Supply chain distribution optimization system based on rule engine

Abstract

The invention discloses a supply chain distribution optimization system based on a rule engine, and relates to the technical field of reverse logistics, and the system can evaluate the quality condition of returned commodities in time through the combination of a quality inspection data collection module and a rule engine module, makes a quick decision, and improves the efficiency. Therefore, the problem of low efficiency caused by untimely processing in a traditional system is avoided. The system solves the problem of unreasonable resource allocation through a dynamic adjustment mechanism. And a pressure evaluation unit and an efficiency evaluation unit in the reverse logistics optimization module can automatically adjust a resource allocation strategy according to real-time data. For example, when the system detects that the reverse logistics pressure exceeds a normal range, a multi-stage recovery network is automatically designed, and a processing path is optimized; when the reverse logistics efficiency is not qualified, processing equipment can be added, and the overall processing capacity is improved. The dynamic adjustment mechanism not only solves the problem of unbalanced resource allocation in a traditional system, but also effectively improves the overall processing efficiency of reverse logistics.

Description

Technical Field

[0001] The present invention relates to the technical field of reverse logistics, and in particular to a supply chain distribution optimization system based on a rule engine. Background Art

[0002] Reverse logistics refers to the process of returning goods or materials from consumers or end users to the front end of the supply chain (such as manufacturers, distributors or retailers). This process is the opposite of traditional "forward logistics" and aims to handle and manage returned goods, waste products or packaging materials.

[0003] Traditional reverse logistics systems often lack real-time data monitoring and analysis capabilities, making the return processing process often inefficient. Delays can occur in the receipt, inspection, processing, and redistribution of returned goods, leading to long reverse logistics cycles and erratic processing times. This inefficiency not only impacts the overall operation of the supply chain but also increases inventory and management costs. Reverse logistics resource allocation is often based on fixed standards and experience, lacking dynamic adjustment mechanisms. This can lead to inefficient resource allocation during high return volumes or high-pressure situations, resulting in resource constraints in some links and idle resources in others. This imbalance limits overall processing capacity and further reduces reverse logistics efficiency. Summary of the Invention

[0004] In view of the deficiencies of the existing technology, the present invention provides a supply chain distribution optimization system based on a rule engine to solve the problems mentioned in the background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a supply chain distribution optimization system based on a rule engine, including a quality inspection data acquisition module, a rule engine module, a reverse logistics monitoring module, a reverse logistics analysis module and a reverse logistics optimization module;

[0006] The quality inspection data collection module is used to collect quality parameters of reverse logistics return orders in the supply chain and generate quality inspection data groups;

[0007] The rule engine module is used to establish and train a quality analysis model, analyze and calculate the quality inspection data set to construct a quality inspection index Zjzl. If the quality inspection index Zjzl is higher than or equal to the quality threshold X, the product is returned to the original supply chain manufacturer. If the quality inspection index Zjzl is lower than the quality threshold X, the product is deemed to be recycled. The rule engine module also dynamically loads and applies preset reverse logistics optimization rules.

[0008] The reverse logistics monitoring module is used to monitor and record status data information related to the reverse logistics process in real time and generate a reverse logistics monitoring data group;

[0009] The reverse logistics analysis module is used to extract features from the reverse logistics-related data information in the reverse logistics monitoring data group to obtain a return volume factor Tqty, a recycling frequency factor Rfreq, a logistics cost factor Lcost, and a processing time factor Ptime, and to obtain a reverse logistics pressure coefficient Rpr by associating the return volume factor Tqty with the recycling frequency factor Rfreq, and to obtain a logistics efficiency coefficient Lef by associating the logistics cost factor Lcost with the processing time factor Ptime;

[0010] The reverse logistics optimization module is used to pre-set the pressure threshold V and the efficiency threshold W, and compare and analyze the logistics pressure coefficient Rpr with the pressure threshold V to obtain the logistics pressure evaluation result, and compare and analyze the logistics efficiency coefficient Lef with the efficiency threshold W to obtain the logistics efficiency result. According to the logistics pressure evaluation result and the logistics efficiency result, the reverse logistics processing flow and resource allocation strategy are automatically adjusted and controlled.

[0011] Preferably, the quality inspection data acquisition module includes a defect monitoring unit and a quality parameter recording unit;

[0012] The defect monitoring unit is used to collect defect information of returned products in real time using testing tools and image recognition technology;

[0013] The quality parameter recording unit is used to record parameter information related to the quality of the returned product in real time, including product category, production batch, quality inspection date and testing method;

[0014] The quality inspection data group includes: the quality score Q of the returned product score , Check the actual product size deviation value A when returning defect , Detect the actual product defect area D when returning dev , Check the actual product performance test value P when returning test And the actual product packaging defect area Xn detected when returning test .

