Agricultural product supply chain management system based on SaaS architecture

Abstract

The application discloses a kind of agricultural product supply chain management systems based on SaaS architecture, including batch record module, environmental exposure processing module, quality analysis module, order matching module and abnormal positioning module;Batch record module establishes agricultural product batch life cycle record;Environmental exposure processing module divides environmental exposure time period and determines environmental exposure state;Quality analysis module determines batch quality state value and remaining shelf life;Order matching module determines candidate performance batch and effective inventory according to batch quality state value, remaining shelf life and target order data, and generates performance execution signal;Abnormal positioning module determines abnormal node according to the quality state change between supply chain nodes and environmental exposure record, and generates abnormal positioning signal;The application realizes the dynamic determination of agricultural product quality state, actual performance ability and supply chain abnormal node.

Description

Technical Field

[0001] This invention relates to the field of agricultural product supply chain information management technology, specifically to an agricultural product supply chain management system based on a SaaS architecture. Background Technology

[0002] Agricultural product supply chain management technology refers to the unified management of batch information, environmental conditions, quality data, inventory status, and order fulfillment processes of agricultural products from production, storage, transportation to sales through information collection, data processing, and network communication, in order to achieve informatization and collaboration in the agricultural product supply chain process.

[0003] Existing patent CN114581108A discloses a supply chain end-to-end node traceability system and method, which records product end-to-end node data through a traceability SaaS system to achieve full lifecycle monitoring of the product supply chain. However, this solution mainly focuses on the collection, recording, and traceability of supply chain node information, and does not address how to determine the batch quality status and remaining shelf life based on the historical environmental exposure process of agricultural products, or further determine the dynamic effective inventory level and locate abnormal supply chain nodes by combining specific order fulfillment conditions. Summary of the Invention

[0004] The purpose of this invention is to provide an agricultural product supply chain management system based on a SaaS architecture to solve the problems in the existing agricultural product supply chain management process, which usually relies on book inventory, production date, or current environmental parameters for inventory management and order allocation. This fails to fully reflect the cumulative impact of historical environmental changes experienced by agricultural products at different supply chain nodes on their quality status, resulting in inconsistencies between book inventory and actual fulfillment capabilities, as well as the difficulty in locating specific abnormal supply chain nodes after agricultural products experience quality abnormalities.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A SaaS-based agricultural product supply chain management system includes a batch recording module, an environmental exposure handling module, a quality analysis module, an order matching module, and an anomaly location module. The batch recording module is used to establish a batch lifecycle record for agricultural products. The environmental exposure processing module is used to divide the environmental exposure time period according to the agricultural product batch life cycle record and determine the environmental exposure status corresponding to each environmental exposure time period. The quality analysis module is used to determine the batch quality status value and remaining shelf life of agricultural products based on the environmental exposure status corresponding to each environmental exposure time period. The order matching module is used to determine candidate fulfillment batches and the effective inventory quantity corresponding to each candidate fulfillment batch based on the batch quality status value, remaining shelf life and target order data, determine the target fulfillment batch based on the effective inventory quantity, and generate a fulfillment execution signal; The anomaly location module is used to determine the abnormal node based on the quality status changes of agricultural product batches at different supply chain nodes and the corresponding environmental exposure records, and to generate anomaly location signals.

[0006] Furthermore, the batch recording module writes the node entry events, node departure events, environmental sampling data, quality inspection data, quantity change data, and order association data generated by the same agricultural product batch at different supply chain nodes into the agricultural product batch lifecycle record in chronological order, based on the agricultural product batch identifier.

[0007] Furthermore, the batch recording module adjusts the environmental data sampling period according to the degree of change between adjacent temperature sampling values, and performs time consistency verification on the timestamps corresponding to the data uploaded by different data acquisition terminals, so that the data generated by different supply chain nodes are written into the agricultural product batch life cycle record according to a unified timeline.

[0008] Furthermore, the environmental exposure processing module divides the supply chain process experienced by agricultural product batches into multiple environmental exposure time periods based on changes in supply chain nodes and changes in adjacent environmental sampling data, and determines the environmental exposure level based on the relationship between the environmental parameters corresponding to each environmental exposure time period and preset environmental parameters.

[0009] Furthermore, when the environmental parameters during any environmental exposure period exceed the damage threshold of the corresponding agricultural product and continue to reach the preset duration, an irreversible damage marker is set for that environmental exposure period. The irreversible damage marker is retained in the environmental exposure record of the corresponding agricultural product batch after the environmental parameters subsequently recover to a suitable environmental range.

[0010] Furthermore, when the corresponding environmental exposure time period belongs to a transportation node, the root mean square value of vibration is determined based on the vibration acceleration data, and when the root mean square value of vibration exceeds a preset vibration threshold within a consecutive preset number of calculation windows, a transportation disturbance mark is set for the corresponding environmental exposure time period.

[0011] Furthermore, the quality analysis module determines the stage quality loss based on the environmental exposure level, duration, humidity, transportation disturbance, and irreversible damage corresponding to each environmental exposure time period, and determines the batch quality status value according to the initial quality status of the agricultural product batch and the quality loss at each stage, in the chronological order of each environmental exposure time period.

[0012] Furthermore, the quality analysis module determines the basic remaining shelf life based on the difference between the batch quality status value and the minimum marketable quality threshold, as well as the preset baseline environmental loss parameters. When an agricultural product batch has irreversible damage markings, the basic remaining shelf life is reduced according to the corresponding remaining life reduction parameters to obtain the remaining shelf life.

[0013] Furthermore, when new actual quality test data is obtained, the quality analysis module determines the measured quality status based on the new actual quality test data, compares the measured quality status with the predicted quality status obtained from environmental exposure records, and corrects the environmental loss parameters within the corresponding time interval based on the difference when the difference exceeds a preset calibration threshold.

[0014] Furthermore, the order matching module determines the fulfillment constraints based on the type of agricultural product corresponding to the target order, the target delivery location, the latest delivery time, and the minimum delivery quality requirements. It also determines the expected fulfillment period based on the transportation time between the current supply chain node location of the agricultural product batch and the target delivery location, and excludes agricultural product batches with a remaining shelf life not exceeding the expected fulfillment period.

[0015] Furthermore, the order matching module determines the expected delivery quality status value based on the batch quality status value of the remaining agricultural product batches, the expected fulfillment period, and the environmental loss parameters during the expected fulfillment period, and determines the agricultural product batches whose expected delivery quality status value meets the minimum delivery quality requirements of the target order as candidate fulfillment batches.

[0016] Furthermore, the order matching module determines the effective inventory quantity of each candidate fulfillment batch for the target order based on the book inventory quantity corresponding to the candidate fulfillment batch, the batch quality status value, the relationship between the expected fulfillment cycle and the remaining shelf life, and the irreversible damage status.

[0017] Furthermore, the order matching module sorts multiple candidate fulfillment batches according to the expected delivery quality status value, remaining shelf life, and expected transportation time corresponding to the candidate fulfillment batches, and sequentially accumulates the effective inventory corresponding to each candidate fulfillment batch until the accumulated effective inventory reaches the target order demand quantity. The candidate fulfillment batches participating in the accumulation are then determined as the target fulfillment batches, and a fulfillment execution signal is generated.

[0018] Furthermore, the anomaly location module determines the node state mutation amount based on the entry quality state value and exit quality state value of agricultural product batches when passing through different supply chain nodes, and identifies supply chain nodes whose node state mutation amount exceeds the preset normal change threshold of the corresponding node type as anomaly candidate nodes.

