed (with different statistical proportions), the proportional weighted fusion method (the second predicted transportation loss rate acquisition scheme) is adopted, integrating the prediction results of high-proportion target node clusters and dynamically weighting them according to statistical proportions to obtain the second predicted transportation loss rate. This can accurately predict the transportation loss rate and improve the accuracy of using the predicted transportation loss rate for procurement quantity forecasting.

[0172] S7. Based on the predicted transportation time, target span duration, predicted transportation loss rate and current inventory, obtain the predicted purchase quantity, and realize intelligent prediction of agricultural product purchase quantity based on the predicted purchase time and predicted purchase quantity.

[0173] It should be explained that the method of obtaining the predicted purchase quantity based on the predicted transportation time, target span duration, predicted transportation loss rate, and current inventory level includes:

[0174] The predicted purchase quantity is obtained based on the predicted transportation time, target span duration, predicted transportation loss rate, current inventory level, and a pre-constructed purchase quantity prediction calculation formula, wherein the purchase quantity prediction calculation formula is as follows:

[0175]

[0176] in, This indicates the predicted purchase quantity. This indicates the predicted transportation time. This represents the predicted transportation loss rate. This indicates the duration of the target span.

[0177] For example, suppose the predicted purchase time is 10:00 AM on June 20th, and the predicted purchase quantity is 100 kg. The supermarket's purchasing manager initiates the purchase of agricultural products at 10:00 AM on June 20th, with a purchase quantity of 100 kg. The predicted purchase quantity, taking into account transportation losses and dynamic inventory depletion, is the total amount of agricultural products that need to be purchased in advance to ensure that agricultural products meet demand within a target time period (such as the procurement cycle or sales cycle). By quantifying the impact of transportation losses and inventory depletion, it avoids stockouts due to insufficient procurement or waste caused by over-procurement, thus achieving accurate prediction of the purchase quantity.

[0178] As can be seen, this invention achieves accurate procurement forecasting for agricultural products through multi-objective optimization. First, it dynamically calculates the predicted average loss rate and predicted average reduction value based on historical inventory records, further determining when to purchase (calculating the predicted purchase time). Second, it comprehensively analyzes the impact of multiple indicators (such as temperature and duration) on losses based on historical transportation records, calculating the predicted transportation loss rate and determining the purchase quantity (predicted purchase volume). Finally, it collaboratively predicts inventory losses and transportation losses, generating a comprehensive optimal procurement decision from the multi-objective perspective of agricultural product inventory and transportation. This invention, by comprehensively considering transportation time, loss rate, and inventory dynamics, accurately predicts purchase volume, achieving dynamic supply and demand balance through predicted purchase time and quantity, avoiding supply and demand imbalances, improving supply chain efficiency, and realizing intelligent procurement forecasting for agricultural products.