[0015] Preferably, the rule engine module includes a model building unit and a returned product quality analysis unit;

[0016] The model building unit is used to build a quality analysis model through machine learning technology, including linear regression and support vector machine, and divide the quality inspection data group into a training set and a test set for training and testing, and then apply the training and testing to the quality analysis model;

[0017] The returned product quality analysis unit is used to perform data cleaning, missing value processing, and dimensionless processing on the quality inspection data group through the quality analysis model, and then calculate the quality inspection index Zjzl using the following formula:

[0018]

[0019] Where Q std represents the quality standard threshold, Q score Indicates the quality score of the returned product, Q weright The weight of the returned product quality score, A tol Indicates the allowable size deviation threshold, A defect Indicates the actual product size deviation value detected when returning goods, D tol Denotes the allowable defect area threshold, D dev Indicates the actual product defect area detected during return, P tol Indicates the performance tolerance threshold, P test Indicates the actual product performance test value when returning the product, Xn tol Indicates the allowable packaging defect area threshold, Xn test Indicates the actual product packaging defect area detected during return.

[0020] Preferably, the rule engine module further includes a first evaluation unit and a first strategy unit;

[0021] The first evaluation unit is configured to compare and analyze the quality inspection index Zjzl of each return order with the quality threshold X to obtain a first evaluation result, including:

[0022] If the quality inspection index Zjzl ≥ the quality threshold X, it means that the product of the return order meets the return criteria. The first policy unit generates a return to original manufacturer policy, which includes returning the product to the original manufacturer for further processing, repair, or rework.

[0023] If the quality inspection index Zjzl is less than the quality threshold X, it means that the product of the return order does not meet the return standard. The first policy unit generates a recycling policy to send the product back to the recycling point or processing center for disassembly, repair or destruction.

[0024] Preferably, the return quantity factor Tqty is calculated using the following formula:

[0025]

[0026] Where R thl Indicates the return quantity of a single order, S sl represents the sales volume of a single order, D return Indicates the historical return quantity of the product category to which the order belongs, D total Indicates the total sales quantity of the product category to which the order belongs.

[0027] Preferably, the recovery frequency factor Rfreq is calculated by the following formula:

[0028]

[0029] Where R thl represents the order recovery interval, σ season represents the standard deviation of the return volume fluctuation of this product category in different seasons, μ season represents the average return volume of the product category in different seasons, W total Indicates the total return weight of the order, W item Indicates the weight of a single product in this order.

[0030] Preferably, the logistics cost factor Lcost is calculated by the following formula:

[0031]

[0032] Where C trans represents the shipping cost of a single order, C storage represents the storage cost of a single order, C handling represents the processing cost of a single order, R thl represents the return quantity of a single order, C fuel is the combustion cost during transportation, D distance is the transportation distance of the order, C total is the total logistics cost of a single order.

[0033] Preferably, the processing time factor Ptime is calculated by the following formula:

[0034]

[0035] Where, T processing Indicates the return processing time of a single order, T inspection Indicates the quality inspection time of a single order, T repackaging represents the repackaging time of a single order, R thl represents the return quantity of a single order, T delay Indicates the delay time in the return process, T optimal Indicates the ideal time threshold for return processing.

[0036] Preferably, after the return quantity factor Tqty and the recycling frequency factor Rfreq are dimensionlessly processed, the reverse logistics pressure coefficient Rpr is calculated using the following correlation formula:

[0037] Rpr=Mlxl*b1+Ljqhz*b2+B1

[0038] Wherein, b1 and b2 represent the preset proportional coefficients of the return quantity factor Tqty and the recycling frequency factor Rfreq respectively; B1 is the first correction constant;

[0039] After dimensionless processing of the logistics cost factor Lcost and the processing time factor Ptime, the logistics efficiency coefficient Lef is calculated using the following correlation formula:

[0040] Lef=Lcost*b3+Ptime*b4+B2

[0041] In the formula, b3 and b4 represent the preset proportional coefficients of the logistics cost factor Lcost and the processing time factor Ptime respectively; B2 is the second correction constant.

[0042] Preferably, the reverse logistics optimization module includes a pressure evaluation unit and an efficiency evaluation unit;

[0043] The pressure assessment unit is used to pre-set a pressure threshold V and compare and analyze the logistics pressure coefficient Rpr with the pressure threshold V to determine whether the current reverse logistics state is normal or not, so as to obtain a logistics pressure assessment result, including:

[0044] When the logistics pressure coefficient Rpr ≤ the pressure threshold V, it indicates that the current reverse logistics pressure is within the normal range, and the resource allocation or processing flow will not be automatically adjusted, and the reverse logistics order will be processed normally;

[0045] When the logistics pressure coefficient Rpr exceeds the pressure threshold V, it indicates that the current reverse logistics pressure exceeds the normal range. The system automatically triggers optimization measures and adjusts the processing strategy, including: designing a multi-level recycling network, including a main recycling center and regional recycling points, and optimizing the path from the return point to the nearest regional recycling point and then to the main recycling center;

[0046] The efficiency evaluation unit is used to preset an efficiency threshold W and compare and analyze the logistics efficiency coefficient Lef with the efficiency threshold W to obtain a logistics efficiency result, including:

[0047] When the logistics efficiency coefficient Lef ≥ efficiency threshold W, it means that the current reverse logistics efficiency is qualified.