[0019] Furthermore, the anomaly localization module determines the anomaly contribution value based on the duration of the temperature anomaly, the degree of temperature exceeding the limit, the state of irreversible damage, the state of transportation disturbance, and the degree of dwell time corresponding to the anomaly candidate node. The anomaly candidate node whose anomaly contribution value exceeds the preset anomaly confirmation threshold is identified as an anomaly node. The anomaly type is determined based on the environmental exposure record corresponding to the anomaly node, and an anomaly localization signal is generated.

[0020] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention takes agricultural product batches as the data management object, and writes the node flow data, environmental sampling data and quality testing data generated by the same agricultural product batch at different supply chain nodes into the agricultural product batch life cycle record. According to the changes in supply chain nodes and environmental parameters, the supply chain process is divided into multiple environmental exposure time periods, so that the determination of the quality status of agricultural products not only depends on the current environmental parameters, but also reflects the cumulative impact of historical environmental exposure processes.

[0021] This invention determines the stage quality loss based on the environmental exposure status corresponding to different environmental exposure time periods, and determines the batch quality status value and remaining shelf life accordingly. For historical environmental processes that have reached the damage conditions, irreversible damage markers are set so that after the environment returns to normal, the quality damage that has already occurred still participates in the determination of subsequent quality status and remaining shelf life, thereby improving the accuracy of agricultural product status analysis.

[0022] This invention determines candidate fulfillment batches and effective inventory based on batch quality status values, remaining shelf life, and fulfillment conditions corresponding to target orders. This allows the same agricultural product batch to generate different actual fulfillable inventory for target orders with different delivery distances, different expected fulfillment cycles, and different minimum delivery quality requirements, reducing fulfillment deviations caused by matching orders solely based on book inventory.

[0023] This invention also filters out abnormal candidate nodes based on changes in quality status between different supply chain nodes, and determines the abnormal contribution value and abnormal node by combining the environmental exposure records within the abnormal candidate nodes. This enables reverse positioning from the final quality abnormality of agricultural products to specific supply chain nodes, and establishes a continuous data association between supply chain abnormality analysis and the aforementioned environmental data collection, quality status analysis, and order fulfillment process. Attached Figure Description

[0024] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings: Figure 1 This is a system block diagram of the present invention. Detailed Implementation

[0025] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] This invention provides an agricultural product supply chain management system based on a SaaS architecture. The system is deployed on a cloud server and communicates with business terminals, environmental data acquisition terminals, and quality testing terminals located at agricultural product production nodes, storage nodes, transportation nodes, transit nodes, and sales nodes via a network.

[0027] In this embodiment of the invention, the SaaS architecture refers to an agricultural product supply chain management program uniformly deployed by a cloud server, where different supply chain entities access the agricultural product supply chain management program through corresponding tenant accounts, and the cloud server executes a data processing architecture for receiving agricultural product batch data, storing data, calculating data, matching orders, and issuing business signals.

[0028] In this system, different tenants establish connections with their respective supply chain data through their tenant identifiers. The data uploaded by different supply chain nodes all carry the corresponding tenant identifier, agricultural product batch identifier, supply chain node identifier, and data generation timestamp, enabling the cloud server to complete data management of different agricultural product batches in a unified operating environment.

[0029] In this embodiment of the invention, the agricultural product supply chain node refers to any actual logistics node among the production node, pre-cooling node, processing node, warehousing node, transportation node, transit node, and sales node that agricultural products pass through from the completion of harvesting to the completion of target order delivery.

[0030] This embodiment primarily focuses on perishable agricultural products whose quality changes over time, with variations in environmental temperature, humidity, and transportation disturbances. These include fruits and vegetables, aquatic products, fresh meat, and edible fungi. It should be noted that this invention does not limit the specific type of agricultural product; any agricultural product whose quality changes due to historical environmental exposure during its supply chain circulation can be managed using the supply chain management method described herein.

[0031] The SaaS-based agricultural product supply chain management system includes a cloud server, a database server, a communication interface, and data acquisition terminals set up at different nodes in the supply chain.

[0032] The cloud server includes at least one processor and a memory. The memory stores a supply chain management program. When the processor executes the supply chain management program, it processes agricultural product batch data, environmental sampling data, quality testing data, order data, and supply chain node data.

[0033] The data acquisition terminal includes a processor, a memory, a communication interface, and sensors. Depending on the type of the corresponding supply chain node, the sensors include at least one of a temperature sensor, a humidity sensor, and an acceleration sensor.

[0034] The quality testing terminal is used to obtain quality testing data for corresponding batches of agricultural products. Taking fruits and vegetables as an example, the quality testing data includes at least two of the following: hardness, soluble solids content, weight loss rate, color value, and spoilage rate.

[0035] The system includes a batch recording module, an environmental exposure handling module, a quality analysis module, an order matching module, and an anomaly location module.

[0036] The batch recording module outputs the agricultural product batch lifecycle record, and the environmental exposure processing module calls the agricultural product batch lifecycle record to generate an environmental exposure record table. The quality analysis module determines the batch quality status value and remaining shelf life based on the environmental exposure record table and quality inspection data. The order matching module determines the candidate fulfillment batch, effective inventory, and target fulfillment batch based on the batch quality status value, remaining shelf life, and target order data, and generates a fulfillment execution signal. The anomaly location module determines the anomaly node based on the quality status changes and environmental exposure records corresponding to different supply chain nodes, and generates anomaly location signals.

[0037] Therefore, the modules do not perform different management functions independently. Instead, the data output by the previous module serves as the basis for the data processing of the next module, forming a continuous data processing chain for the same batch of agricultural products, from environmental data collection, quality status determination, order fulfillment to anomaly reverse location.

[0038] The batch record module is used to establish a batch lifecycle record for agricultural products.

[0039] In this embodiment of the invention, the agricultural product batch lifecycle record refers to a set of data records that store, in chronological order, the node entry events, node departure events, environmental sampling data, quality inspection data, quantity change data, and order-related data generated during the period from the establishment of the same agricultural product batch to the completion of delivery, order delivery, or damage reporting.

[0040] Each agricultural product batch lifecycle record includes at least a batch identifier field, a node identifier field, an event type field, a timestamp field, a quantity field, and a status data field.

[0041] The batch identifier field is used to record the unique identification data of the corresponding agricultural product batch; The node identifier field is used to record the supply chain node corresponding to the current data generation. The event type field is used to record one of the following event types: batch creation, entry node, departure node, quality inspection, quantity change, order locking, completion of outbound shipment, and order delivery. The timestamp field is used to record the generation time of the corresponding event or sampled data; The quantity field is used to record the quantity of agricultural products entering the warehouse, the quantity leaving the warehouse, the quantity reported as damaged, or the quantity locked in the order for a batch; The status data field is used to record temperature, humidity, transportation vibration data, and quality inspection data obtained at the corresponding time point or during the corresponding time period.

[0042] When agricultural products are harvested or enter the system for the first time, the data acquisition terminal set up at the corresponding supply chain node sends a batch creation request to the cloud server.

[0043] The batch creation request must include at least the tenant identifier, agricultural product type, variety, place of origin identifier, harvest time, initial quantity, initial quality test data, and current supply chain node identifier.

[0044] After receiving the batch creation request, the cloud server generates an agricultural product batch identifier based on the agricultural product type, harvest date, and sequence number, and establishes an agricultural product batch lifecycle record corresponding to the agricultural product batch identifier in the database server.

[0045] For example, when there are multiple independently managed batches of the same agricultural product harvested on the same day, the cloud server adds sequential numbers to the different agricultural product batches in turn, so that different agricultural product batches correspond to different agricultural product batch identifiers.

