Park digital intelligent logistics supply system

Abstract

The invention discloses a park digital intelligent logistics supply system, relates to the technical field of park digital intelligent logistics supply, is used for solving the problems of data isolation, decision lag and unreasonable resource allocation in existing park logistics supply, and comprises a data acquisition module, a data processing module and an intelligent analysis and decision module. The data acquisition module performs multi-dimensional data acquisition through a preset multi-element sensing network; the data processing module performs feature extraction based on the collected data and performs calculation to obtain a health degree evaluation index; the intelligent analysis and decision module performs analysis and judgment based on the health degree evaluation index and generates a corresponding strategy combination; according to the invention, data acquisition, multi-dimensional feature analysis and intelligent strategy generation are carried out through the multi-element sensing network, dynamic monitoring of the logistics operation state, dynamic balance of supply and demand and collaborative optimization are realized, and the logistics operation efficiency and supply chain toughness of the park are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of digital logistics supply for industrial parks, and specifically to a digital logistics supply system for industrial parks. Background Art

[0002] As a core area for upgrading modern industrial park supply chains, digitalization of industrial parks is rapidly being adopted in scenarios such as industrial parks and e-commerce warehouses. Through IoT sensors, automated sorting equipment, and intelligent scheduling algorithms, digitalization of industrial parks enables visual management of the entire material flow process, providing data support for inventory optimization and production data collaboration. However, existing technologies still have many problems:

[0003] In the data integration and decision-making stages, most solutions rely on single-dimensional warehousing or transportation data analysis and lack the deep integration of multi-source heterogeneous data, making it difficult to accurately capture the patterns of supply and demand fluctuations, and abnormal responses lag behind actual changes in demand. At the same time, resource scheduling is rigidly based on static threshold rules and cannot generate flexible solutions based on inventory levels and equipment operating status, resulting in data isolation, delayed decision-making, and irrational resource allocation in the park's logistics supply.

[0004] In order to solve the above-mentioned defects, a technical solution is now provided. Summary of the Invention

[0005] The purpose of this invention is to solve the problems of data isolation, delayed decision-making and unreasonable resource allocation in the existing digital logistics supply of parks, and to propose a digital logistics supply system for parks.

[0006] The objectives of the present invention can be achieved through the following technical solutions: a digital and intelligent logistics supply system for a park, comprising a data acquisition module for collecting multi-dimensional data on logistics, inventory, orders, and production status within the park, as well as external market environment source data through a multi-sensing network, to generate unified park logistics operation data;

[0007] The data processing module is used to pre-process the park logistics operation data and extract time series features, calculate the inventory health comprehensive index, order health comprehensive index and production health comprehensive index; and calculate the health evaluation index HOL;

[0008] The intelligent analysis and decision-making module analyzes and judges and generates corresponding strategy combinations based on the comparison results of the health assessment index HOL and the preset threshold, combined with the inventory health comprehensive index, order health comprehensive index and production health comprehensive index.

[0009] Furthermore, the data collection module includes internal collection nodes in the park and external market collection nodes. The specific data collection process is as follows:

[0010] The internal collection nodes of the park are composed of IoT sensors, automatic identification units, intelligent visual monitoring units and production line status monitoring units;

[0011] The IoT sensor consists of a temperature and humidity sensor, a position sensor, and an environmental sensor, which is used to collect logistics status data and internal environmental parameters of the park. Based on the internal environmental parameters and logistics status data, the IoT sensor collects the logistics, inventory, order, and production status within the park to obtain the corresponding park logistics operation data.

[0012] The automatic identification unit collects material in-and-out information and logistics flow trajectory data through RFID readers, QR code scanners and smart tag identifiers;

[0013] The intelligent visual monitoring unit collects information on cargo stacking status, handling efficiency, and personnel flow through a distributed camera array and edge computing layer;

[0014] The external market collection node collects market environment source data through the e-commerce data interface, the National Bureau of Statistics public data interface and third-party data service providers;

[0015] The obtained multi-dimensional data is time-synchronized and associated through a distributed clock synchronization protocol to generate unified park logistics operation data.

[0016] Furthermore, the data processing module performs data preprocessing and feature extraction on the park logistics operation data. The specific processing process includes:

[0017] The data preprocessing includes using the LSTM neural network prediction filling method to predict and fill missing values, using the isolation forest algorithm to detect outliers in multidimensional features and correcting the abnormal data in the multidimensional features, and using the RobustScaler method to perform data standardization and nonlinear normalization on the data;

[0018] The feature extraction includes using a time series feature extraction method to extract time dimension features from pre-processed park logistics operation data to obtain corresponding multidimensional feature data; the multidimensional feature data includes inventory features, order features and production features, and captures parameter change trends and fluctuation characteristics; and the health assessment index HOL is calculated based on the multidimensional feature data.