[0179] To address the problems described in the background art, this invention receives agricultural product procurement instructions, establishes an intelligent procurement quantity prediction system based on these instructions, acquires historical inventory records and historical transportation records based on this system, and obtains a target type, wherein the target type is the type of agricultural product to be procured. Based on the target type, a historical inventory node set is identified in the historical inventory records, wherein the historical inventory node set contains multiple historical inventory nodes, and each historical inventory node includes historical span duration, historical initial inventory quantity, historical loss quantity, and historical reduction value. Based on the historical inventory node set, a predicted average loss rate, a predicted average reduction value, and a target span duration are obtained. Thus, this invention improves the inventory turnover rate of agricultural products per unit time by dividing the shelf life of agricultural products into intervals to obtain the predicted average reduction value, and by maximizing the predicted average reduction value, thereby optimizing the inventory management of agricultural products and reducing unsold losses. The invention obtains the current inventory level and predicted transportation time. Based on the current inventory level, predicted transportation time, predicted average loss rate, and predicted average reduction value, the predicted procurement time is obtained. It is evident that this invention determines the predicted procurement time by comprehensively considering the current inventory level, predicted transportation time, predicted average loss rate, and predicted average reduction value. By combining the current time with the predicted transportation time, procurement can be arranged in advance to avoid stockouts, thereby improving the efficiency of the agricultural product supply chain and ensuring a stable supply of agricultural products. M transportation indicators are obtained. Based on the target type and the M transportation indicators, a set of historical transportation nodes is identified in the historical transportation record database. This set of historical transportation nodes contains multiple historical transportation nodes, and each historical transportation node contains a transportation loss rate and M transportation indicator values, where M is an integer greater than 1. Based on the set of historical transportation nodes, the M transportation indicators, and the predicted procurement time, the predicted transportation loss rate is obtained. It is evident that this invention constructs a primary / secondary node set by rotating the primary indicators, quantifying the independent impact of each transportation indicator on the transportation loss rate. When node clusters are evenly distributed (with the same statistical proportion), the first predicted transportation loss rate acquisition scheme is adopted. The node cluster with the most consistent secondary indicators is selected to fit the curve, highlighting the influence of the primary indicators, thus obtaining the first predicted transportation loss rate. When node clusters are unevenly distributed (with different statistical proportions), the proportional weighted fusion method (the second predicted transportation loss rate acquisition scheme) is adopted. The prediction results of high-proportion target node clusters are integrated and dynamically weighted according to the statistical proportions to obtain the second predicted transportation loss rate. This method can accurately predict the transportation loss rate and improve the accuracy of procurement quantity forecasting using the predicted transportation loss rate. Based on the predicted transportation time, target span duration, predicted transportation loss rate, and current inventory, the predicted procurement quantity is obtained. Based on the predicted procurement time and predicted procurement quantity, intelligent prediction of agricultural product procurement quantity is achieved. It is evident that this embodiment accurately predicts procurement quantity by comprehensively considering transportation time, loss rate, and inventory dynamics. By predicting the procurement time and predicted procurement quantity, a dynamic balance between supply and demand is achieved, avoiding supply and demand imbalances, improving supply chain efficiency, and realizing intelligent procurement forecasting for agricultural products.Therefore, this invention can improve the accuracy and flexibility of intelligent forecasting of agricultural product procurement volume.

[0180] like Figure 2 The diagram shown is a functional block diagram of an intelligent agricultural product procurement quantity prediction system based on multi-objective optimization provided by an embodiment of the present invention.

[0181] The intelligent agricultural product procurement quantity forecasting system 100 based on multi-objective optimization described in this invention can be installed in an electronic device. Depending on the functions implemented, the intelligent agricultural product procurement quantity forecasting system 100 may include a historical inventory acquisition module 101, a predicted reduction acquisition module 102, a transportation loss prediction module 103, and a predicted procurement quantity acquisition module 104. The modules described in this invention can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, stored in the memory of the electronic device.

[0182] The historical inventory acquisition module 101 is used to receive agricultural product purchase instructions, confirm the purchase quantity intelligent prediction system based on the agricultural product purchase instructions, acquire the historical inventory record library and the historical transportation record library based on the purchase quantity intelligent prediction system, and acquire the target type, wherein the target type is the type of agricultural product to be purchased.

[0183] Based on the target type, a set of historical inventory nodes is identified in the historical inventory record database. The set of historical inventory nodes contains multiple historical inventory nodes, and each historical inventory node contains the historical span duration, historical initial inventory quantity, historical loss quantity, and historical reduction value.

[0184] The predicted reduction acquisition module 102 is used to acquire the predicted average loss rate, the predicted average reduction value and the target span duration based on the historical inventory node set.

[0185] Obtain the current inventory level and predicted transportation time. Based on the current inventory level, predicted transportation time, predicted average loss rate, and predicted average reduction value, obtain the predicted procurement time.

[0186] The transportation loss prediction module 103 is used to obtain M transportation indicators and, based on the target type and the M transportation indicators, identify a set of historical transportation nodes in the historical transportation record database. The set of historical transportation nodes contains multiple historical transportation nodes, and each historical transportation node contains a transportation loss rate and M transportation indicator values, where M is an integer greater than 1.

[0187] The predicted transportation loss rate is obtained based on the historical transportation node set, M transportation indicators, and the predicted procurement time.

[0188] The predicted purchase volume acquisition module 104 is used to acquire the predicted purchase volume based on the predicted transportation time, target span duration, predicted transportation loss rate and current inventory, and realize intelligent prediction of agricultural product purchase volume based on the predicted purchase time and predicted purchase volume.

[0189] In detail, the modules in the intelligent agricultural product procurement quantity prediction system 100 based on multi-objective optimization described in this embodiment of the invention employ the same methods as described above. Figure 1 The method uses the same technical means as the intelligent prediction method for agricultural product procurement volume based on multi-objective optimization described in the article, and can produce the same technical effect, so it will not be elaborated here.