[0046] When a batch of agricultural products enters a new supply chain node, the data acquisition terminal set up at the supply chain node reads the batch identifier of the agricultural products and generates a node entry signal.

[0047] In this embodiment of the invention, the node entry signal refers to the computer data signal generated by the data acquisition terminal after recognizing that the corresponding agricultural product batch has entered the current supply chain node, and includes at least the agricultural product batch identifier, the current node identifier, and the node entry time.

[0048] After receiving the node entry signal, the cloud server writes the node entry event into the corresponding agricultural product batch lifecycle record.

[0049] When an agricultural product batch leaves the current supply chain node, the data acquisition terminal generates a node departure signal and sends the agricultural product batch identifier, the current node identifier, and the node departure time to the cloud server, which then writes the node departure event into the corresponding agricultural product batch lifecycle record.

[0050] When agricultural product batches are in the production, pre-cooling, storage or transit stages, temperature and humidity sensors collect ambient temperature and humidity data according to a preset sampling cycle.

[0051] During the transportation phase of agricultural product batches, in addition to collecting ambient temperature and humidity data, the on-board data acquisition terminal also collects transportation vibration data through an acceleration sensor.

[0052] In this embodiment, the basic sampling period for environmental data is set to 1 to 10 minutes.

[0053] To avoid the loss of critical data during temperature abrupt changes due to sampling at fixed time intervals, the data acquisition terminal further adjusts the sampling period based on the degree of change between adjacent temperature sampling values.

[0054] Specifically, when the absolute value of the difference between two adjacent temperature samples is greater than 2℃, the data acquisition terminal adjusts the current sampling period to half of the original sampling period; when the range of three consecutive temperature samples is less than 1℃, the original sampling period is restored.

[0055] For example, the basic sampling period is 10 minutes. When the difference between two adjacent temperature samples reaches 3°C, the subsequent sampling period is adjusted to 5 minutes. If the temperature continues to change rapidly and the difference between adjacent samples exceeds the corresponding threshold again, the sampling frequency will continue to be increased according to the preset rules.

[0056] By using the above method, a low data sampling frequency is maintained during periods of stable temperature change, while the data sampling density is increased when a cold chain is interrupted or there are sudden changes in ambient temperature, in order to preserve time information about the onset, development, and recovery of temperature anomalies.

[0057] The quality inspection terminal obtains the quality inspection data of the corresponding agricultural product batch when the agricultural product is harvested, enters the storage node, leaves the storage node, or the target order is delivered, and sends the agricultural product batch identification, quality inspection data, and inspection time to the cloud server.

[0058] The cloud server writes environmental sampling data, transportation vibration data, and quality inspection data into the corresponding agricultural product batch lifecycle record based on the agricultural product batch identifier carried in the corresponding data, and arranges them according to timestamps.

[0059] Furthermore, to avoid clock errors between different data acquisition terminals causing misalignment in the time sequence of data uploaded by different supply chain nodes, the cloud server performs time consistency verification on the data timestamps.

[0060] Specifically, the cloud server compares the local timestamp generated by the data acquisition terminal with the server receiving timestamp generated when the cloud server receives the corresponding data.

[0061] When the absolute value of the difference between the two is no more than 30 seconds, the local timestamp is used as the basis for data sorting. When the absolute value of the difference between the two exceeds 30 seconds, the cloud server reads the most recent time calibration record of the data acquisition terminal, corrects the local timestamp according to the time offset in the time calibration record, and uses the corrected timestamp as the basis for data sorting.

[0062] Through the above data writing and time calibration processes, data generated by the same batch of agricultural products at different nodes in the supply chain are stored according to a unified timeline.

[0063] The batch recording module ultimately outputs the agricultural product batch lifecycle record for the corresponding agricultural product batch and sends the agricultural product batch lifecycle record to the environmental exposure processing module.

[0064] It should be noted that this invention does not store independent environmental data for each supply chain node and then query it temporarily when needed. Instead, it uses the agricultural product batch identifier as the main index and continuously writes the node flow information, environmental information, quality change information, and quantity change information of the same agricultural product batch into the same agricultural product batch life cycle record.

[0065] Therefore, the data used in the subsequent quality analysis is not the single temperature data of the agricultural product at the current moment, but continuous historical data that can characterize what kind of environment the batch of agricultural product has experienced, how long it has lasted, and whether any environmental anomalies have occurred.

[0066] The environmental exposure processing module is used to divide the supply chain process experienced by agricultural product batches into multiple environmental exposure time periods based on the agricultural product batch life cycle records, and to determine the environmental exposure status corresponding to each environmental exposure time period.

[0067] In this embodiment of the invention, the environmental exposure status refers to a set of data used to characterize the degree of environmental impact experienced by a batch of agricultural products during the corresponding environmental exposure time period, including the environmental exposure level, as well as irreversible damage markers and transportation disturbance markers determined based on actual environmental data.

[0068] In this embodiment of the invention, the environmental exposure time period refers to a continuous data interval in which agricultural product batches are at the same supply chain node within a continuous time range, and the main environmental parameters do not exceed the preset change conditions.

[0069] The environmental exposure processing module first obtains the agricultural product batch lifecycle record corresponding to the batch of agricultural products to be analyzed from the batch record module, and then reads the node entry event, node departure event, temperature sampling data, humidity sampling data and transportation vibration data in sequence according to the timestamp.

[0070] When a batch of agricultural products enters a new node in the supply chain, the moment of entry into the node is taken as the starting point of a new environmental exposure period.

[0071] Within the same supply chain node, when the absolute value of the difference between adjacent temperature samples exceeds the first temperature change threshold, the absolute value of the difference between adjacent humidity samples exceeds the first humidity change threshold, or the transportation vibration data meets the preset transportation disturbance segmentation conditions, the data sampling time that first meets the corresponding conditions is taken as the starting time of a new environmental exposure period.

[0072] In this embodiment, the first temperature change threshold is set to 3°C, and the first humidity change threshold is set to 15 percentage points.

[0073] The preset transportation disturbance segmentation condition refers to the condition that the root mean square value of the vibration corresponding to the transportation vibration data exceeds a preset vibration threshold corresponding to the current agricultural product type for two consecutive sampling periods. When the preset transportation disturbance segmentation condition is met, the sampling time when the preset vibration threshold is first exceeded is taken as the starting time of the new environmental exposure period.

[0074] Based on the above division rules, the environmental exposure treatment module divides the entire agricultural product supply chain process into multiple sequentially arranged environmental exposure time periods.

[0075] For each environmental exposure period, the environmental exposure processing module must at least determine the start time, end time, duration, average temperature, maximum temperature, average humidity, minimum humidity, and the corresponding supply chain node.

[0076] The database server pre-stores environmental parameter tables corresponding to different types of agricultural products.

[0077] The environmental parameter table includes at least the correspondence between agricultural product types, suitable temperature ranges, suitable humidity ranges, temperature warning thresholds, temperature damage thresholds, and duration thresholds.

[0078] The environmental exposure processing module calls up the corresponding environmental parameter table based on the current agricultural product type, and compares the environmental parameters within each environmental exposure time period with the corresponding environmental parameter table to determine the corresponding environmental exposure level.

[0079] In this embodiment of the invention, the environmental exposure level refers to the state level determined based on the degree to which the actual environmental parameters deviate from the suitable environmental conditions for the corresponding agricultural product within an environmental exposure period.

[0080] Taking strawberries as an example, the suitable temperature range is set to 0℃ to 4℃, the temperature warning threshold is set to 8℃, and the temperature damage threshold is set to 15℃.