[0019] Furthermore, the specific calculation process of the health evaluation index HOL includes:

[0020] The inventory characteristics include warehouse management efficiency, storage rationality index, inventory density fluctuation rate and expired material ratio, which are normalized and then inserted into the formula The comprehensive index of inventory health, ZHS, is obtained, where KXL represents the warehouse management efficiency; HLS represents the storage rationality index, which is obtained by matching the goods stacking density with the space utilization rate; HJD represents the inventory density fluctuation rate; GQW represents the ratio of expired materials to the total number of materials, which is the proportion of expired materials; ω1, ω2, ω3, and ω4 are the weighted influencing factors of warehouse management efficiency, storage rationality index, inventory density fluctuation rate, and expired material proportion, respectively.

[0021] The order characteristics include order fulfillment time SJ, delayed delivery rate FH, order status abnormality rate YC and supply chain response delay XC, which are normalized and then inserted into the formula The comprehensive order health index OFS is obtained, where u1, u2, u3, and u4 are the influencing weight factors of delayed shipment rate, order fulfillment time, order status abnormality rate, and supply chain response delay, respectively;

[0022] The production characteristics include production line efficiency, equipment failure rate, material consumption matching and production capacity fluctuation rate; after normalization, enter the formula The comprehensive production health index SCS is obtained, where XL is the production line efficiency, GZ is the equipment failure rate, WL is the material consumption matching degree, and BD is the production capacity fluctuation rate; p1, p2, p3, and p4 are the influencing weight factors of production line efficiency XL, equipment failure rate GZ, material consumption matching degree WL, and production capacity fluctuation rate BD, respectively;

[0023] After normalization, the inventory health comprehensive index ZHS, order health comprehensive index OFS and production health comprehensive index SCS are put into the formula The health evaluation index HOL is obtained, HOL∈[μ,λ], where α, β, and γ are the influencing weight factors of the inventory health comprehensive index, order health comprehensive index, and production health comprehensive index, respectively.

[0024] Furthermore, the specific operation steps of the intelligent analysis and decision-making module include:

[0025] When the health evaluation index HOL is within the preset threshold range, it is judged that the current park health is in a normal state. Then, time series analysis and machine learning methods are used to optimize the demand forecast model; the inventory structure is adjusted based on the ABC classification management method to optimize the safety stock level; production parameters and marketing strategies are adjusted to maintain a dynamic balance between supply and demand;

[0026] When the health assessment index HOL exceeds the preset threshold range, it is judged that there is a problem with the current park, causing the health of the current park to be in an abnormal state. The inventory health comprehensive index, order health comprehensive index and production health comprehensive index are then analyzed.

[0027] Furthermore, the specific analysis steps for the inventory health comprehensive index, order health comprehensive index and production health comprehensive index are as follows:

[0028] Compare the inventory health index ZHS with the preset threshold and generate a corresponding strategy combination based on the comparison results. The specific steps are as follows:

[0029] When ZHS is higher than the threshold and OFS and SCS are within the threshold, the inventory status is judged to be abnormal. The inventory management efficiency, storage rationality index, inventory density fluctuation rate, and expired material ratio in the inventory health index are extracted and compared with the preset thresholds respectively. If it is higher than the preset threshold, it is judged as an abnormal cause and an inventory strategy combination is generated;

[0030] The order health comprehensive index (OFS) and the production health index (SCS) are analyzed. When the OFS exceeds the preset threshold and the SCS is within the preset threshold, the order health is judged to be abnormal. The order fulfillment time, delayed shipment rate, order status abnormality rate, and supply chain response delay in the order health comprehensive index are extracted and compared with the corresponding preset thresholds. If they are higher than the preset threshold, it is judged as an abnormal cause and the corresponding order strategy combination is generated.

[0031] When the SCS exceeds the preset threshold and the OFS is within the preset threshold, the production status is judged to be abnormal. The production line efficiency, equipment failure rate, material consumption matching, and production capacity fluctuation rate in the comprehensive production health index are extracted and compared with the corresponding preset thresholds. If it is higher than the preset threshold, it is judged to be an abnormal cause and the corresponding production strategy combination is generated;

[0032] If OFS and SCS exceed the preset thresholds at the same time, the order strategy combination and production strategy combination will be automatically executed.