[0190] like Figure 3 The diagram shown is a schematic representation of an electronic device for implementing a multi-objective optimization-based intelligent prediction method for agricultural product procurement volume, according to an embodiment of the present invention.

[0191] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and capable of running on the processor 10, such as a program for intelligent prediction of agricultural product purchase volume based on multi-objective optimization.

[0192] The memory 11 includes at least one type of readable storage medium, such as flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the electronic device 1, such as the portable hard drive of the electronic device 1. In other embodiments, the memory 11 can be an external storage device of the electronic device 1, such as a plug-in portable hard drive, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device 1. Furthermore, the memory 11 includes both internal storage units and external storage devices of the electronic device 1. The memory 11 can be used not only to store application software and various types of data installed on the electronic device 1, such as the code of a multi-objective optimization-based intelligent prediction method for agricultural product procurement quantities, but also to temporarily store data that has been output or will be output.

[0193] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device, connecting various components of the entire electronic device through various interfaces and lines. It executes programs or modules stored in the memory 11 (e.g., a multi-objective optimization-based intelligent prediction method for agricultural product procurement quantities) and calls data stored in the memory 11 to perform various functions of the electronic device 1 and process data.

[0194] The bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 can be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize the connection and communication between the memory 11 and at least one processor 10, etc.

[0195] Figure 3 Only electronic devices with components are shown; those skilled in the art will understand that... Figure 3 The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0196] For example, although not shown, the electronic device 1 may also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power supplies, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0197] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, a Bluetooth interface, etc.), which is typically used to establish communication connections between the electronic device 1 and other electronic devices.

[0198] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), and optionally, a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device 1 and to display a visual user interface.

[0199] The program for intelligent prediction of agricultural product purchase volume based on multi-objective optimization, stored in the memory 11 of the electronic device 1, is a combination of multiple instructions. When run in the processor 10, it can achieve the following:

[0200] Receive agricultural product procurement instructions, confirm the procurement quantity intelligent prediction system based on the agricultural product procurement instructions, obtain historical inventory record database and historical transportation record database based on the procurement quantity intelligent prediction system, and obtain the target type, wherein the target type is the type of agricultural product to be procured;

[0201] Based on the target type, a set of historical inventory nodes is identified in the historical inventory record database. The set of historical inventory nodes contains multiple historical inventory nodes, and each historical inventory node contains the historical span duration, historical initial inventory quantity, historical loss quantity, and historical reduction value.

[0202] Based on the historical inventory node set, the predicted average loss rate, predicted average reduction value, and target span duration are obtained.

[0203] Obtain the current inventory level and predicted transportation time. Based on the current inventory level, predicted transportation time, predicted average loss rate, and predicted average reduction value, obtain the predicted procurement time.

[0204] Get M transportation indicators, and based on the target type and the M transportation indicators, identify a set of historical transportation nodes in the historical transportation record database. The set of historical transportation nodes contains multiple historical transportation nodes, and each historical transportation node contains a transportation loss rate and M transportation indicator values, where M is an integer greater than 1.

[0205] The predicted transportation loss rate is obtained based on the historical transportation node set, M transportation indicators, and the predicted procurement time.

[0206] The predicted purchase quantity is obtained based on the predicted transportation time, target span duration, predicted transportation loss rate, and current inventory. Based on the predicted purchase time and predicted purchase quantity, intelligent prediction of agricultural product purchase quantity is realized.

[0207] Specifically, the processor 10's implementation method for the above instructions can be found in [reference needed]. Figures 1 to 3 The descriptions of the relevant steps in the corresponding embodiments are not repeated here.

[0208] Furthermore, if the modules / units integrated in the electronic device 1 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0209] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor of an electronic device, can perform the following:

[0210] Receive agricultural product procurement instructions, confirm the procurement quantity intelligent prediction system based on the agricultural product procurement instructions, obtain historical inventory record database and historical transportation record database based on the procurement quantity intelligent prediction system, and obtain the target type, wherein the target type is the type of agricultural product to be procured;

[0211] Based on the target type, a set of historical inventory nodes is identified in the historical inventory record database. The set of historical inventory nodes contains multiple historical inventory nodes, and each historical inventory node contains the historical span duration, historical initial inventory quantity, historical loss quantity, and historical reduction value.

[0212] Based on the historical inventory node set, the predicted average loss rate, predicted average reduction value, and target span duration are obtained.

