Supply chain logistics real-time scheduling system and method

Abstract

The invention discloses a supply chain logistics real-time scheduling system and method, and relates to the technical field of production scheduling, and the system specifically comprises a production plan dynamic adjustment module, a digital twin simulation module, a dynamic response and optimization module, a space-time coupling resource scheduling model, a cross-enterprise cooperation module, and a visualization and decision support module. PLTI instructions are dynamically generated through digital twinborn simulation and a dynamic response mechanism, in-plant and out-plant real-time collaboration and in-plant and out-plant combined modeling are achieved, the inventory turnover rate is increased by 25%, the inventory turnover rate is increased through a space-time coupling resource scheduling model, an in-plant storage model and an Amap API road network are constructed through a Unity 3D engine, instruction conflict probability is rehearsed, and the real-time performance of the system is improved. According to the method, logistics scheduling information is dynamically updated every 15 minutes based on a production plan, the logistics scheduling conflict probability is reduced by 83% through digital twin rehearsal, and real-time collaboration and intelligent scheduling of logistics inside and outside a factory are achieved through digital twin simulation, a dynamic response mechanism and a space-time coupling resource scheduling model.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of production scheduling, in particular to a supply chain logistics real-time scheduling system and method. BACKGROUND

[0002] Under the background of Industry 4.0 and intelligent manufacturing, supply chain logistics is rapidly developing, and supply chain logistics is a strategic function for the smooth realization of logistics related to economic activities, coordination of production, supply activities, sales activities and logistics activities, and comprehensive management. However, the supply chain logistics faces the following core challenges: lack of dynamics: production plan is frequently adjusted, such as single insertion, equipment failure, demand fluctuation, traditional static logistics plan cannot quickly respond, such as fixed 4-hour scheduling, lack of coordination: lack of real-time coordination between in-plant logistics and out-of-plant logistics, leading to inventory accumulation and production line downtime, lack of intelligence: path planning relies on human experience, AGV empty running rate reaches 20%, lack of digital twin pre-play capability, high risk of equipment conflict, Chinese patent discloses a kind of advanced scheduling method based on production plan, application number is: CN202311607957.8, the present application can reduce production scheduling time, reduce production and marketing coordination meeting time, provide immediate rearrangement function when customer inserts single or order changes, control capacity and material, and make complex manufacturing management transparent, which can predict future capacity and long-term material procurement;

[0003] However, in the prior art, only the internal resources of the factory are modeled, and the external nodes such as supplier capacity, logistics service provider capacity, distribution center inventory, etc. are not digitally mapped, resulting in poor coordination outside the factory. In the prior art, there is no real-time coordination protocol, and the warning information is transmitted manually. No automatic negotiation mechanism is established, and in the prior art, there is no dynamic knowledge graph, and the self-evolution ability is lacking. The experience accumulation relies on manual maintenance. SUMMARY

[0004] The present application provides a supply chain logistics real-time scheduling system and method, which can effectively solve the problem of lack of digital mapping of external nodes such as supplier capacity, logistics service provider capacity, distribution center inventory, etc. in the prior art, resulting in poor coordination outside the factory. In the prior art, there is no real-time coordination protocol, and the warning information is transmitted manually. No automatic negotiation mechanism is established, and in the prior art, there is no dynamic knowledge graph, and the self-evolution ability is lacking. The experience accumulation relies on manual maintenance.

[0005] To achieve the above purpose, the present application provides the following technical scheme: a supply chain logistics real-time scheduling system, which realizes real-time coordination of in-plant and out-of-plant logistics in a dynamic environment, optimizes logistics scheduling through intelligent means, reduces the risk of inventory accumulation and production line downtime, and solves the real-time scheduling problem of supply chain logistics in dynamic production planning and multi-source data coordination.

[0006] Specifically, it includes production plan dynamic adjustment module, digital twin simulation module, dynamic response and optimization module, space-time coupling resource scheduling model, cross-enterprise collaboration module and visualization and decision support module.