[0033] Furthermore, the specific analysis step 2 of the inventory health comprehensive index, order health comprehensive index and production health comprehensive index is as follows:

[0034] When ZHS is lower than the preset threshold, it is judged that the current park inventory is insufficient, and the order health comprehensive index OFS and production health index SCS are analyzed;

[0035] If OFS exceeds the preset threshold and SCS is within the preset threshold, the order health is judged to be abnormal. The order fulfillment time, delayed shipment rate, order status abnormality rate and supply chain response delay in the order health comprehensive index are extracted and compared with the corresponding preset thresholds. If it is higher than the preset threshold, it is judged as an abnormal cause and the corresponding order processing plan is generated;

[0036] If the SCS exceeds the preset threshold and the OFS is within the preset threshold, the production status is judged to be abnormal. The production line efficiency, equipment failure rate, material consumption matching, and production capacity fluctuation rate in the comprehensive production health index are extracted and compared with the corresponding preset thresholds. If it is higher than the preset threshold, it is judged to be an abnormal cause and a corresponding production treatment plan is generated;

[0037] If OFS and SCS exceed the corresponding preset thresholds at the same time, the order processing plan and production processing plan will be automatically executed.

[0038] Furthermore, the strategy combination includes:

[0039] The inventory strategy portfolio includes building a market demand forecast model based on historical sales data, market trends, and unexpected events. Based on the output of the market demand forecast model, a cross-regional inventory allocation plan is generated to allocate excess inventory to high-demand areas. The inventory is managed differently by combining the ABC classification method with the product life cycle stage.

[0040] The order strategy combination includes generating picking routes based on the ant colony algorithm to optimize sorting time for orders with long fulfillment times. If the current order volume is less than 40% of the average over the past cycle time, and the proportion of orders from large customers exceeds 60%, it is judged as a small order quantity. For small order quantities, based on the threshold exceeded by the ZHS, the business department is issued a business instruction to increase the corresponding order and conduct tiered promotion activities based on inventory products.

[0041] The production strategy combination includes: if excessive production efficiency leads to overcapacity, the production efficiency of the relevant products will be reduced to 50% of the original level, and the production quantity of the relevant products will be reduced; excessive production efficiency means that the production efficiency exceeds the preset threshold multiple times within the cycle time; overcapacity means that the inventory level exceeds 120% of the safety stock;

[0042] If material consumption does not match, digital twin technology is used to self-correct process parameters;

[0043] The order processing solution includes optimizing production scheduling through order priority. Order priority is obtained by comprehensively analyzing order value, customer level, contract constraints, delivery time, and strategic significance. After normalization, the priority score is substituted into the formula YXJ = t1·DZ + t2·KH + t3·HT + t4·JH + t5·ZL, where DZ is the order value, KH is the customer level, HT is the contract constraint, JH is the delivery time, and ZL is the strategic significance. t1, t2, t3, t4, and t5 are the weighting factors of order value, customer level, contract constraints, delivery time, and strategic significance, respectively, and t1+t2+t3+t4+t5=1. Priority sorting is performed based on the priority score, and the priority is recalculated every n hours to trigger a production scheduling update.

[0044] Based on order priority, high-priority orders are produced first, and inventory is dispatched from nearby warehouses to meet order demand.

[0045] Furthermore, the treatment plan includes:

[0046] The order processing solution includes optimizing production scheduling through order priority. Order priority is obtained by comprehensively analyzing order value, customer level, contract constraints, delivery time, and strategic significance. After normalization, the priority score is substituted into the formula YXJ = t1·DZ + t2·KH + t3·HT + t4·JH + t5·ZL, where DZ is the order value, KH is the customer level, HT is the contract constraint, JH is the delivery time, and ZL is the strategic significance. t1, t2, t3, t4, and t5 are the weighting factors of order value, customer level, contract constraints, delivery time, and strategic significance, respectively, and t1+t2+t3+t4+t5=1. Priority sorting is performed based on the priority score, and the priority is recalculated every n hours to trigger a production scheduling update.

[0047] Based on order priority, high-priority orders are produced first, and inventory is dispatched from nearby warehouses to meet order demand;

[0048] The production processing solution includes comparing the equipment operation data collected in real time by deployed IoT sensors with the preset operating parameter thresholds. When deviations in the operating data are detected, adjustment instructions are automatically issued.

[0049] The production shift was adjusted from the standard two-shift system to a three-shift system;

[0050] Resource scheduling recommendations include optimizing the raw material supply chain, shortening the raw material procurement cycle, increasing raw material inventory, activating emergency supply channels, and adjusting the supplier structure;

[0051] When the material consumption matching degree exceeds the preset threshold, the raw material inventory is detected to be lower than the safety stock threshold and the real-time consumption is higher than the inventory replenishment. It is judged that the raw materials in stock are insufficient, and the raw material inventory is dispatched from the nearby warehouse to ensure the continuity of production.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] The present invention collects park logistics operation data by constructing a multi-element perception network, adopts LSTM neural network filling and isolation forest algorithm to process multi-source data anomalies, combines the health assessment index HOL to collaboratively monitor inventory, orders and production links, analyzes and judges the health assessment index, generates adjustment strategy combinations, and forms flexible resource allocation capabilities; through the multi-objective collaborative optimization algorithm, adaptive adjustment of resource scheduling and intelligent generation of strategy combinations are realized, ensuring the dynamic balance of supply and demand, and effectively improving the park's operational efficiency and supply resilience.