[0048] When the logistics efficiency coefficient Lef is less than the efficiency threshold W, it means that the current reverse logistics efficiency is unqualified. The system automatically triggers optimization measures and adjusts the processing strategy, including: increasing the return reception, sorting, inspection, repair and repackaging equipment by 10% and storing them in the warehouse after processing at the nearest regional recycling point.

[0049] The present invention provides a supply chain distribution optimization system based on a rule engine. It has the following beneficial effects:

[0050] (1) This supply chain distribution optimization system based on a rule engine, through the quality inspection data acquisition module and the rule engine module, can acquire and analyze the quality parameters of reverse logistics return orders in real time and calculate the quality inspection index Zjzl. This data-driven analysis ensures strict control of the quality of returned products and can accurately determine whether they should be returned to the original supply chain manufacturer or recycled, thereby improving the accuracy of quality control and processing efficiency. The automated rule engine generates a return to the original manufacturer strategy or a recycling strategy, reducing the errors and delays of manual judgment and improving the reliability and consistency of processing results.

[0051] (2) The supply chain distribution optimization system based on the rule engine dynamically adjusts the processing strategy through the pressure evaluation unit and the efficiency evaluation unit. When the reverse logistics pressure exceeds the standard, the system designs a multi-level recycling network and optimizes the processing path; when the efficiency is unsatisfactory, the system adds processing equipment. This optimization mechanism can effectively reduce logistics costs, improve processing speed, optimize resource allocation, and ensure that resources are used rationally. By calculating the return volume factor Tqty, the recycling frequency factor Rfreq, the logistics cost factor Lcost, and the processing time factor Ptime, the system can comprehensively evaluate the pressure and efficiency of reverse logistics, and thus formulate targeted optimization measures. This comprehensive consideration of various factors helps to improve overall logistics efficiency and reduce unnecessary expenses.

[0052] (3) The supply chain distribution optimization system based on the rule engine has real-time data monitoring and analysis functions, which enables each link in the return processing process to be accurately tracked and recorded, significantly reducing the instability of the processing cycle caused by delays. The combination of the quality inspection data acquisition module and the rule engine module can timely evaluate the quality status of returned goods and make quick decisions, thereby avoiding the inefficiency problem caused by untimely processing in traditional systems. The system solves the problem of unreasonable resource allocation through a dynamic adjustment mechanism. The pressure assessment unit and efficiency assessment unit in the reverse logistics optimization module can automatically adjust the resource allocation strategy based on real-time data. For example, when the system detects that the reverse logistics pressure exceeds the normal range, it will automatically design a multi-level recycling network and optimize the processing path; when the reverse logistics efficiency is unqualified, it will increase processing equipment to improve the overall processing capacity. This dynamic adjustment mechanism not only solves the problem of unbalanced resource allocation in traditional systems, but also effectively improves the overall processing efficiency of reverse logistics. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 This is a flowchart diagram of the supply chain distribution optimization system based on the rule engine of the present invention. DETAILED DESCRIPTION

[0054] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0055] Example 1

[0056] See also Figure 1 , the present invention provides a supply chain distribution optimization system based on a rule engine, including a quality inspection data acquisition module, a rule engine module, a reverse logistics monitoring module, a reverse logistics analysis module and a reverse logistics optimization module;

[0057] The quality inspection data collection module is used to collect quality parameters of reverse logistics return orders in the supply chain and generate quality inspection data groups;

[0058] The rule engine module is used to establish and train a quality analysis model, analyze and calculate the quality inspection data set to construct a quality inspection index Zjzl. If the quality inspection index Zjzl is higher than or equal to the quality threshold X, the product is returned to the original supply chain manufacturer. If the quality inspection index Zjzl is lower than the quality threshold X, the product is deemed to be recycled. The rule engine module also dynamically loads and applies preset reverse logistics optimization rules.

[0059] The reverse logistics monitoring module is used to monitor and record status data information related to the reverse logistics process in real time and generate a reverse logistics monitoring data group;

[0060] The reverse logistics analysis module is used to extract features from the reverse logistics-related data information in the reverse logistics monitoring data group to obtain a return volume factor Tqty, a recycling frequency factor Rfreq, a logistics cost factor Lcost, and a processing time factor Ptime, and to obtain a reverse logistics pressure coefficient Rpr by associating the return volume factor Tqty with the recycling frequency factor Rfreq, and to obtain a logistics efficiency coefficient Lef by associating the logistics cost factor Lcost with the processing time factor Ptime;

[0061] The reverse logistics optimization module is used to pre-set the pressure threshold V and the efficiency threshold W, and compare and analyze the logistics pressure coefficient Rpr with the pressure threshold V to obtain the logistics pressure evaluation result, and compare and analyze the logistics efficiency coefficient Lef with the efficiency threshold W to obtain the logistics efficiency result. According to the logistics pressure evaluation result and the logistics efficiency result, the reverse logistics processing flow and resource allocation strategy are automatically adjusted and controlled.