[0007] According to the above technical solution, the production plan dynamic adjustment module dynamically adjusts the production plan according to the failure of the order equipment, perceives the changes in the production environment in real time, dynamically adjusts the production schedule, and provides accurate instructions for downstream logistics, including an input interface, a dynamic adjustment engine and an output interface;

[0008] The input interface accesses data on market demand changes, equipment status, and raw material supply fluctuations in real time. Market demand changes include order increases and decreases, and urgent orders; equipment status includes equipment failures and maintenance warnings; and raw material supply fluctuations include raw material inventory and delivery delays.

[0009] The dynamic adjustment engine uses a machine learning model to analyze historical order and equipment data to predict short-term demand trends and generate a schedulable production plan based on constraints, including production capacity, inventory capacity, and equipment availability.

[0010] The output interface sends an updated production scheduling instruction to the downstream logistics module, and the production scheduling instruction is the updated product order production time.

[0011] According to the above technical solution, the digital twin simulation module uses the Unity3D engine to build a logistics supply chain warehousing model inside and outside the factory, combined with the Amap API road network simulation, to preview the probability of command conflicts, including a three-dimensional simulation environment and conflict preview functions;

[0012] The three-dimensional simulation environment uses the Unity3D engine to build a warehouse layout model within the factory, including shelves, AGV paths, and loading and unloading areas;

[0013] At the same time, the AutoNavi Map API is integrated to simulate the off-site road network, traffic conditions, and the geographic location of supply customers;

[0014] The conflict rehearsal function simulates potential conflicts between different logistics instructions, where logistics instructions include AGV scheduling and transportation routes, and potential conflicts include path intersection and equipment occupancy.

[0015] According to the above technical solution, the dynamic response and optimization module is based on the rolling horizon optimization algorithm RHO and the Markov decision process MDP. It dynamically generates pull logistics timing instructions PLTI every 15 minutes, adjusts logistics instructions in real time, balances resources inside and outside the factory, and suppresses demand fluctuations.

[0016] Specifically, the rolling horizon optimization algorithm (RHO) uses a 15-minute time window to dynamically generate pull-type logistics timing instructions, and dynamically adjusts the logistics instructions (PLTI) to adapt to real-time changes in production plans. Combined with the Markov decision process (MDP), it predicts future logistics demand through state transition probabilities and optimizes the decision sequence.

[0017] According to the above technical solution, the spatiotemporal coupled resource scheduling model jointly optimizes the in-factory storage allocation and the out-factory transportation path VRP. The in-factory joint optimization uses a clustering algorithm to classify and store goods to reduce the AGV pickup distance. The out-factory joint optimization uses the vehicle routing problem (VRP) algorithm to plan the shortest path from the supplier to the factory.

[0018] By adopting a two-stage robust optimization approach, we can suppress the bullwhip effect caused by demand fluctuations and improve supply chain resilience. The first stage determines the baseline plan, and the second stage adjusts to cope with fluctuations. Through a spatiotemporal coupled resource scheduling model, we can reduce transportation costs and improve inventory turnover.

[0019] According to the above technical solution, the cross-enterprise collaboration module blockchain records performance data, triggers exception processing, and realizes data sharing and exception processing automation in the upstream and downstream of the supply chain, including blockchain evidence storage, smart contracts and real-time collaboration protocols;

[0020] The blockchain evidence storage records logistics performance data through the blockchain to ensure that the data cannot be tampered with, and supports supply chain traceability and responsibility tracing. Logistics performance data includes cargo delivery time and quality inspection results;

[0021] The smart contract automatically triggers exception handling, including activation of alternative plans and compensation clauses when suppliers delay delivery;

[0022] The real-time collaboration protocol uses the MQTT protocol to achieve real-time data exchange between systems inside and outside the factory, eliminating information islands and reducing data synchronization delays.

[0023] According to the above technical solution, the visualization and decision support module previews instructions and provides real-time data dashboards and intelligent early warnings to assist managers in decision-making;

[0024] The digital twin large screen displays the real-time logistics status inside and outside the factory, including AGV location, vehicle trajectory, and inventory levels. It also supports multi-dimensional data analysis, including equipment utilization and route transportation efficiency.

[0025] Intelligent early warning is to prompt potential risks by setting threshold alarms. Potential risks include inventory below the safety value and AGV failure rate exceeding the limit. When the alarm is triggered, the early warning information is transmitted to the responsible personnel through remote notification to trigger the emergency plan.