[0054] Attached figure judgment

[0055] In order to facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings;

[0056] Figure 1 This is the overall system block diagram of the present invention. DETAILED DESCRIPTION

[0057] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0058] It should be understood that the terms "include" and "comprising" used in the judgment and claims of this disclosure indicate the presence of described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0059] It should also be understood that the terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to limit the disclosure. As used in this disclosure and in the claims, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should be further understood that the term "and / or" as used in this disclosure and in the claims refers to any and all possible combinations of one or more of the associated listed items, including and including these combinations.

[0060] like Figure 1 As shown, a digital logistics supply system for a park includes: a data acquisition module, a data processing module, and an intelligent analysis and decision-making module.

[0061] The data collection module collects multi-dimensional data on logistics, inventory, orders, and production status within the park through a pre-set multi-sensing network to obtain the corresponding park logistics operation data; it also records relevant market environment source data and stores it in a pre-built park data center. The specific process of collecting park logistics operation data includes:

[0062] The data acquisition module includes internal collection nodes in the park and external market collection nodes; the internal collection nodes in the park are composed of an Internet of Things sensor network, an automatic identification unit, an intelligent visual monitoring unit and a production line status monitoring unit; among them, the Internet of Things sensors are composed of temperature and humidity sensors, position sensors, pressure sensors and environmental monitoring sensors distributed in various areas, which are used to collect environmental parameters and logistics status data in the park; based on the internal environmental parameters and logistics status data of the park, the logistics, inventory, orders and production status in the park are collected to obtain the corresponding park logistics operation data; the park logistics operation data includes real-time inventory quantity, inventory distribution, inventory status, order processing status, production line efficiency and environmental data; the automatic identification unit includes an RFID reader, a QR code scanner and an intelligent label identifier; it is used to collect material information, material in and out information and logistics flow trajectory data; the intelligent visual monitoring unit is composed of a distributed camera array and an edge computing layer, which is used to collect the cargo stacking status, cargo handling efficiency and personnel flow in the park; the production line status monitoring unit collects the operating parameters, production capacity data and material consumption rate of each production line through sensors connected to the production equipment;

[0063] External market collection nodes include e-commerce data interfaces, the National Bureau of Statistics' public data interfaces, and third-party data service providers, which are used to obtain market source data, including demand trends, competitive product dynamics, and raw material price fluctuations.

[0064] The obtained multi-dimensional data is synchronized and associated with the data through a distributed clock synchronization protocol to generate unified park logistics operation data;

[0065] The data processing module performs data preprocessing and feature extraction on the collected park logistics operation data based on the collected park operation data. The specific process of data preprocessing and feature extraction on the collected park operation data includes: obtaining the collected park operation data, and preprocessing the data. The data preprocessing includes using the LSTM neural network prediction filling method to predict and fill in missing values, using the isolation forest algorithm to detect abnormal points in multidimensional features and correct abnormal data in multidimensional features, and using the RobustScaler method to perform data standardization and nonlinear normalization on the data; using the time series feature extraction method to extract the time dimension features of the preprocessed park logistics operation data to obtain the corresponding multidimensional feature data; the multidimensional feature data includes inventory features, order features and production features, and captures parameter change trends and fluctuation characteristics; inventory features include warehouse management efficiency, storage rationality index, inventory density fluctuation rate and the proportion of expired materials; after normalization processing, the formula is inserted The calculation results in the comprehensive inventory health index ZHS, where KXL represents the warehouse management efficiency; HLS represents the storage rationality index, which is obtained by matching the goods stacking density with the space utilization rate; HJD represents the inventory density fluctuation rate; GQW represents the ratio of expired materials to the total number of materials, which is the proportion of expired materials; ω1, ω2, ω3, and ω4 are the weighted influencing factors of warehouse management efficiency, storage rationality index, inventory density fluctuation rate, and the proportion of expired materials, respectively.

[0066] Order characteristics include order fulfillment time SJ, delayed delivery rate FH, order status abnormality rate YC and supply chain response delay XC; after normalization, they are entered into the formula The order health comprehensive index OFS is calculated, where u1, u2, u3, and u4 are the influencing weight factors of delayed shipment rate, order fulfillment time, order status abnormality rate, and supply chain response delay, respectively;

[0067] Production characteristics include production line efficiency, equipment failure rate, material consumption matching, and production capacity fluctuation rate; after normalization, they are entered into the formula The production health comprehensive index SCS is calculated, where XL is the production line efficiency, GZ is the equipment failure rate, WL is the material consumption matching degree, and BD is the production capacity fluctuation rate; p1, p2, p3, and p4 are the influencing weight factors of production line efficiency XL, equipment failure rate GZ, material consumption matching degree WL, and production capacity fluctuation rate BD, respectively;

[0068] After normalization, the inventory health comprehensive index ZHS, order health comprehensive index OFS and production health comprehensive index SCS are put into the formula The health evaluation index HOL is obtained, HOL∈[μ,λ], where α, β, and γ are the influencing weight factors of the inventory health comprehensive index, order health comprehensive index, and production health comprehensive index, respectively.