[0081] When the average temperature of an environmental exposure period does not exceed 4°C, it is defined as a Level 1 environmental exposure period. When the average temperature is above 4℃ but not exceeding 8℃, it is defined as a secondary environmental exposure period. When the average temperature is above 8°C but not exceeding 15°C, it is defined as a Level III environmental exposure period. When the average temperature is above 15℃, it is defined as a Level IV environmental exposure period.

[0082] For other agricultural products, corresponding parameters are configured based on their suitable temperature range and temperature damage patterns.

[0083] Furthermore, the environmental exposure treatment module does not only determine the environmental exposure status based on the average temperature, but also further determines whether a continuous overheating process that reaches the damage condition occurs during the environmental exposure period.

[0084] When there are consecutive temperature sampling values ​​exceeding the corresponding temperature damage threshold for agricultural products within an environmental exposure period, the environmental exposure processing module determines the duration of the continuous overheating process.

[0085] When the duration reaches the preset duration threshold for the corresponding agricultural product, an irreversible damage marker is set for the environmental exposure period.

[0086] In this embodiment of the invention, the irreversible damage marker refers to a data marker used to characterize the historical environmental state of an agricultural product batch that has experienced a damage threshold exceeding the corresponding agricultural product damage threshold and has continued for a preset duration.

[0087] Once the irreversible damage marker is set, it will not be canceled even if the environmental parameters subsequently return to a suitable environmental range.

[0088] For example, when the ambient temperature of a batch of strawberries exceeds 15°C for 30 consecutive minutes, the environmental exposure treatment module sets an irreversible damage marker for the corresponding environmental exposure period.

[0089] Even if the batch of strawberries is subsequently re-exposed to a low-temperature environment below 4°C, the corresponding irreversible damage markers are still retained.

[0090] This is because the subsequent return to normal temperature only indicates that the current environmental conditions have been restored, and does not mean that the quality damage caused by the previous high-temperature process has been reversed and eliminated.

[0091] Furthermore, when the corresponding environmental exposure time period belongs to a transportation node, the environmental exposure processing module reads the vibration acceleration data within that environmental exposure time period and determines the root mean square value of vibration using one minute as a calculation window.

[0092] Furthermore, when the root mean square value of vibration for three consecutive calculation windows exceeds the preset vibration threshold, a transportation disturbance marker is set for the corresponding environmental exposure time period.

[0093] In this embodiment of the invention, the transport disturbance marker refers to a data marker used to characterize that a batch of agricultural products experiences continuous mechanical vibrations exceeding the permissible range during the corresponding transport time period.

[0094] Subsequently, the environmental exposure processing module establishes environmental exposure records for each environmental exposure period.

[0095] The environmental exposure records are arranged in chronological order to form an environmental exposure record table.

[0096] The environmental exposure record form shall include at least the environmental exposure time period number, supply chain node identifier, start time, end time, duration, average temperature, highest temperature, average humidity, lowest humidity, environmental exposure level, irreversible damage marker, and transportation disturbance marker.

[0097] The environmental exposure processing module outputs the environmental exposure record table to the quality analysis module.

[0098] It should be noted that the environmental exposure treatment in this invention is not simply calculating the average temperature that agricultural products have experienced. Instead, it first divides the historical environmental process into multiple environmental exposure time periods with clear time boundaries based on changes in supply chain nodes and environmental parameters, and then determines the environmental exposure level and special damage state in different time periods.

[0099] Therefore, even if two batches of agricultural products have the same temperature at the current moment, as long as the temperature levels, abnormal durations, or transportation disturbances they have experienced are different, their corresponding environmental exposure records will still be different, thus providing a data basis for obtaining different batch quality status values ​​in the future.

[0100] The quality analysis module is used to determine the stage quality loss corresponding to each environmental exposure period based on the environmental exposure status, duration, and quality testing data, and further determine the batch quality status value and remaining shelf life of the agricultural product batch.

[0101] In this embodiment of the invention, the stage quality loss refers to a data value determined based on the environmental exposure level, duration, humidity, transportation disturbance, and irreversible damage state of an environmental exposure period, used to characterize the relative quality loss caused by that environmental exposure period.

[0102] The database server pre-stores environmental degradation parameter tables corresponding to different types of agricultural products.

[0103] The environmental loss parameter table includes at least the correspondence between agricultural product type, environmental exposure level, basic loss value per unit time, humidity correction parameter, additional parameters for transportation disturbance, and irreversible damage increment.

[0104] The quality analysis module reads each environmental exposure time period from the environmental exposure record table in chronological order.

[0105] For any given environmental exposure period, the quality analysis module first reads the corresponding basic loss value per unit time from the environmental loss parameter table based on the current agricultural product type and environmental exposure level.

[0106] Subsequently, the basic quality loss is determined based on the basic loss value per unit time and the duration of the current environmental exposure period.

[0107] For example, all other things being equal, the higher the environmental exposure level, the greater the basic loss value per unit time; the longer the environmental exposure period, the greater the corresponding basic quality loss.

[0108] Furthermore, when the average humidity during an environmental exposure period is outside the suitable humidity range for the corresponding agricultural product, the quality analysis module determines the humidity correction amount based on the degree to which the average humidity deviates from the boundary value of the suitable humidity range, and adds the humidity correction amount to the basic quality loss amount.

[0109] When the corresponding environmental exposure time period has a transportation disturbance marker, the quality analysis module reads the cumulative duration of the vibration root mean square value exceeding the preset vibration threshold, and determines the additional loss due to transportation disturbance based on the additional parameters of transportation disturbance.

[0110] When the corresponding environmental exposure time period has an irreversible damage marker, the quality analysis module calls the irreversible damage increment corresponding to the current agricultural product type and environmental exposure level, and directly adds the irreversible damage increment to the quality loss amount of the corresponding stage.

[0111] Therefore, the stage quality loss corresponding to an environmental exposure period is determined at least by the basic quality loss and corrected for actual humidity anomalies, transportation disturbances, and irreversible damage.

[0112] It should be noted that the increase in irreversible damage will not be removed from the cumulative quality loss as agricultural products subsequently re-enter a suitable temperature environment.

[0113] After the quality analysis module determines the stage quality loss corresponding to all environmental exposure time periods, it determines the current batch quality status value based on the initial quality status of the agricultural product batch and the quality loss at each stage.

[0114] In this embodiment of the invention, the batch quality status value refers to a data value that characterizes the degree of quality maintenance of agricultural products at the current moment, determined based on the initial quality test data of the agricultural product batch and the stage quality loss corresponding to the multiple environmental exposure time periods that the agricultural product batch has experienced.

[0115] In practice, 100 is used as the initial quality benchmark value.

[0116] When the batch recording module obtains complete initial quality inspection data, the quality analysis module calls a preset quality indicator conversion parameter table based on the current agricultural product type. This quality indicator conversion parameter table records at least the correspondence between agricultural product type, quality inspection indicators, indicator benchmark ranges, and corresponding quality correction values. The quality analysis module queries the corresponding quality correction values ​​for each of the obtained quality inspection indicators, including hardness, soluble solids content, weight loss rate, color value, and spoilage rate, and corrects the initial quality benchmark value based on these correction values ​​to obtain the initial quality status of the agricultural product batch.

[0117] For example, for strawberries, when the actual firmness, soluble solids content and spoilage rate fall into the corresponding preset index ranges, the quality correction values ​​corresponding to each preset index range are obtained, and each quality correction value is calculated with the initial quality benchmark value to obtain the initial quality status of the corresponding agricultural product batch.

[0118] Subsequently, the quality analysis module deducts the corresponding stage quality loss from the initial quality status in chronological order of each environmental exposure period to obtain the current batch quality status value.