[0062] In this embodiment, the quality parameters of the reverse logistics return order are obtained in real time through the quality inspection data acquisition module, and the quality analysis model is established and trained using the rule engine module. The system can efficiently analyze and calculate the quality inspection data group, obtain the quality inspection index Zjzl and evaluation, and can quickly determine whether to return the goods to the original supply chain manufacturer or determine whether they need to be recycled; the system can send high-quality returned goods back to the original supply chain manufacturer and recycle low-quality goods, ensuring that quality control is more stringent and reliable.

[0063] The system optimizes reverse logistics processes and reduces unnecessary expenses by comprehensively considering factors such as return volume, recycling frequency, logistics costs, and processing time. Dynamically optimizing resource allocation and processing strategies helps reduce overall logistics costs and improve cost control.

[0064] Example 2: This example is an explanation of Example 1. Figure 1 ,Specifically, the quality inspection data acquisition module includes a defect monitoring unit and a quality parameter recording unit;

[0065] The defect monitoring unit is used to collect defect information of returned products in real time using testing tools and image recognition technology;

[0066] The quality parameter recording unit is used to record parameter information related to the quality of the returned product in real time, including product category, production batch, quality inspection date and testing method;

[0067] The quality inspection data group includes: the quality score Q of the returned product score , Check the actual product size deviation value A when returning defect , Detect the actual product defect area D when returning dev , Check the actual product performance test value P when returning test And the actual product packaging defect area Xn detected when returning test .

[0068] The quality inspection data set includes information such as the quality score of the returned product, the actual product size deviation value, the defect area, performance test values, and the packaging defect area. This detailed data can be used to comprehensively assess the quality of the returned product and provide a solid data foundation for further quality analysis and decision-making. Specifically, it includes:

[0069] Quality Score: Comprehensively reflects the overall quality level of the returned product, helping to quickly determine whether the product meets quality standards.

[0070] Dimensional deviation values: used to assess whether a product meets dimensional specifications and help identify problems in the manufacturing process.

[0071] Defect Area: Provides detailed information about product defects to help understand the severity of the defects.

[0072] Performance test value: Evaluates whether the actual performance of the product meets the standards and supports detailed analysis of product functions.

[0073] Packaging defect area: Identify packaging problems and help improve packaging design and handling processes.

[0074] In this embodiment, the defect monitoring unit uses advanced testing tools and image recognition technology to collect defect information of returned products in real time and efficiently. Through image recognition technology, the system can accurately identify and record various types of defects of the product, including appearance flaws, and then test functional problems through performance testing tools, thereby improving the accuracy and timeliness of defect detection. This real-time monitoring capability helps to reduce errors and delays in manual inspection, making defect information more reliable and comprehensive. The quality parameter recording unit is responsible for recording detailed parameter information related to the quality of returned products in real time, including product category, production batch, quality inspection date and inspection method. Such detailed records facilitate comprehensive quality analysis and traceability of returned products, ensuring that the root cause of product quality problems can be accurately located. These records also support subsequent quality management and improvement work, help identify potential problems in product quality, and thus improve production processes and quality control.

[0075] Example 3, this example is explained in Example 1, please refer to Figure 1 ,Specifically, the rule engine module includes a model building unit and a ,returned product quality analysis unit;

[0076] The model building unit is used to establish a quality analysis model through machine learning techniques, including linear regression and support vector machines, and divides the quality inspection data set into a training set and a test set for training and testing, and then applies it to the quality analysis model; the model building unit establishes a quality analysis model through machine learning techniques (such as linear regression and support vector machines). By dividing the quality inspection data set into a training set and a test set for training and testing, the accuracy and reliability of the model are ensured. This method can use a large amount of data for in-depth analysis, discover potential quality problem patterns, and improve the accuracy and scientificity of quality analysis by continuously optimizing the model.

[0077] The returned product quality analysis unit is used to perform data cleaning, missing value processing, and dimensionless processing on the quality inspection data group through the quality analysis model, and then calculate the quality inspection index Zjzl using the following formula:

[0078]

[0079] Where Q std represents the quality standard threshold, Qscore Indicates the quality score of the returned product, Q weright The weight of the returned product quality score, A tol Indicates the allowable size deviation threshold, A defect Indicates the actual product size deviation value detected when returning goods, D tol Denotes the allowable defect area threshold, D dev Indicates the actual product defect area detected during return, P tol Indicates the performance tolerance threshold, P test Indicates the actual product performance test value when returning the product, Xn tol Indicates the allowable packaging defect area threshold, Xn test Indicates the actual product packaging defect area detected during return.