[0069] The intelligent analysis and decision-making module compares the health evaluation index HOL with the preset threshold to classify and judge the current park health. It then analyzes and proposes a strategy combination based on the inventory health comprehensive index, order health comprehensive index, and production health comprehensive index. The specific steps are as follows:

[0070] Based on the comparison between the health assessment index HOL and the preset threshold, when HOL is within the preset threshold range, the current park health is judged to be in a normal state. Time series analysis and machine learning methods are then used to analyze historical sales data to enhance demand forecast accuracy and respond in advance to possible supply and demand fluctuations. Based on the ABC classification management method, the inventory structure is adjusted and the safety stock level is optimized. Based on the uncertainty of demand forecast, a safety stock is set to cope with possible demand fluctuations. Production parameters and marketing strategies are fine-tuned to maintain a dynamic balance between supply and demand. When HOL exceeds the preset threshold range, it is judged that there is a problem with the current park, resulting in the current park health being in an abnormal state. Further analysis is then conducted on the inventory health comprehensive index ZHS, the order health comprehensive index OFS, and the production health comprehensive index SCS.

[0071] Compare the inventory health index ZHS with the preset threshold and generate a corresponding strategy combination. The specific steps are as follows:

[0072] When ZHS is above the threshold and OFS and SCS are within the threshold, the inventory status is judged to be abnormal. The inventory health index, including warehouse management efficiency, storage rationality index, inventory density fluctuation rate, and expired material ratio, is extracted and compared with the preset threshold. If it is above the preset threshold, it is judged as an abnormal cause and an inventory strategy combination is generated. This includes building a market demand forecast model based on historical sales data, market trends, and emergencies. Based on the output of the market demand forecast model, a cross-regional inventory allocation plan is generated to allocate excess inventory to high-demand areas. In addition, the ABC classification method is combined with the product life cycle stage to implement differentiated inventory management.

[0073] Further analysis of the order health comprehensive index OFS and production health index SCS; the specific steps include:

[0074] When OFS exceeds the preset threshold and SCS is within the preset threshold, it is judged that there is an abnormality in the order health, and the order fulfillment time, delayed delivery rate, order status abnormality rate and supply chain response delay in the comprehensive index of order health are extracted and compared with the corresponding preset thresholds respectively. If it is higher than the preset threshold, it is judged as the abnormal reason and the corresponding order strategy combination is generated. The order strategy combination includes: for orders with long fulfillment time, a picking path is generated based on the ant colony algorithm to optimize the sorting time; for small order quantities, based on the threshold exceeded by ZHS, an instruction to increase the corresponding order business is issued to the business department and graded promotion activities are carried out based on inventory products; graded promotions are carried out by defining products and adopting different promotion methods based on the definition results; product definitions include general products, slow-moving products and near-expiry products; promotion methods include: for general products in inventory, bundling sales are used for promotion, such as combining product A and product B and selling them at a preferential price; for slow-moving products in inventory, limited-time discounts are used for promotion, such as selling slow-moving products at a discount within a certain period of time; for near-expiry products in inventory, public donations or discount clearance promotions are used for promotion;

[0075] When SCS exceeds the preset threshold and OFS is within the preset threshold, the production status is judged to be abnormal. The production line efficiency, equipment failure rate, material consumption matching, and production capacity fluctuation rate in the comprehensive production health index are extracted and compared with the corresponding preset thresholds. If it is higher than the preset threshold, it is judged to be an abnormal cause and the corresponding production strategy combination is generated, including:

[0076] If excessive production efficiency leads to overcapacity, the production efficiency of the relevant products will be reduced to 50% of the original level, and the production quantity of the relevant products will be reduced. Excessive production efficiency means that the production efficiency exceeds the preset threshold multiple times within the cycle time. Overcapacity means that the inventory exceeds 120% of the safety stock.