[0119] For example, if the initial quality status of a batch of agricultural products is 100, the quality loss during the first environmental exposure period is 2, the quality loss during the second environmental exposure period is 5, and the quality loss during the third environmental exposure period is 3, then the current batch quality status is 90 if no other correction conditions are present.

[0120] Furthermore, when the quality inspection terminal obtains new actual quality inspection data, the quality analysis module determines the measured quality status based on the new actual quality inspection data and compares the measured quality status with the predicted quality status calculated based on environmental exposure records.

[0121] When the absolute value of the difference between the two does not exceed the preset calibration threshold, the current environmental loss parameter is retained; When the absolute value of the difference between the two exceeds the preset calibration threshold, the quality analysis module determines the original stage quality loss corresponding to each environmental exposure time period between the most recent quality inspection time and the current quality inspection time, and calculates the proportion of each original stage quality loss to the cumulative stage quality loss in that time interval; according to the proportion, the difference between the predicted quality status and the measured quality status is allocated to each environmental exposure time period, and the unit time base loss value of the corresponding environmental exposure level is corrected according to the allocation result.

[0122] In this embodiment, the preset calibration threshold is set to 5 quality status units.

[0123] Through the above processing, the actual quality test results can be used to correct the environmental loss parameters used in subsequent batches of the same agricultural product, avoiding the long-term reliance of all quality analysis results on fixed preset parameters.

[0124] After determining the quality status value of the current batch, the quality analysis module further determines the remaining shelf life.

[0125] In this embodiment of the invention, the remaining shelf life refers to the estimated time required for the batch quality status value to drop to the minimum marketable quality threshold, assuming the corresponding agricultural product continues to be stored in a preset baseline storage environment, starting from the current calculation time.

[0126] The database server pre-stores the minimum marketable quality threshold and baseline environmental degradation parameters for different types of agricultural products.

[0127] The quality analysis module determines the basic remaining shelf life based on the difference between the current batch quality status value and the minimum marketable quality threshold, as well as the baseline environmental degradation parameters of the corresponding agricultural products.

[0128] When an agricultural product batch has an irreversible damage mark, the quality analysis module further calls the remaining shelf life reduction parameter corresponding to the irreversible damage mark to reduce the basic remaining shelf life.

[0129] For example, if the current batch quality status value of a certain agricultural product is 78, the corresponding minimum marketable quality threshold is 60, and the basic remaining shelf life is determined to be 72 hours based on the baseline environmental loss parameters.

[0130] When an agricultural product batch has irreversible damage markings and the corresponding remaining shelf life reduction parameter is 0.8, the final remaining shelf life is determined to be 57.6 hours.

[0131] The quality analysis module outputs the batch quality status value and remaining shelf life to the order matching module.

[0132] Meanwhile, the batch quality status value and the stage quality loss corresponding to each environmental exposure time period are also stored in the database server, and serve as the data basis for the anomaly localization module to determine the quality status changes of different supply chain nodes.

[0133] It should be noted that this invention does not directly calculate the remaining shelf life based on the harvest date or the current temperature. Instead, it first determines the quality loss that agricultural products have already suffered based on multiple historical environmental exposure periods, and then determines the quality status value of the current batch by combining the irreversible damage status.

[0134] Therefore, when agricultural products are restored to low temperatures after a cold chain interruption, the quality damage that has already occurred still participates in the calculation of the remaining shelf life, avoiding an overestimation of the remaining shelf life based solely on the current environmental conditions.

[0135] The order matching module is used to receive target orders, determine candidate fulfillment batches and the effective inventory quantity corresponding to each candidate fulfillment batch based on the fulfillment constraints corresponding to the target orders, as well as the batch quality status value and remaining shelf life corresponding to different agricultural product batches, determine the target fulfillment batch, and generate a fulfillment execution signal.

[0136] When any tenant terminal submits an agricultural product order, the order matching module receives the target order data.

[0137] The target order data includes at least the type of agricultural product, the quantity required, the target delivery location, the latest delivery time, and the minimum delivery quality requirements.

[0138] In this embodiment of the invention, the performance constraint refers to a set of data determined based on the type of agricultural product in the target order, the latest delivery time, the target delivery location, and the minimum delivery quality requirements, used to determine whether a batch of agricultural products has the ability to complete the corresponding target order.

[0139] The order matching module first queries the database server for all agricultural product batches that have not yet been shipped and have a book inventory greater than zero, based on the type of agricultural product corresponding to the target order, to form an initial batch set.

[0140] In this embodiment of the invention, the book inventory refers to the current undelivered quantity determined based on the initial inbound quantity, the quantity of each outbound batch of agricultural products, the quantity of reported losses, and the locked quantity.

[0141] For each agricultural product batch in the initial batch set, the order matching module obtains the current supply chain node location, book inventory, batch quality status value, remaining shelf life, and irreversible damage status.

[0142] Subsequently, the order matching module calls the preset transportation timeliness data table to determine the estimated transportation time based on the current supply chain node location of the agricultural product batch and the target order delivery location.

[0143] The preset transportation timeliness data table includes at least the starting area, the target area, the transportation mode, and historical transportation time data.

[0144] The order matching module reads the most recent preset number of completed shipment records under the same starting area, target area, and shipping method. After removing the maximum and minimum shipping times, it determines the estimated shipping time based on the remaining historical shipping times.

[0145] In this embodiment, the 20 most recent completed transport records are read.

[0146] If there are fewer than 5 valid historical transport records, the standard transport time between the corresponding regions is retrieved from the database server.

[0147] The order matching module further adds sorting time, loading waiting time, and delivery buffer time to the estimated transportation time to obtain the estimated fulfillment cycle.

[0148] The first round of batch screening was then carried out.

[0149] If the remaining shelf life of an agricultural product batch is not greater than the expected fulfillment period, the agricultural product batch is excluded from the initial batch set.

[0150] For the remaining agricultural product batches, the order matching module determines the expected delivery quality status value of the agricultural product batch at the expected delivery time based on the current batch quality status value, the expected fulfillment period, and the environmental loss parameters corresponding to the expected fulfillment period.

[0151] When the expected delivery quality status value is lower than the minimum delivery quality requirement corresponding to the target order, the corresponding agricultural product batch will be excluded; When the expected delivery quality status value meets the minimum delivery quality requirements corresponding to the target order, the corresponding agricultural product batch will be identified as a candidate fulfillment batch.

[0152] In this embodiment of the invention, the candidate fulfillment batch refers to an agricultural product batch with a remaining shelf life greater than the expected fulfillment period and an expected delivery quality status value that meets the minimum delivery quality requirements of the target order.

[0153] Subsequently, the order matching module further determines the effective inventory level of each candidate fulfillment batch for the current target order.

[0154] In this embodiment of the invention, the effective inventory refers to the amount of inventory that a candidate fulfillment batch can actually be used for fulfilling the current target order, provided that the expected fulfillment cycle and minimum delivery quality requirements of the target order are met.

[0155] The effective inventory level is different from the book inventory level.

[0156] The same batch of agricultural products has a fixed book inventory, but for different target orders, different effective inventory can be obtained due to differences in delivery distance, expected fulfillment period or minimum delivery quality requirements.

[0157] Specifically, the order matching module first determines the basic availability ratio corresponding to the candidate fulfillment batches.

[0158] When the corresponding candidate fulfillment batch has quality inspection data at the current moment, the basic usable ratio is determined based on the ratio between the number of samples that reach the minimum marketable quality threshold in the actual inspection samples and the total number of inspection samples.

[0159] When the corresponding candidate fulfillment batch does not have quality inspection data at the current time, the preset quality status-availability ratio mapping table is queried based on the batch quality status value of the candidate fulfillment batch to determine the corresponding basic availability ratio.