[0080] The rule engine module further includes a first evaluation unit and a first strategy unit;

[0081] The first evaluation unit is configured to compare and analyze the quality inspection index Zjzl of each return order with the quality threshold X to obtain a first evaluation result, including:

[0082] If the quality inspection index Zjzl ≥ the quality threshold X, it means that the product of the return order meets the return criteria. The first policy unit generates a return to original manufacturer policy, which includes returning the product to the original manufacturer for further processing, repair, or rework.

[0083] If the quality inspection index Zjzl is less than the quality threshold X, it means that the product of the return order does not meet the return standard. The first policy unit generates a recycling policy to send the product back to the recycling point or processing center for disassembly, repair or destruction.

[0084] In this embodiment, the first evaluation unit compares and analyzes the quality inspection index Zjzl of each return order against a preset quality threshold X to generate a first evaluation result. This comparative analysis effectively distinguishes whether products meet return criteria, ensuring that only qualified products are returned to the original manufacturer for further processing. Products that do not meet the standards will be subject to a recycling strategy and sent back to a recycling point or processing center for disassembly, repair, or destruction. This dynamically adjusted strategy reduces the influx of substandard products, optimizes the return processing process, and improves resource utilization. By comparing the quality inspection index with the quality threshold, the rules engine module intelligently generates a return-to-original manufacturer strategy or a recycling strategy. This optimized decision-making process helps ensure that returned products are handled appropriately, reducing potential losses due to product quality issues. It also provides scientific, well-founded decision support for all links in the supply chain. The automated rules engine module reduces reliance on manual judgment and minimizes errors caused by human intervention. This automated processing not only improves system stability and consistency, but also increases the reliability of processing results, ensuring the efficient operation of the reverse logistics process.

[0085] Example 4: This example is an explanation of Example 1. Figure 1 Specifically, the return quantity factor Tqty is calculated using the following formula:

[0086]

[0087] Where R thl Indicates the return quantity of a single order, S sl represents the sales volume of a single order, D return Indicates the historical return quantity of the product category to which the order belongs, D total Indicates the total sales quantity of the product category to which the order belongs.

[0088] The return volume factor, Tqty, takes into account the returns and sales volume of individual orders, as well as historical return data for the product category. This comprehensive calculation objectively reflects product return trends, avoiding the limitations of relying on a single data point and providing a more accurate assessment of returns.

[0089] The recovery frequency factor Rfreq is calculated by the following formula:

[0090]

[0091] Where R thl represents the order recovery interval, σ season represents the standard deviation of the return volume fluctuation of this product category in different seasons, μ seasonrepresents the average return volume of the product category in different seasons, W total Indicates the total return weight of the order, W item Indicates the weight of a single product in this order.

[0092] The recycling frequency factor, Rfreq, provides a precise basis for assessing recycling frequency by combining recycling intervals and return volume fluctuations. This comprehensive calculation method allows for more accurate forecasting and management of recycling demand, avoiding the limitations of relying solely on fixed time intervals or return volumes. The calculated recycling frequency factor can help companies optimize the allocation of recycling resources. For example, during periods of high recycling demand, companies can pre-arrange more recycling resources and personnel to ensure a smooth recycling process. By monitoring and analyzing the recycling frequency factor, companies can identify product return patterns in different seasons and implement targeted quality improvements. Understanding the changing trends in recycling frequency can help identify product problems and take measures to improve product quality.

[0093] The logistics cost factor Lcost is calculated by the following formula:

[0094]

[0095] Where C trans represents the shipping cost of a single order, C storage represents the storage cost of a single order, C handling represents the processing cost of a single order, R thl represents the return quantity of a single order, C fuel is the combustion cost during transportation, D distance is the transportation distance of the order, C total is the total logistics cost of a single order.

[0096] The logistics cost factor, Lcost, combines transportation, warehousing, and handling costs, as well as any potential fuel costs during transportation. This comprehensive calculation method allows companies to fully understand the actual logistics costs of each order, avoiding the limitations of focusing on only a subset of costs. By calculating the logistics cost factor in detail, companies can identify and optimize various cost links. For example, by reducing transportation costs or improving warehousing management, companies can reduce overall logistics costs and optimize resource allocation. For example, if fuel costs during transportation are too high, measures can be taken to reduce fuel losses, thereby saving costs. Calculating the logistics cost factor helps companies improve their cost management strategies. Based on cost data, companies can adjust warehousing and handling processes, improve transportation routes and methods, and reduce overall logistics costs.