[0077] If material consumption does not match, digital twin technology is used to self-correct process parameters;

[0078] If both OFS and SCS exceed the corresponding preset thresholds at the same time, the order strategy combination and production strategy combination will be automatically executed;

[0079] When ZHS is lower than the preset threshold, it is judged that the current park inventory is insufficient, and further analysis is conducted on the order health comprehensive index OFS and the production health index SCS;

[0080] If OFS exceeds the preset threshold and SCS is within the preset threshold, it is judged that there is an abnormality in the order health. The order fulfillment time, delayed delivery rate, order status abnormality rate and supply chain response delay in the order health comprehensive index are extracted and compared with the corresponding preset thresholds respectively. If it is higher than the preset threshold, it is judged as the cause of the abnormality and the corresponding order processing plan is generated, including optimizing the production schedule through order priority. The order priority is comprehensively analyzed by order value, customer level, contract constraints, delivery time and strategic significance, and after normalization, it is substituted into the formula The priority score is calculated by YXJ = t1·DZ + t2·KH + t3·HT + t4·JH + t5·ZL, where DZ is the order value, KH is the customer level, HT is the contract constraint, JH is the delivery time, and ZL is the strategic significance. t1, t2, t3, t4, and t5 are the weighting factors for order value, customer level, contract constraint, delivery time, and strategic significance, respectively. Furthermore, t1+t2+t3+t4+t5=1. Priority sorting is performed based on the priority score. Priorities are recalculated every n hours, triggering an update to the production scheduling plan.

[0081] Based on order priority, high-priority orders are produced first, and inventory is dispatched from nearby warehouses to meet order demand;

[0082] If SCS exceeds the preset threshold and OFS is within the preset threshold, it is judged that there is an abnormality in the production status, and the production line efficiency, equipment failure rate, material consumption matching and production capacity fluctuation rate in the comprehensive production health index are extracted and compared with the corresponding preset thresholds respectively. If it is higher than the preset threshold, it is judged as the cause of the abnormality and a corresponding production processing plan is generated, including comparing the equipment operation data collected in real time by the deployed IoT sensors with the preset operation parameter thresholds. When the operation data deviation is detected, the adjustment instruction is automatically issued. The adjustment instruction includes adjusting the motor speed by +5% each time, the energy consumption parameter by +15% and the pressure parameter by +5%. The production shift generates manpower demand based on order fluctuations and the actual number of workers. The demand is to adjust workers' working hours and increase production shifts from the standard two-shift system to a three-shift system. Resource scheduling suggestions include optimizing the raw material supply chain, negotiating with suppliers to shorten the raw material procurement cycle, increasing raw material inventory, and activating emergency supply channels, such as finding backup suppliers or activating safety stocks in inventory; as well as adjusting the supplier structure, establishing long-term cooperative relationships with high-quality suppliers and reducing dependence on low-quality suppliers; when the material consumption matching degree exceeds the preset threshold, the raw material inventory is detected to be lower than the safety stock threshold and the real-time consumption is higher than the inventory replenishment, judging that the raw material inventory is insufficient, and the raw material inventory is dispatched from the adjacent warehouse to ensure production continuity; the adjacent warehouse is the warehouse with the shortest distance and sufficient inventory;

[0083] If OFS and SCS exceed the corresponding preset thresholds at the same time, the order processing plan and production processing plan will be automatically executed.

[0084] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to specific embodiments. Obviously, many modifications and variations are possible based on the contents of this judgment. These embodiments are selected and described in detail in this judgment to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A digital logistics supply system for a park, characterized by: include: The data collection module is used to collect multi-dimensional data on logistics, inventory, orders, and production status within the park, as well as external market environment source data through a multi-sensing network, to generate unified park logistics operation data; The data processing module is used to pre-process the park's logistics operation data and extract time series features, and calculate the comprehensive index of inventory health, order health, and production health; And calculate the health assessment index HOL; The intelligent analysis and decision-making module analyzes and judges and generates corresponding strategy combinations based on the comparison results of the health assessment index HOL and the preset threshold, combined with the inventory health comprehensive index, order health comprehensive index and production health comprehensive index.

2. A digital logistics supply system for a park according to claim 1, characterized in that: The data collection module includes internal collection nodes in the park and external market collection nodes. The specific data collection process is as follows: The internal collection nodes of the park are composed of IoT sensors, automatic identification units, intelligent visual monitoring units and production line status monitoring units; The IoT sensor consists of a temperature and humidity sensor, a position sensor, and an environmental sensor, which is used to collect logistics status data and internal environmental parameters of the park. Based on the internal environmental parameters and logistics status data, the IoT sensor collects the logistics, inventory, order, and production status within the park to obtain the corresponding park logistics operation data. The automatic identification unit collects material in-and-out information and logistics flow trajectory data through RFID readers, QR code scanners and smart tag identifiers; The intelligent visual monitoring unit collects information on cargo stacking status, handling efficiency, and personnel flow through a distributed camera array and edge computing layer; The external market collection node collects market environment source data through the e-commerce data interface, the National Bureau of Statistics public data interface and third-party data service providers; The obtained multi-dimensional data is time-synchronized and associated through a distributed clock synchronization protocol to generate unified park logistics operation data.