[0160] The preset quality status-availability ratio mapping table is established based on the correspondence between the batch quality status values ​​of historical batches of similar agricultural products and the corresponding actual outbound qualification rates.

[0161] Subsequently, the order matching module adjusts the fulfillment time of the basic available ratio based on the relationship between the expected fulfillment period and the remaining shelf life; when there is an irreversible damage mark on the candidate fulfillment batch, the adjusted available ratio is further adjusted according to the corresponding irreversible damage correction parameter to obtain the final available ratio.

[0162] The order matching module calculates the book inventory of the candidate fulfillment batch with the final available ratio to obtain the effective inventory of the candidate fulfillment batch for the current target order.

[0163] For example, in one specific implementation, when quality inspection data at the current moment is missing, if the batch quality status value is not lower than 90, the corresponding basic usable ratio is 1; if the batch quality status value is not lower than 80 and lower than 90, the corresponding basic usable ratio is 0.95; if the batch quality status value is not lower than 70 and lower than 80, the corresponding basic usable ratio is 0.85.

[0164] For example, a certain batch of strawberries has a book inventory of 1000kg.

[0165] For the first target order that is relatively close in distance and has a short expected fulfillment period, the effective inventory level is 900 kg; For second target orders with long transportation distances and expected fulfillment periods, the effective inventory level may be reduced to 700 kg.

[0166] Therefore, this invention does not set a fixed effective inventory for each batch of agricultural products, but dynamically determines the effective inventory based on the specific fulfillment conditions of the target order.

[0167] Furthermore, the order matching module sorts multiple candidate fulfillment batches according to their expected delivery quality status value, remaining shelf life, and expected transportation time.

[0168] First, determine whether the expected delivery quality status value meets the sum of the minimum delivery quality requirements of the target order and the preset quality safety margin.

[0169] For candidate fulfillment batches that meet the aforementioned conditions, they are sorted from shortest to longest remaining shelf life.

[0170] In this embodiment, the preset quality safety margin is 5 quality status units.

[0171] For example, if the minimum delivery quality requirement for the target order is 70, then priority will be given to selecting from candidate fulfillment batches with an expected delivery quality status value of 75 or higher.

[0172] When the difference in remaining shelf life between two candidate fulfillment batches is less than a preset time difference threshold, their estimated transportation times are further compared, and the candidate fulfillment batch with the shorter estimated transportation time is selected first.

[0173] In this embodiment, the preset time difference threshold is set to 6 hours.

[0174] The order matching module accumulates the effective inventory corresponding to each candidate fulfillment batch according to the above sorting results.

[0175] When the cumulative effective inventory reaches the target order demand quantity, one or more candidate fulfillment batches that participated in the accumulation will be determined as the target fulfillment batch.

[0176] When the effective inventory corresponding to the last target fulfillment batch is greater than the quantity of unmet order demand, only the inventory quantity equal to the quantity of unmet order demand is locked.

[0177] Subsequently, the order matching module establishes the association between the target order and each target fulfillment batch, and writes the corresponding locked quantity into the corresponding agricultural product batch lifecycle record.

[0178] Once the target fulfillment batch is determined, the order matching module generates a fulfillment execution signal.

[0179] In this embodiment of the invention, the fulfillment execution signal refers to a computer control signal generated based on the matching result between the target order and the target fulfillment batch, and sent to the corresponding warehousing terminal, transportation scheduling terminal or sorting terminal to trigger the outbound, sorting, loading or transportation operations of agricultural products.

[0180] The performance execution signal includes at least the target order identifier, the target performance batch identifier, the locked quantity, the target delivery location, the latest departure time, and the estimated transportation time.

[0181] After receiving the fulfillment execution signal, the warehouse terminal outputs the corresponding target fulfillment batch identifier and locked quantity on the display terminal.

[0182] When automated sorting equipment is set up at the storage node, the storage terminal sends the storage location code and locked quantity corresponding to the target fulfillment batch to the automated sorting equipment, which then performs the corresponding agricultural product picking operation.

[0183] When the sum of the effective inventory corresponding to all candidate fulfillment batches is less than the target order demand quantity, the order matching module does not generate a normal fulfillment execution signal, but instead generates an insufficient inventory signal.

[0184] The inventory shortage signal includes at least the target order identifier, the target order demand quantity, the cumulative effective inventory, and the difference between the two.

[0185] When the book inventory meets the target order demand, but the effective inventory is insufficient due to the remaining shelf life or the expected delivery quality status value not meeting the fulfillment constraints, the order matching module further generates a quality constraint insufficiency signal.

[0186] The quality constraint deficiency signal includes at least the excluded batch identifier, the corresponding book inventory quantity, and the exclusion reason code.

[0187] The exclusion reason codes include insufficient remaining shelf life and insufficient expected delivery quality.

[0188] It should be noted that this invention does not directly allocate orders based on book inventory levels, nor does it simply adopt first-in-first-out or first-to-expires-first-out rules.

[0189] This invention calculates effective inventory by taking into account the current quality status of agricultural product batches, remaining shelf life, expected fulfillment period, and irreversible damage status, while ensuring that the expected delivery quality meets the requirements of the target order.

[0190] Therefore, the same batch of agricultural products generates different effective inventory levels for different orders, enabling the actual fulfillment capacity of agricultural products to dynamically change with delivery distance, fulfillment time, and quality requirements.

[0191] The anomaly location module is used to determine candidate nodes for anomalies based on the quality status changes of agricultural product batches at different supply chain nodes, and to determine the anomaly node and anomaly type based on the environmental exposure records within the candidate nodes for anomalies, thereby generating an anomaly location signal.

[0192] When the quality inspection terminal obtains new quality inspection data for a batch of agricultural products, the anomaly location module determines the current measured quality status value based on the new quality inspection data.

[0193] Subsequently, the anomaly localization module reads the measured quality status value corresponding to the most recent upstream quality inspection node of the agricultural product batch and obtains all supply chain nodes passed between the two quality inspection times.

[0194] For each node in the supply chain, the anomaly location module establishes a corresponding node status record.

[0195] The node status record includes at least the node identifier, node entry time, node exit time, quality status value upon entering the node, quality status value upon leaving the node, node dwell time, highest temperature within the node, cumulative duration exceeding the suitable temperature range, irreversible damage marker, and transport disturbance marker.

[0196] When a supply chain node has actual entry quality inspection data and actual exit quality inspection data, the entry quality status value and exit quality status value are determined directly based on the actual quality inspection data.

[0197] When a supply chain node does not undergo direct quality inspection, the anomaly location module calls the stage quality loss amount determined by the quality analysis module. Based on the quality status value when the node enters and the stage quality loss amount generated during the corresponding environmental exposure period of the node, the expected departure quality status value is determined.

[0198] The anomaly localization module determines the abrupt change in the node's state based on the entry and exit quality status values ​​of a supply chain node.

[0199] In this embodiment of the invention, the node state mutation amount refers to the difference between the quality state value of the same agricultural product batch when it enters a supply chain node and the quality state value when it leaves the supply chain node, which is used to indicate the degree of quality change of agricultural products within the corresponding supply chain node.

[0200] The database server pre-stores normal change thresholds for different types of agricultural products and different supply chain node types.

[0201] The anomaly localization module compares the node state mutation amount corresponding to each supply chain node with the corresponding preset normal change threshold.

[0202] If the number of node state mutations corresponding to a supply chain node does not exceed the corresponding preset normal change threshold, it will not be identified as an abnormal candidate node.

[0203] When the number of sudden changes in the state of a supply chain node exceeds the corresponding preset normal change threshold, the supply chain node is identified as an abnormal candidate node.