[0213] Obtain the current inventory level and predicted transportation time. Based on the current inventory level, predicted transportation time, predicted average loss rate, and predicted average reduction value, obtain the predicted procurement time.

[0214] Get M transportation indicators, and based on the target type and the M transportation indicators, identify a set of historical transportation nodes in the historical transportation record database. The set of historical transportation nodes contains multiple historical transportation nodes, and each historical transportation node contains a transportation loss rate and M transportation indicator values, where M is an integer greater than 1.

[0215] The predicted transportation loss rate is obtained based on the historical transportation node set, M transportation indicators, and the predicted procurement time.

[0216] The predicted purchase quantity is obtained based on the predicted transportation time, target span duration, predicted transportation loss rate, and current inventory. Based on the predicted purchase time and predicted purchase quantity, intelligent prediction of agricultural product purchase quantity is realized.

[0217] In the embodiments provided by this invention, it should be understood that the disclosed devices, systems, and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative, and actual implementations may have other classification methods.

[0218] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0219] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0220] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0221] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for intelligent forecasting of agricultural product procurement volume based on multi-objective optimization, characterized in that, The method includes: Receive agricultural product procurement instructions, confirm the procurement quantity intelligent prediction system based on the agricultural product procurement instructions, obtain historical inventory record database and historical transportation record database based on the procurement quantity intelligent prediction system, and obtain the target type, wherein the target type is the type of agricultural product to be procured; Based on the target type, a set of historical inventory nodes is identified in the historical inventory record database. The set of historical inventory nodes contains multiple historical inventory nodes, and each historical inventory node contains the historical span duration, historical initial inventory quantity, historical loss quantity, and historical reduction value. Based on the historical inventory node set, the predicted average loss rate, predicted average reduction value, and target span duration are obtained. Obtain the current inventory level and predicted transportation time. Based on the current inventory level, predicted transportation time, predicted average loss rate, and predicted average reduction value, obtain the predicted procurement time. Get M transportation indicators, and based on the target type and the M transportation indicators, identify a set of historical transportation nodes in the historical transportation record database. The set of historical transportation nodes contains multiple historical transportation nodes, and each historical transportation node contains a transportation loss rate and M transportation indicator values, where M is an integer greater than 1. The predicted transportation loss rate is obtained based on the historical transportation node set, M transportation indicators, and the predicted procurement time. The predicted purchase quantity is obtained based on the predicted transportation time, target span duration, predicted transportation loss rate, and current inventory. Based on the predicted purchase time and predicted purchase quantity, intelligent prediction of agricultural product purchase quantity is realized.

2. The intelligent forecasting method for agricultural product procurement volume based on multi-objective optimization as described in claim 1, characterized in that, The process of obtaining the predicted average loss rate, predicted average reduction value, and target span duration based on the historical inventory node set includes: The shelf life of agricultural products is obtained by dividing the shelf life of agricultural products into multiple shelf life intervals based on a preset division method. Each shelf life interval includes an interval start value, an interval end value, and an interval mid value. Based on multiple shelf-life time intervals, the historical inventory node set is divided into multiple analysis inventory node sets. The shelf-life time intervals correspond one-to-one with the analysis inventory node sets, and the historical span duration corresponding to each analysis inventory node in the analysis inventory node set is located within the shelf-life time interval. Based on the multiple sets of analyzed inventory nodes and multiple shelf-life intervals, the predicted average loss rate, predicted average reduction value, and target span duration are obtained.

3. The intelligent forecasting method for agricultural product procurement volume based on multi-objective optimization as described in claim 2, characterized in that, The process of obtaining the predicted average loss rate, predicted average reduction value, and target span duration based on the multiple sets of analyzed inventory nodes and multiple shelf-life intervals includes: An analysis inventory node set is extracted sequentially from multiple analysis inventory node sets, and the corresponding freshness period interval is determined in multiple freshness period intervals based on the analysis inventory node set. The average reduction value is obtained based on the analysis of the inventory node set, the shelf life interval corresponding to the analysis of the inventory node set, and the pre-constructed reduction value calculation formula. The average reduction values ​​are summarized to obtain the set of average reduction values. The average reduction value with the largest value in the set is the predicted average reduction value. The set of analysis inventory nodes corresponding to the predicted average reduction value is the predicted inventory node set. The midpoint of the shelf life interval corresponding to the predicted average reduction value is the target span duration. Predicted inventory nodes are extracted sequentially from the predicted inventory node set, and an initial average loss rate is obtained based on the predicted inventory nodes, wherein the initial average loss rate is the ratio of the historical loss amount to the historical span duration in the predicted inventory nodes. The initial average loss rates are summarized to obtain multiple initial average loss rates, and the average of the multiple initial average loss rates is used as the predicted average loss rate.