3. A digital logistics supply system for a park according to claim 2, characterized in that: The data processing module performs data preprocessing and feature extraction on the park logistics operation data. The specific processing process includes: The data preprocessing includes using the LSTM neural network prediction filling method to predict and fill missing values, using the isolation forest algorithm to detect outliers in multidimensional features and correcting the abnormal data in the multidimensional features, and using the RobustScaler method to perform data standardization and nonlinear normalization on the data; The feature extraction includes using a time series feature extraction method to extract time dimension features from pre-processed park logistics operation data to obtain corresponding multidimensional feature data; the multidimensional feature data includes inventory features, order features and production features, and captures parameter change trends and fluctuation characteristics; and the health assessment index HOL is calculated based on the multidimensional feature data.

4. A digital logistics supply system for a park according to claim 3, characterized in that: The specific calculation process of the health evaluation index HOL includes: The inventory characteristics include warehouse management efficiency, storage rationality index, inventory density fluctuation rate and expired material ratio, which are normalized and then inserted into the formula The comprehensive index of inventory health, ZHS, is obtained, where KXL represents the warehouse management efficiency; HLS represents the storage rationality index, which is obtained by matching the goods stacking density with the space utilization rate; HJD represents the inventory density fluctuation rate; GQW represents the ratio of expired materials to the total number of materials, which is the proportion of expired materials; ω1, ω2, ω3, and ω4 are the weighted influencing factors of warehouse management efficiency, storage rationality index, inventory density fluctuation rate, and expired material proportion, respectively. The order characteristics include order fulfillment time SJ, delayed delivery rate FH, order status abnormality rate YC and supply chain response delay XC, which are normalized and then inserted into the formula The comprehensive order health index OFS is obtained, where u1, u2, u3, and u4 are the influencing weight factors of delayed shipment rate, order fulfillment time, order status abnormality rate, and supply chain response delay, respectively; The production characteristics include production line efficiency, equipment failure rate, material consumption matching and production capacity fluctuation rate; after normalization, enter the formula The comprehensive production health index SCS is obtained, where XL is the production line efficiency, GZ is the equipment failure rate, WL is the material consumption matching degree, and BD is the production capacity fluctuation rate; p1, p2, p3, and p4 are the influencing weight factors of production line efficiency XL, equipment failure rate GZ, material consumption matching degree WL, and production capacity fluctuation rate BD, respectively; After normalization, the inventory health comprehensive index ZHS, order health comprehensive index OFS and production health comprehensive index SCS are put into the formula The health evaluation index HOL is obtained, HOL∈[μ,λ], where α, β, and γ are the influencing weight factors of the inventory health comprehensive index, order health comprehensive index, and production health comprehensive index, respectively.

5. A digital logistics supply system for a park according to claim 4, characterized in that: The specific operation steps of the intelligent analysis and decision-making module include: When the health evaluation index HOL is within the preset threshold range, it is judged that the current park health is in a normal state. Then, time series analysis and machine learning methods are used to optimize the demand forecast model; the inventory structure is adjusted based on the ABC classification management method to optimize the safety stock level; production parameters and marketing strategies are adjusted to maintain a dynamic balance between supply and demand; When the health assessment index HOL exceeds the preset threshold range, it is judged that there is a problem with the current park, causing the health of the current park to be in an abnormal state. The inventory health comprehensive index, order health comprehensive index and production health comprehensive index are then analyzed.

6. A digital logistics supply system for a park according to claim 5, characterized in that: The specific analysis steps for the inventory health comprehensive index, order health comprehensive index and production health comprehensive index are as follows: Compare the inventory health index ZHS with the preset threshold and generate a corresponding strategy combination based on the comparison results. The specific steps are as follows: When ZHS is higher than the threshold and OFS and SCS are within the threshold, the inventory status is judged to be abnormal. The inventory management efficiency, storage rationality index, inventory density fluctuation rate, and expired material ratio in the inventory health index are extracted and compared with the preset thresholds respectively. If it is higher than the preset threshold, it is judged as an abnormal cause and an inventory strategy combination is generated; The order health comprehensive index (OFS) and the production health index (SCS) are analyzed. When the OFS exceeds the preset threshold and the SCS is within the preset threshold, the order health is judged to be abnormal. The order fulfillment time, delayed shipment rate, order status abnormality rate, and supply chain response delay in the order health comprehensive index are extracted and compared with the corresponding preset thresholds. If they are higher than the preset threshold, it is judged as an abnormal cause and the corresponding order strategy combination is generated. When the SCS exceeds the preset threshold and the OFS is within the preset threshold, the production status is judged to be abnormal. The production line efficiency, equipment failure rate, material consumption matching, and production capacity fluctuation rate in the comprehensive production health index are extracted and compared with the corresponding preset thresholds. If it is higher than the preset threshold, it is judged to be an abnormal cause and the corresponding production strategy combination is generated; If OFS and SCS exceed the preset thresholds at the same time, the order strategy combination and production strategy combination will be automatically executed.