[0204] In this embodiment of the invention, the abnormal candidate node refers to a supply chain node whose state mutation amount exceeds the preset normal change threshold for the corresponding node type, and which needs to be further determined based on the node's internal environmental data to determine whether it constitutes the main source of the anomaly.

[0205] For each candidate node for anomaly, the anomaly localization module further calls the environmental exposure records generated during the corresponding node period.

[0206] The anomaly location module obtains the cumulative duration of exceeding the suitable temperature range within the anomaly candidate node, the difference between the highest temperature within the anomaly candidate node and the corresponding agricultural product temperature damage threshold, the cumulative duration of exceeding the suitable humidity range, the irreversible damage status, the transportation disturbance status, and the difference between the actual dwell time and the preset standard dwell time.

[0207] Subsequently, the abnormal contribution value corresponding to the abnormal candidate node was determined based on the above data.

[0208] In this embodiment of the invention, the abnormal contribution value refers to a data value determined based on the duration of temperature anomalies, the degree of temperature exceeding limits, the duration of humidity anomalies, the state of irreversible damage, the state of transportation disturbance, and the degree of dwell time within the abnormal candidate nodes, and is used to compare the degree of influence of different abnormal candidate nodes on the final quality anomalies.

[0209] Specifically, the database server pre-stores the upper limit of the duration of temperature anomalies, the upper limit of the degree of temperature exceeding the limit, the upper limit of the duration of humidity anomalies, and the upper limit of the dwell timeout, corresponding to the types of agricultural products and the types of supply chain nodes.

[0210] The anomaly localization module calculates the ratio between the actual temperature anomaly duration and the upper limit of the temperature anomaly duration, the ratio between the actual temperature exceedance degree and the upper limit of the temperature exceedance degree, the ratio between the actual humidity anomaly duration and the upper limit of the humidity anomaly duration, and the ratio between the actual dwell timeout duration and the upper limit of the dwell timeout duration. Ratios greater than 1 are set to 1, thereby obtaining the normalized temperature anomaly duration, normalized temperature exceedance degree, normalized humidity anomaly duration, and normalized dwell timeout degree within the range of 0 to 1.

[0211] Subsequently, the data were weighted according to the following weights: duration of temperature anomaly, degree of temperature exceeding the limit, duration of humidity anomaly, and degree of dwell time exceeding the limit, respectively, with weights of 0.35, 0.30, 0.15, and 0.20.

[0212] When an abnormal candidate node has an irreversible damage marker, a first fixed increment is added to the above weighted result.

[0213] When an abnormal candidate node has a transport disturbance flag, a second fixed increment is added to the above weighted result.

[0214] In this embodiment, the first fixed increment is set to 0.3, and the second fixed increment is set to 0.15.

[0215] The anomaly localization module determines the anomaly contribution value corresponding to the anomaly candidate node based on the above processing results.

[0216] When the abnormal contribution value of an abnormal candidate node exceeds the preset abnormal confirmation threshold, the abnormal candidate node is identified as an abnormal node.

[0217] In this embodiment, the preset anomaly confirmation threshold is set to 0.6.

[0218] When there are multiple abnormal nodes, the anomaly localization module arranges them from high to low according to the anomaly contribution value corresponding to each abnormal node, forming an abnormal node sequence.

[0219] Among them, the abnormal node with the highest abnormal contribution value is identified as the primary abnormal node.

[0220] Subsequently, the anomaly localization module determines the anomaly type based on the environmental exposure records corresponding to the primary anomaly node.

[0221] The anomaly types include temperature anomalies, humidity anomalies, transportation disturbance anomalies, and stay timeout anomalies.

[0222] When an abnormal node simultaneously meets the judgment conditions corresponding to multiple abnormal types, multiple abnormal types are recorded at the same time.

[0223] The anomaly localization module generates anomaly localization signals based on the identified anomaly nodes and anomaly types.

[0224] In this embodiment of the invention, the abnormal location signal refers to a computer data signal generated based on changes in the quality status of agricultural product batches and abnormal data of the internal environment of the corresponding supply chain nodes, used to represent the supply chain nodes that cause abnormal agricultural product quality and the corresponding abnormality type.

[0225] The anomaly location signal includes at least the agricultural product batch identifier, the anomaly node identifier, the time interval of the anomaly occurrence, the anomaly type, and the anomaly contribution value.

[0226] The cloud server sends the anomaly location signal to the corresponding tenant terminal and supply chain management terminal through the communication interface. The corresponding terminal displays the agricultural product batch, the anomaly node, the time interval of the anomaly, and the anomaly type.

[0227] Furthermore, when the primary abnormal node is a storage node, the abnormal location module reads other agricultural product batches stored by that storage node within the time interval of the abnormality.

[0228] When other agricultural product batches overlap with the current abnormal agricultural product batch within the time interval of the abnormality, the abnormality location module generates a related risk signal.

[0229] In this embodiment of the invention, the associated risk signal refers to a computer data signal generated based on the overlapping residence time of different agricultural product batches within the storage node after an environmental anomaly is determined to have occurred at a certain storage node. This signal is used to indicate that other agricultural product batches have a common environmental exposure risk.

[0230] The associated risk signals include at least the warehouse node identifier, the time interval of the anomaly occurrence, and the identifiers of other agricultural product batches with overlapping time periods.

[0231] The cloud server will send the associated risk signals to the corresponding warehouse terminal to trigger advance quality testing of the relevant agricultural product batches.

[0232] It should be noted that this invention does not simply retrieve historical logistics records for manual review after the agricultural products have been found to have spoiled. Instead, it first filters out abnormal candidate nodes based on the amount of node state mutations corresponding to different supply chain nodes, and then determines the abnormal contribution value based on the environmental exposure data inside the abnormal candidate nodes.

[0233] Therefore, the anomaly location results are not simply a reverse lookup based on the final quality results, but rather form a data association with the aforementioned batch lifecycle records, environmental exposure record tables, and batch quality status values.

[0234] Especially after identifying a storage node as an anomaly, further queries can be made to identify other agricultural product batches with common environmental exposure experiences within the same anomaly time interval. This allows the anomaly location results to directly trigger advance quality inspections of other potentially risky batches.

[0235] In this embodiment of the invention, the batch recording module, environmental exposure processing module, quality analysis module, order matching module, and anomaly location module achieve data interaction through the internal data bus of the cloud server or the database server.

[0236] The batch record module outputs the batch lifecycle record of agricultural products. The environmental exposure processing module reads the batch lifecycle records of the agricultural products, divides the environmental exposure time periods according to changes in supply chain nodes and environmental parameters, and outputs an environmental exposure record table. The quality analysis module reads the environmental exposure record table and quality test data to determine the stage quality loss, batch quality status value and remaining shelf life. The order matching module reads the batch quality status value, remaining shelf life, and target order data to determine the candidate fulfillment batch, effective inventory, and target fulfillment batch, and outputs the fulfillment execution signal. The anomaly localization module reads the agricultural product batch life cycle record, environmental exposure record table, stage quality loss amount and batch quality status value, determines the node status mutation amount, anomaly candidate node, anomaly contribution value and anomaly node, and outputs anomaly localization signal.

[0237] This creates the following continuous data link: The original data collected from each agricultural product batch is written into the agricultural product batch lifecycle record. The agricultural product batch lifecycle records were converted into multiple environmental exposure records with clearly defined time boundaries; Multiple environmental exposure records were further converted into stage quality loss; Multiple stages of quality loss are used to determine batch quality status values ​​and remaining shelf life; The batch quality status value, remaining shelf life, and fulfillment constraints corresponding to the target order are used together to determine candidate fulfillment batches and effective inventory levels. Effective inventory levels are further used to determine target fulfillment batches and generate fulfillment execution signals; Once new quality testing data is generated, existing environmental exposure records and quality status change data are further used to screen for abnormal candidate nodes and identify abnormal nodes.