4. The intelligent forecasting method for agricultural product procurement volume based on multi-objective optimization as described in claim 3, characterized in that, The process of obtaining the predicted procurement time based on current inventory levels, predicted transportation time, predicted average loss rate, and predicted average reduction value includes: The predicted duration is obtained based on the current inventory level, the predicted average loss rate, the predicted average reduction value, and the pre-built prediction duration calculation formula. Obtain the current time, and based on the current time, the predicted duration, and the predicted transportation time, obtain the predicted procurement time.

5. The intelligent forecasting method for agricultural product procurement volume based on multi-objective optimization as described in claim 4, characterized in that, The method of obtaining the predicted transportation loss rate based on the historical transportation node set, M transportation indicators, and the predicted procurement time includes: From the M transportation indicators, transportation indicators are extracted sequentially and then removed. The transportation indicators are taken as the primary indicators, and the M-1 transportation indicators after removal are taken as the M-1 secondary indicators. Based on the main indicators, M-1 secondary indicators and historical transportation node set, a main node set and a secondary node set are obtained. Each main node in the main node set includes a transportation loss rate and a main indicator value with an indicator type. Each secondary node in the secondary node set includes M-1 secondary indicator values. The main nodes in the main node set and the secondary nodes in the secondary node set correspond one-to-one. A pre-defined clustering algorithm is used to perform clustering operations on the secondary node set to obtain multiple node clusters; The number of minor nodes in each of the multiple node clusters is counted separately to obtain multiple statistical quantities; Multiple statistical ratios are obtained based on multiple statistical quantities, wherein each statistical ratio corresponds one-to-one with a statistical quantity, and the statistical ratio is the ratio of the statistical quantity corresponding to the statistical ratio to the sum of multiple statistical quantities. Determine whether all statistical proportions are the same across multiple statistical proportions; If they are the same, a first predicted transportation loss rate acquisition scheme is constructed based on multiple node clusters and the main node set, and a first predicted transportation loss rate is obtained based on the first predicted transportation loss rate acquisition scheme. Otherwise, a second predicted transportation loss rate acquisition scheme is constructed based on multiple node clusters and the main node set, and a second predicted transportation loss rate is obtained based on the second predicted transportation loss rate acquisition scheme. The first predicted transportation loss rate or the second predicted transportation loss rate is used as the single indicator corresponding to the main indicator to predict the transportation loss rate. After confirming that the corresponding single-indicator predicted transportation loss rate has been obtained for each major indicator, the single-indicator predicted transportation loss rates are summarized to obtain multiple single-indicator predicted transportation loss rates. The average of the multiple single-indicator predicted transportation loss rates is calculated to obtain the predicted transportation loss rate.

6. The intelligent forecasting method for agricultural product procurement volume based on multi-objective optimization as described in claim 5, characterized in that, The first predicted transportation loss rate acquisition scheme is constructed based on multiple node clusters and a main node set. The first predicted transportation loss rate is obtained based on the first predicted transportation loss rate acquisition scheme, including: A first predicted transportation loss rate acquisition scheme is constructed based on multiple node clusters, wherein the first predicted transportation loss rate acquisition scheme is as follows: Perform the following operation on each of the multiple node clusters: Calculate the similarity between secondary nodes in a node cluster to obtain the intra-cluster bias. The intra-cluster deviations are summarized to obtain an intra-cluster deviation set. The cluster of nodes corresponding to the smallest intra-cluster deviation in the intra-cluster deviation set is the analysis cluster. Based on the analysis cluster, multiple main nodes are identified in the main node set, and the main nodes among the multiple main nodes correspond one-to-one with the secondary nodes among the multiple secondary nodes in the analysis cluster. Perform the following operation on each of the multiple primary nodes: The transportation loss rate and key indicator values ​​are extracted from the main nodes, and the transportation loss rate and key indicator values ​​are correlated to obtain fitted coordinate points, wherein the fitted coordinate points correspond one-to-one with the main nodes. By summing the fitted coordinate points, a set of fitted coordinate points is obtained; All the fitted coordinate points in the fitted coordinate point set are mapped to a pre-constructed coordinate system to obtain a mapped coordinate point set. The mapped coordinate point set is then fitted to the first predicted curve using a pre-constructed polynomial fitting model. Based on the predicted procurement time, a first predicted indicator value with an indicator type is obtained, wherein the indicator type of the first predicted indicator value is the same as the indicator type of the main indicator value. Using the first predicted index value, the retrieval index value is identified in the first predicted curve, and the transportation loss rate corresponding to the retrieval index value is the first predicted transportation loss rate.