[0097] The processing time factor Ptime is calculated by the following formula:

[0098]

[0099] Where, T processing Indicates the return processing time of a single order, T inspection Indicates the quality inspection time of a single order, T repackaging represents the repackaging time of a single order, R thl represents the return quantity of a single order, T delay Indicates the delay time in the return process, T optimal Indicates the ideal time threshold for return processing.

[0100] Calculating the processing time factor (Ptime) provides detailed time data for the return processing process, helping to identify and improve efficiency in the return handling process. By shortening processing time at each stage, overall return processing efficiency can be improved. The calculated processing time factor helps companies identify bottlenecks and delays in the process. Based on this data, companies can optimize the processing flow, reduce unnecessary waiting and delays, and improve overall process smoothness. By analyzing the processing time factor, companies can understand the contribution of different stages to the overall processing time. This helps to rationally allocate resources, ensuring that each stage in the return processing process has sufficient resources to support it, thereby reducing processing time. Calculating the processing time factor can help companies optimize warehouse management. By reducing the time spent on repackaging and quality inspection, inventory backlogs can be reduced and storage costs can be lowered.

[0101] After the return volume factor Tqty and the recycling frequency factor Rfreq are dimensionlessly processed, the reverse logistics pressure coefficient Rpr is calculated using the following correlation formula:

[0102] Rpr=Mlxl*b1+Ljqhz*b2+B1

[0103] Wherein, b1 and b2 represent the preset proportional coefficients of the return quantity factor Tqty and the recycling frequency factor Rfreq respectively; B1 is the first correction constant;

[0104] After dimensionless processing of the logistics cost factor Lcost and the processing time factor Ptime, the logistics efficiency coefficient Lef is calculated using the following correlation formula:

[0105] Lef=Lcost*b3+Ptime*b4+B2

[0106] In the formula, b3 and b4 represent the preset proportional coefficients of the logistics cost factor Lcost and the processing time factor Ptime respectively; B2 is the second correction constant.

[0107] In this embodiment, by calculating the reverse logistics pressure coefficient Rpr and the logistics efficiency coefficient Lef, a comprehensive assessment of the reverse logistics system's pressure level and processing efficiency can be achieved. This assessment helps companies identify sources of pressure and efficiency bottlenecks in the processing process, facilitating the development of appropriate optimization measures. The reverse logistics pressure coefficient Rpr provides a quantitative indicator of pressure levels, enabling companies to adjust resource allocation based on actual pressure conditions. Effective resource allocation avoids resource waste, ensures the rational utilization of resources at all stages, and thus improves overall processing capacity.

[0108] Example 5. This example is explained in Example 1. Please refer to Figure 1 ,Specifically, the reverse logistics optimization module includes a pressure evaluation unit and an ,efficiency evaluation unit;

[0109] The pressure assessment unit is used to pre-set a pressure threshold V and compare and analyze the logistics pressure coefficient Rpr with the pressure threshold V to determine whether the current reverse logistics state is normal or not, so as to obtain a logistics pressure assessment result, including:

[0110] When the logistics pressure coefficient Rpr ≤ the pressure threshold V, it indicates that the current reverse logistics pressure is within the normal range, and the resource allocation or processing flow will not be automatically adjusted, and the reverse logistics order will be processed normally;

[0111] When the logistics pressure coefficient Rpr exceeds the pressure threshold V, it indicates that the current reverse logistics pressure exceeds the normal range. The system automatically triggers optimization measures and adjusts the processing strategy, including: designing a multi-level recycling network, including a main recycling center and regional recycling points, and optimizing the path from the return point to the nearest regional recycling point and then to the main recycling center;

[0112] The efficiency evaluation unit is used to preset an efficiency threshold W and compare and analyze the logistics efficiency coefficient Lef with the efficiency threshold W to obtain a logistics efficiency result, including:

[0113] When the logistics efficiency coefficient Lef ≥ efficiency threshold W, it means that the current reverse logistics efficiency is qualified.

[0114] When the logistics efficiency coefficient Lef is less than the efficiency threshold W, it means that the current reverse logistics efficiency is unqualified. The system automatically triggers optimization measures and adjusts the processing strategy, including: increasing the return reception, sorting, inspection, repair and repackaging equipment by 10% and storing them in the warehouse after processing at the nearest regional recycling point.

[0115] In this embodiment, when Rpr ≤ V, the system maintains the current processing flow to ensure normal operation; when Rpr > V, the system will automatically trigger optimization measures. This dynamic adjustment mechanism can promptly address the problem of excessive pressure and prevent processing delays or resource waste caused by excessive pressure. When the pressure exceeds the standard, the system will design a multi-level recovery network, including a main recovery center and regional recovery points. This multi-level recovery network optimizes the path from the return point to the recovery point, improving the resource utilization efficiency. By optimizing the path from the return point to the nearest regional recovery point and then to the main recovery center, the system reduces the logistics cost and processing time and enhances the overall processing capacity. The processing strategy is automatically adjusted to ensure efficient processing of reverse logistics orders even under high pressure, avoiding a decline in processing capacity.