7. A digital logistics supply system for a park according to claim 6, characterized in that: The specific analysis step 2 of the inventory health comprehensive index, order health comprehensive index and production health comprehensive index is as follows: When ZHS is lower than the preset threshold, it is judged that the current park inventory is insufficient, and the order health comprehensive index OFS and production health index SCS are analyzed; If OFS exceeds the preset threshold and SCS is within the preset threshold, the order health is judged to be abnormal. The order fulfillment time, delayed shipment rate, order status abnormality rate and supply chain response delay in the order health comprehensive index are extracted and compared with the corresponding preset thresholds. If it is higher than the preset threshold, it is judged as an abnormal cause and the corresponding order processing plan is generated; If the SCS exceeds the preset threshold and the OFS is within the preset threshold, the production status is judged to be abnormal. The production line efficiency, equipment failure rate, material consumption matching, and production capacity fluctuation rate in the comprehensive production health index are extracted and compared with the corresponding preset thresholds. If it is higher than the preset threshold, it is judged to be an abnormal cause and a corresponding production treatment plan is generated; If OFS and SCS exceed the corresponding preset thresholds at the same time, the order processing plan and production processing plan will be automatically executed.

8. The digital logistics supply system for a park according to claim 6 is characterized in that: The strategy combination includes: The inventory strategy portfolio includes building a market demand forecast model based on historical sales data, market trends, and unexpected events. Based on the output of the market demand forecast model, a cross-regional inventory allocation plan is generated to allocate excess inventory to high-demand areas. The inventory is managed differently by combining the ABC classification method with the product life cycle stage. The order strategy combination includes generating picking routes based on the ant colony algorithm to optimize sorting time for orders with long fulfillment times. If the current order volume is less than 40% of the average over the past cycle time, and the proportion of orders from large customers exceeds 60%, it is judged as a small order quantity. For small order quantities, based on the threshold exceeded by the ZHS, the business department is issued a business instruction to increase the corresponding order and conduct tiered promotion activities based on inventory products. The production strategy combination includes: if excessive production efficiency leads to overcapacity, the production efficiency of the relevant products will be reduced to 50% of the original level, and the production quantity of the relevant products will be reduced; excessive production efficiency means that the production efficiency exceeds the preset threshold multiple times within the cycle time; overcapacity means that the inventory level exceeds 120% of the safety stock; If material consumption does not match, digital twin technology is used to self-correct process parameters; The order processing solution includes optimizing production scheduling through order priority. Order priority is obtained by comprehensively analyzing order value, customer level, contract constraints, delivery time, and strategic significance. After normalization, the priority score is substituted into the formula YXJ = t1·DZ + t2·KH + t3·HT + t4·JH + t5·ZL, where DZ is the order value, KH is the customer level, HT is the contract constraint, JH is the delivery time, and ZL is the strategic significance. t1, t2, t3, t4, and t5 are the weighting factors of order value, customer level, contract constraints, delivery time, and strategic significance, respectively, and t1+t2+t3+t4+t5=1. Priority sorting is performed based on the priority score, and the priority is recalculated every n hours to trigger a production scheduling update. Based on order priority, high-priority orders are produced first, and inventory is dispatched from nearby warehouses to meet order demand.

9. The digital logistics supply system for a park according to claim 7, characterized in that: The treatment plan includes: The order processing solution includes optimizing production scheduling through order priority. Order priority is obtained by comprehensively analyzing order value, customer level, contract constraints, delivery time, and strategic significance. After normalization, the priority score is substituted into the formula YXJ = t1·DZ + t2·KH + t3·HT + t4·JH + t5·ZL, where DZ is the order value, KH is the customer level, HT is the contract constraint, JH is the delivery time, and ZL is the strategic significance. t1, t2, t3, t4, and t5 are the weighting factors of order value, customer level, contract constraints, delivery time, and strategic significance, respectively, and t1+t2+t3+t4+t5=1. Priority sorting is performed based on the priority score, and the priority is recalculated every n hours to trigger a production scheduling update. Based on order priority, high-priority orders are produced first, and inventory is dispatched from nearby warehouses to meet order demand; The production processing solution includes comparing the equipment operation data collected in real time by deployed IoT sensors with the preset operating parameter thresholds. When deviations in the operating data are detected, adjustment instructions are automatically issued. The production shift was adjusted from the standard two-shift system to a three-shift system; Resource scheduling recommendations include optimizing the raw material supply chain, shortening the raw material procurement cycle, increasing raw material inventory, activating emergency supply channels, and adjusting the supplier structure; When the material consumption matching degree exceeds the preset threshold, the raw material inventory is detected to be lower than the safety stock threshold and the real-time consumption is higher than the inventory replenishment. It is judged that the raw materials in stock are insufficient, and the raw material inventory is dispatched from the nearby warehouse to ensure the continuity of production.

Images