7. The intelligent forecasting method for agricultural product procurement volume based on multi-objective optimization as described in claim 6, characterized in that, The second predicted transportation loss rate acquisition scheme is constructed based on multiple node clusters and the main node set. The second predicted transportation loss rate is obtained based on the second predicted transportation loss rate acquisition scheme, including: A second scheme for obtaining the predicted transportation loss rate is constructed based on multiple node clusters, as shown below: Based on a preset statistical ratio threshold, one or more target node clusters are identified from multiple node clusters, wherein the statistical ratio corresponding to the target node cluster is greater than the statistical ratio threshold. Perform the following operation on each of one or more target node clusters: The second prediction curve is obtained based on the target node cluster and the main node set; The second prediction curves are summarized to obtain the second prediction curve set, wherein the second prediction curves correspond one-to-one with the statistical proportions; For each second prediction curve in the second prediction curve set, the following operation is performed: The second forecast indicator value is obtained based on the predicted procurement time. The initial second predicted transportation loss rate is obtained based on the second predicted index value and the second predicted curve, wherein the initial second predicted transportation loss rate corresponds one-to-one with the statistical proportion. By associating the initial second predicted transportation loss rate with the statistical ratio, an initial node is obtained; By summing up the initial nodes, we obtain the initial node set; The second predicted transportation loss rate is obtained based on the initial node set.

8. The intelligent forecasting method for agricultural product procurement volume based on multi-objective optimization as described in claim 7, characterized in that, The step of obtaining the second predicted transportation loss rate based on the initial node set includes: The second predicted transportation loss rate is obtained based on the initial node set and the pre-constructed second predicted transportation loss rate calculation formula.

9. The intelligent forecasting method for agricultural product procurement volume based on multi-objective optimization as described in claim 8, characterized in that, The method of obtaining the predicted purchase quantity based on predicted transportation time, target span duration, predicted transportation loss rate, and current inventory includes: The predicted purchase quantity is obtained based on the predicted transportation time, target span duration, predicted transportation loss rate, current inventory level, and pre-built purchase quantity prediction calculation formula.

10. A smart forecasting system for agricultural product procurement volume based on multi-objective optimization, characterized in that, The system includes: The historical inventory acquisition module is used to receive agricultural product purchase instructions, confirm the purchase quantity intelligent prediction system based on the agricultural product purchase instructions, acquire the historical inventory record database and the historical transportation record database based on the purchase quantity intelligent prediction system, and acquire the target type, wherein the target type is the type of agricultural product to be purchased; Based on the target type, a set of historical inventory nodes is identified in the historical inventory record database. The set of historical inventory nodes contains multiple historical inventory nodes, and each historical inventory node contains the historical span duration, historical initial inventory quantity, historical loss quantity, and historical reduction value. The predicted reduction acquisition module is used to acquire the predicted average loss rate, the predicted average reduction value, and the target span duration based on the historical inventory node set. Obtain the current inventory level and predicted transportation time. Based on the current inventory level, predicted transportation time, predicted average loss rate, and predicted average reduction value, obtain the predicted procurement time. The transportation loss prediction module is used to obtain M transportation indicators and, based on the target type and the M transportation indicators, identify a set of historical transportation nodes in the historical transportation record database. The set of historical transportation nodes contains multiple historical transportation nodes, and each historical transportation node contains a transportation loss rate and M transportation indicator values, where M is an integer greater than 1. The predicted transportation loss rate is obtained based on the historical transportation node set, M transportation indicators, and the predicted procurement time. The predicted purchase volume acquisition module is used to obtain the predicted purchase volume based on the predicted transportation time, target span duration, predicted transportation loss rate and current inventory, and realize intelligent prediction of agricultural product purchase volume based on the predicted purchase time and predicted purchase volume.

Images