[0116] When Lef ≥ W, it indicates that the current reverse logistics efficiency is qualified; when Lef < W, the system will automatically trigger optimization measures. This efficiency control mechanism can ensure the high efficiency of reverse logistics processing and promptly correct problems of low efficiency.

[0117] For the situation where the efficiency is unqualified, the system will increase the return receiving, sorting, inspection, repair, and repackaging equipment by 10%. This measure can enhance the processing capacity, shorten the processing time, and improve the overall efficiency.

[0118] After adding processing equipment at the nearest regional recovery point, the system can effectively utilize existing resources, increase the processing speed, and reduce equipment idle time.

[0119] After adding processing equipment, preliminary processing can be carried out at the regional recovery point and the products can be stored in the warehouse, optimizing the warehousing management and logistics process and improving the overall processing efficiency of the system.

[0120] The setting of the threshold value is for the convenience of comparison. Regarding the size of the threshold value, it depends on the amount of sample data and the base quantity set by those skilled in the art for each group of sample data; as long as it does not affect the proportional relationship between the parameters and the quantified values.

[0121] The above formulas are all obtained by collecting a large amount of data for software simulation and selecting a formula close to the true value. The coefficients in the formulas are set by those skilled in the art according to the actual situation. As described above, this is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and should be covered within the protection scope of the present invention.

Claims

1. A rule-based supply chain distribution optimization system, characterized by: It includes quality inspection data collection module, rule engine module, reverse logistics monitoring module, reverse logistics analysis module and reverse logistics optimization module; The quality inspection data collection module is used to collect quality parameters of reverse logistics return orders in the supply chain and generate quality inspection data groups; The rule engine module is used to establish and train a quality analysis model, analyze and calculate the quality inspection data set to construct a quality inspection index Zjzl. If the quality inspection index Zjzl is higher than or equal to the quality threshold X, the product is returned to the original supply chain manufacturer. If the quality inspection index Zjzl is lower than the quality threshold X, the product is deemed to be recycled. The rule engine module also dynamically loads and applies preset reverse logistics optimization rules. The reverse logistics monitoring module is used to monitor and record status data information related to the reverse logistics process in real time and generate a reverse logistics monitoring data group; The reverse logistics analysis module is used to extract features from the reverse logistics-related data information in the reverse logistics monitoring data group to obtain a return volume factor Tqty, a recycling frequency factor Rfreq, a logistics cost factor Lcost, and a processing time factor Ptime, and to obtain a reverse logistics pressure coefficient Rpr by associating the return volume factor Tqty with the recycling frequency factor Rfreq, and to obtain a logistics efficiency coefficient Lef by associating the logistics cost factor Lcost with the processing time factor Ptime; The reverse logistics optimization module is used to pre-set the pressure threshold V and the efficiency threshold W, and compare and analyze the logistics pressure coefficient Rpr with the pressure threshold V to obtain the logistics pressure evaluation result, and compare and analyze the logistics efficiency coefficient Lef with the efficiency threshold W to obtain the logistics efficiency result. According to the logistics pressure evaluation result and the logistics efficiency result, the reverse logistics processing flow and resource allocation strategy are automatically adjusted and controlled.

2. The rule-based supply chain distribution optimization system according to claim 1, characterized in that: The quality inspection data acquisition module includes a defect monitoring unit and a quality parameter recording unit; The defect monitoring unit is used to collect defect information of returned products in real time using testing tools and image recognition technology; The quality parameter recording unit is used to record parameter information related to the quality of the returned product in real time, including product category, production batch, quality inspection date and testing method; The quality inspection data group includes: the quality score Q of the returned product score , Check the actual product size deviation value A when returning defect , Detect the actual product defect area D when returning dev , Check the actual product performance test value P when returning test And the actual product packaging defect area Xn detected when returning test .

3. The rule-based supply chain distribution optimization system according to claim 2, characterized in that: The rule engine module includes a model building unit and a returned product quality analysis unit; The model building unit is used to build a quality analysis model through machine learning technology, including linear regression and support vector machine, and divide the quality inspection data group into a training set and a test set for training and testing, and then apply the training and testing to the quality analysis model; The returned product quality analysis unit is used to perform data cleaning, missing value processing, and dimensionless processing on the quality inspection data group through the quality analysis model, and then calculate the quality inspection index Zjzl using the following formula: Where Q std represents the quality standard threshold, Q score Indicates the quality score of the returned product, Q weright The weight of the returned product quality score, A tol Indicates the allowable size deviation threshold, A defect Indicates the actual product size deviation value detected when returning goods, D tol Denotes the allowable defect area threshold, D dev Indicates the actual product defect area detected during return, P tol Indicates the performance tolerance threshold, P test Indicates the actual product performance test value when returning the product, Xn tol Indicates the allowable packaging defect area threshold, Xn test Indicates the actual product packaging defect area detected during return.