[0026] A real-time scheduling method for supply chain logistics includes the following steps:

[0027] Step 1: Real-time perception of production environment changes and dynamic adjustment of production schedules;

[0028] Step 2: Build a digital simulation environment that combines virtual and real elements to rehearse the probability of command conflicts;

[0029] Step 3: Dynamically adjust logistics instructions to adapt to real-time changes in production plans;

[0030] Step 4: Jointly optimize the allocation of storage space within the factory and the transportation routes outside the factory to reduce the bullwhip effect;

[0031] Step 5: Realize data sharing and exception handling automation across the supply chain;

[0032] Step six: Provide three-dimensional visualization of real-time data and intelligent early warning to assist decision-making.

[0033] According to the above technical solution, in step 1, the production plan is dynamically adjusted according to the failure of the order equipment, the changes in the production environment are perceived in real time, the production schedule is dynamically adjusted, and accurate instructions are provided to the downstream logistics. The data of market demand changes, equipment status, and raw material supply fluctuations are accessed in real time. The short-term demand trend is predicted based on the machine learning model, and a schedulable production plan is generated based on the constraints. The updated production scheduling instructions are issued to the downstream logistics module.

[0034] In step 2, a warehousing model for the in-plant and out-plant logistics supply chain is constructed using the Unity3D engine. This model is combined with the AutoNavi Map API road network simulation to simulate the probability of order conflicts. The Unity3D engine is used to construct an in-plant warehousing layout model, which is integrated with the AutoNavi Map API to simulate the out-plant road network, traffic conditions, and geographic locations of supply customers. This simulation is used to deduce potential conflicts between different logistics orders.

[0035] In step three, based on the rolling horizon optimization algorithm RHO and the Markov decision process MDP, pull logistics timing instructions PLTI are dynamically generated every 15 minutes, logistics instructions are adjusted in real time, resources inside and outside the factory are balanced, demand fluctuations are suppressed, future logistics demand is predicted through state transition probability, and the decision sequence is optimized.

[0036] According to the above technical solution, step 4 involves jointly optimizing the in-plant storage allocation and the off-plant transportation route (VRP), classifying and storing goods using a clustering algorithm to reduce the AGV pickup distance, using the vehicle routing problem (VRP) algorithm to plan the shortest path from the supplier to the factory, and using a two-stage robust optimization to suppress the bullwhip effect caused by demand fluctuations and improve supply chain resilience. Furthermore, a spatiotemporal coupled resource scheduling model is used to reduce transportation costs and improve inventory turnover.

[0037] In step five, the blockchain records contract performance data, triggers exception handling, and implements data sharing and exception handling automation across the supply chain. The blockchain records logistics contract performance data to ensure that the data cannot be tampered with. The smart contract automatically triggers exception handling and implements real-time data exchange between internal and external systems through the MQTT protocol, eliminating information silos and reducing data synchronization delays.

[0038] The aforementioned step six is ​​feasible through rehearsal instructions, providing real-time data dashboards and intelligent early warnings to assist managers in decision-making, using digital twin large screens to display the logistics status inside and outside the factory in real time, supporting multi-dimensional data analysis, and setting threshold alarms to prompt potential risks. The early warning information is transmitted to the responsible personnel through remote notification to trigger the emergency plan.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] 1. Dynamically generate PLTI instructions through digital twin simulation and a dynamic response mechanism, achieving real-time collaboration inside and outside the factory. Joint modeling inside and outside the factory increases inventory turnover by 25%. Inventory turnover is improved through a spatiotemporal coupling resource scheduling model. The Unity3D engine is used to build an in-factory warehousing model + AutoNavi Map API road network simulation, and the probability of instruction conflicts is previewed with an accuracy of ±0.5 seconds. Logistics scheduling information is dynamically updated every 15 minutes based on the production plan. Digital twin previews reduce the probability of logistics scheduling conflicts by 83%. Through digital twin simulation, a dynamic response mechanism, and a spatiotemporal coupling resource scheduling model, real-time collaboration and intelligent scheduling of logistics inside and outside the factory are achieved. This has important application value in improving inventory turnover and reducing the probability of logistics scheduling conflicts, providing an efficient solution for intelligent manufacturing in the Industry 4.0 era.