[0238] It should be noted that the module names mentioned above are used to distinguish different data processing functions and do not mean that each module must be implemented by a separate physical device.

[0239] In one embodiment, each module is implemented by a processor in the same cloud server executing a corresponding computer program. In another embodiment, the batch recording module and the environmental exposure processing module are deployed on the first cloud server, while the quality analysis module, the order matching module, and the anomaly location module are deployed on the second cloud server, and data is transmitted through a database server or a network interface.

[0240] The above implementation methods do not affect the data processing logic between the modules of this invention.

[0241] In this embodiment of the invention, the database server pre-stores suitable temperature ranges, suitable humidity ranges, temperature warning thresholds, temperature damage thresholds, duration thresholds, basic loss values ​​per unit time, minimum marketable quality thresholds, preset normal change thresholds, and transportation vibration thresholds for different types of agricultural products.

[0242] The above parameters are preset during the initial system configuration based on historical storage experiment data, historical transportation data, and quality inspection data for the corresponding agricultural product types.

[0243] Once the system obtains new measured quality inspection data, the quality analysis module corrects the environmental loss parameters for the subsequent use of the corresponding agricultural product batch based on the difference between the predicted quality status and the measured quality status.

[0244] It should be noted that the 30 seconds, 2℃, 3℃, 15 percentage points, the most recent 20 transport records, 5 quality status units, 6 hours, and the anomaly confirmation threshold of 0.6 involved in this embodiment are all used to illustrate a specific executable implementation method and do not constitute a limitation on the scope of protection of this invention.

[0245] For different types of agricultural products, different testing accuracies, and different supply chain environments, corresponding thresholds can be set according to the actual quality change patterns of the corresponding agricultural products.

[0246] Those skilled in the art will understand that all or part of the functions of the above modules can be implemented by computer program instructions and related hardware. The program is stored in a computer-readable storage medium. When the processor executes the program, it completes the data processing process corresponding to the batch recording module, environmental exposure processing module, quality analysis module, order matching module, and anomaly location module.

[0247] The computer-readable storage media include read-only memory, random access memory, solid-state drive, hard disk, and other non-transitory storage media capable of storing computer programs.

[0248] The foregoing has provided a detailed description of an agricultural product supply chain management system based on a SaaS architecture, as provided in the embodiments of the present invention. Specific embodiments have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the embodiments above are intended to help understand the technical solutions and core ideas of the present invention. For those skilled in the art, any adjustments to the specific implementation methods and application scenarios without departing from the technical concept of the present invention should fall within the protection scope of the present invention.

Claims

1. A SaaS-based agricultural product supply chain management system, characterized in that, The system is deployed on a cloud server, and different supply chain entities access the system through corresponding tenant accounts. The system includes: a batch recording module for establishing batch lifecycle records of agricultural products; an environmental exposure processing module for dividing environmental exposure time periods and determining environmental exposure status based on the batch lifecycle records of agricultural products; a quality analysis module for determining batch quality status values ​​and remaining shelf life based on the environmental exposure status; an order matching module for determining candidate fulfillment batches and effective inventory based on the batch quality status values, remaining shelf life, and target order data, and generating fulfillment execution signals; and an anomaly location module for identifying abnormal nodes based on quality status changes and environmental exposure records between supply chain nodes, and generating anomaly location signals.

2. The agricultural product supply chain management system based on a SaaS architecture according to claim 1, characterized in that, The batch recording module writes the node entry events, node departure events, environmental sampling data, quality inspection data, quantity change data, and order association data generated by the same agricultural product batch at different supply chain nodes into the agricultural product batch lifecycle record in chronological order, based on the agricultural product batch identifier.

3. The agricultural product supply chain management system based on a SaaS architecture according to claim 2, characterized in that, The environmental exposure processing module divides the environmental exposure time period according to the changes in supply chain nodes and the changes between adjacent environmental sampling data, and determines the environmental exposure level according to the relationship between the environmental parameters corresponding to each environmental exposure time period and the preset environmental parameters. When environmental parameters exceed the damage threshold of the corresponding agricultural product and continue for a preset duration, an irreversible damage marker is set for the corresponding environmental exposure period.

4. The agricultural product supply chain management system based on a SaaS architecture according to claim 3, characterized in that, The quality analysis module determines the stage quality loss based on the environmental exposure level, duration, humidity, transportation disturbance markers, and irreversible damage markers corresponding to each environmental exposure time period. It also determines the batch quality status value based on the initial quality status of the agricultural product batch and the quality loss at each stage, according to the time sequence of each environmental exposure time period.

5. The agricultural product supply chain management system based on a SaaS architecture according to claim 4, characterized in that, The quality analysis module determines the basic remaining shelf life based on the difference between the batch quality status value and the minimum marketable quality threshold and the preset benchmark environmental loss parameters. When the agricultural product batch has irreversible damage marks, the basic remaining shelf life is reduced according to the corresponding remaining life reduction parameters to obtain the remaining shelf life.

6. The agricultural product supply chain management system based on a SaaS architecture according to claim 1, characterized in that, The order matching module determines the fulfillment constraints based on the type of agricultural product corresponding to the target order, the target delivery location, the latest delivery time, and the minimum delivery quality requirements. It also determines the estimated fulfillment period based on the transportation time between the current supply chain node location of the agricultural product batch and the target delivery location, and excludes agricultural product batches with a remaining shelf life not exceeding the estimated fulfillment period.

7. The agricultural product supply chain management system based on a SaaS architecture according to claim 6, characterized in that, The order matching module determines the expected delivery quality status value based on the batch quality status value of the remaining agricultural product batches, the expected fulfillment period, and the environmental loss parameters corresponding to the expected fulfillment period. The agricultural product batches whose expected delivery quality status value meets the minimum delivery quality requirements of the target order are identified as candidate fulfillment batches.

8. The agricultural product supply chain management system based on a SaaS architecture according to claim 7, characterized in that, The order matching module determines the effective inventory based on the book inventory corresponding to the candidate fulfillment batch, the batch quality status value, the relationship between the expected fulfillment cycle and the remaining shelf life, and the irreversible damage status. The candidate fulfillment batches are sorted according to the expected delivery quality status value, the remaining shelf life, and the expected transportation time, and the corresponding effective inventory is accumulated in sequence until the accumulated effective inventory reaches the target order demand quantity. The target fulfillment batch is then determined and the fulfillment execution signal is generated.

9. The agricultural product supply chain management system based on a SaaS architecture according to claim 1, characterized in that, The anomaly localization module determines the node state mutation amount based on the entry quality state value and exit quality state value corresponding to the agricultural product batch when it passes through different supply chain nodes, and identifies supply chain nodes whose node state mutation amount exceeds the preset normal change threshold of the corresponding node type as anomaly candidate nodes.

10. The agricultural product supply chain management system based on a SaaS architecture according to claim 9, characterized in that, The anomaly location module determines the anomaly contribution value based on the duration of temperature anomaly, the degree of temperature exceeding the limit, the duration of humidity anomaly, the state of irreversible damage, the state of transportation disturbance, and the degree of dwell time corresponding to the anomaly candidate node. The anomaly candidate node whose anomaly contribution value exceeds the preset anomaly confirmation threshold is identified as an anomaly node. The anomaly type is determined based on the environmental exposure record corresponding to the anomaly node, and the anomaly location signal is generated.

Images