4. The rule-based supply chain distribution optimization system according to claim 3, characterized in that: The rule engine module further includes a first evaluation unit and a first strategy unit; The first evaluation unit is configured to compare and analyze the quality inspection index Zjzl of each return order with the quality threshold X to obtain a first evaluation result, including: If the quality inspection index Zjzl ≥ the quality threshold X, it means that the product of the return order meets the return criteria. The first policy unit generates a return to original manufacturer policy, which includes returning the product to the original manufacturer for further processing, repair, or rework. If the quality inspection index Zjzl is less than the quality threshold X, it means that the product of the return order does not meet the return standard. The first policy unit generates a recycling policy to send the product back to the recycling point or processing center for disassembly, repair or destruction.

5. The rule-based supply chain distribution optimization system according to claim 1, characterized in that: The return quantity factor Tqty is calculated using the following formula: Where R thl Indicates the return quantity of a single order, S sl represents the sales volume of a single order, D return Indicates the historical return quantity of the product category to which the order belongs, D total Indicates the total sales quantity of the product category to which the order belongs.

6. The rule-based supply chain distribution optimization system according to claim 1, characterized in that: The recovery frequency factor Rfreq is calculated by the following formula: Where R thl represents the order recovery interval, σ season represents the standard deviation of the return volume fluctuation of this product category in different seasons, μ season represents the average return volume of the product category in different seasons, W total Indicates the total return weight of the order, W item Indicates the weight of a single product in this order.

7. The rule-based supply chain distribution optimization system according to claim 1, characterized in that: The logistics cost factor Lcost is calculated by the following formula: Where C trans represents the shipping cost of a single order, C storage represents the storage cost of a single order, C handling represents the processing cost of a single order, R thl represents the return quantity of a single order, C fuel is the combustion cost during transportation, D distance is the transportation distance of the order, C total is the total logistics cost of a single order.

8. The rule-based supply chain distribution optimization system according to claim 1, characterized in that: The processing time factor Ptime is calculated by the following formula: Where, T processing Indicates the return processing time of a single order, T inspection Indicates the quality inspection time of a single order, T repackaging represents the repackaging time of a single order, R thl represents the return quantity of a single order, T delay Indicates the delay time in the return process, T optimal Indicates the ideal time threshold for return processing.

9. The rule-based supply chain distribution optimization system according to claim 8, characterized in that: After the return volume factor Tqty and the recycling frequency factor Rfreq are dimensionlessly processed, the reverse logistics pressure coefficient Rpr is calculated using the following correlation formula: Rpr=Mlxl*b1+Ljqhz*b2+B1 Wherein, b1 and b2 represent the preset proportional coefficients of the return quantity factor Tqty and the recycling frequency factor Rfreq respectively; B1 is the first correction constant; After dimensionless processing of the logistics cost factor Lcost and the processing time factor Ptime, the logistics efficiency coefficient Lef is calculated using the following correlation formula: Lef=Lcost*b3+Ptime*b4+B2 In the formula, b3 and b4 represent the preset proportional coefficients of the logistics cost factor Lcost and the processing time factor Ptime respectively; B2 is the second correction constant.

10. The rule-based supply chain distribution optimization system according to claim 9, characterized in that: The reverse logistics optimization module includes a pressure evaluation unit and an efficiency evaluation unit; The pressure assessment unit is used to pre-set a pressure threshold V and compare and analyze the logistics pressure coefficient Rpr with the pressure threshold V to determine whether the current reverse logistics state is normal or not, so as to obtain a logistics pressure assessment result, including: When the logistics pressure coefficient Rpr ≤ the pressure threshold V, it indicates that the current reverse logistics pressure is within the normal range, and the resource allocation or processing flow will not be automatically adjusted, and the reverse logistics order will be processed normally; When the logistics pressure coefficient Rpr exceeds the pressure threshold V, it indicates that the current reverse logistics pressure exceeds the normal range. The system automatically triggers optimization measures and adjusts the processing strategy, including: designing a multi-level recycling network, including a main recycling center and regional recycling points, and optimizing the path from the return point to the nearest regional recycling point and then to the main recycling center; The efficiency evaluation unit is used to preset an efficiency threshold W and compare and analyze the logistics efficiency coefficient Lef with the efficiency threshold W to obtain a logistics efficiency result, including: When the logistics efficiency coefficient Lef ≥ efficiency threshold W, it means that the current reverse logistics efficiency is qualified. When the logistics efficiency coefficient Lef is less than the efficiency threshold W, it means that the current reverse logistics efficiency is unqualified. The system automatically triggers optimization measures and adjusts the processing strategy, including: increasing the return reception, sorting, inspection, repair and repackaging equipment by 10% and storing them in the warehouse after processing at the nearest regional recycling point.

Images