[0041] 2. By sensing changes in the production environment in real time, dynamically adjusting production schedules, providing precise instructions for downstream logistics, and achieving minute-level responses through machine learning, production flexibility is significantly improved. A virtual-reality simulation environment is constructed to rehearse conflicts in logistics instructions and optimize routes. Risks are discovered in advance through simulation to reduce conflict rates, adjust logistics instructions in real time, balance resources inside and outside the factory, and suppress demand fluctuations. Through joint optimization, inventory turnover rates are improved, and data sharing and exception handling automation are achieved upstream and downstream of the supply chain. Through blockchain + MQTT, full-link automated collaboration is achieved, exception response time is shortened, and decision-making time is compressed from hours to minutes through large screens and early warnings. Real-time data dashboards and intelligent early warnings are provided to assist managers in decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0043] In the attached figure:

[0044] Figure 1 It is a structural block diagram of the scheduling system of the present invention;

[0045] Figure 2 It is a schematic diagram of the composition of the scheduling system of the present invention;

[0046] Figure 3 It is a flow chart of the steps of the scheduling method of the present invention. DETAILED DESCRIPTION

[0047] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0048] Example: Figure 1-2 As shown, the present invention provides a technical solution, a real-time scheduling system for supply chain logistics, which realizes real-time coordination of logistics inside and outside the factory in a dynamic environment, optimizes logistics scheduling through intelligent means, reduces the risks of inventory backlogs and production line shutdowns, and solves the real-time scheduling problem of supply chain logistics in terms of dynamic production planning and multi-source data coordination;

[0049] Specifically, it includes production plan dynamic adjustment module, digital twin simulation module, dynamic response and optimization module, space-time coupling resource scheduling model, cross-enterprise collaboration module and visualization and decision support module.

[0050] Based on the above technical solution, the production plan dynamic adjustment module dynamically adjusts the production plan according to order equipment failures, perceives changes in the production environment in real time, dynamically adjusts the production schedule, and provides precise instructions for downstream logistics. It includes an input interface, a dynamic adjustment engine, and an output interface.

[0051] The input interface accesses data on market demand changes, equipment status, and raw material supply fluctuations in real time. Market demand changes include order increases and decreases, as well as urgent orders. Equipment status includes equipment failures and maintenance alerts. Raw material supply fluctuations include raw material inventory and delivery delays.

[0052] The dynamic adjustment engine uses machine learning models to analyze historical order and equipment data to predict short-term demand trends and generate schedulable production plans based on constraints such as production capacity, inventory capacity, and equipment availability.

[0053] The output interface sends the updated production scheduling instructions to the downstream logistics module. The production scheduling instructions are the updated product order production time.

[0054] Based on the above technical solution, the digital twin simulation module uses the Unity3D engine to build an in-plant and out-of-plant logistics supply chain warehousing model. Combined with the Amap API road network simulation, it can preview the probability of command conflicts, including a 3D simulation environment and conflict preview functions.

[0055] The 3D simulation environment uses the Unity3D engine to build a warehouse layout model within the factory, including shelves, AGV paths, and loading and unloading areas;

[0056] At the same time, the AutoNavi Map API is integrated to simulate the off-site road network, traffic conditions, and the geographic location of supply customers;

[0057] The conflict rehearsal function simulates potential conflicts among different logistics instructions, including AGV scheduling and transportation routes. Potential conflicts include path intersection and equipment occupancy.

[0058] Based on the above technical solution, the dynamic response and optimization module uses the rolling horizon optimization algorithm RHO and the Markov decision process MDP to dynamically generate pull logistics timing instructions PLTI every 15 minutes, adjust logistics instructions in real time, balance internal and external resources, and suppress demand fluctuations;

[0059] Specifically, the rolling horizon optimization algorithm (RHO) uses a 15-minute time window to dynamically generate pull-type logistics timing instructions, and dynamically adjusts the logistics instructions (PLTI) to adapt to real-time changes in production plans. Combined with the Markov decision process (MDP), it predicts future logistics demand through state transition probabilities and optimizes the decision sequence.

[0060] Based on the above technical solution, a spatiotemporal coupled resource scheduling model jointly optimizes in-factory storage allocation and out-factory transportation routes (VRP). In-factory joint optimization uses a clustering algorithm to classify and store goods, reducing the distance AGVs have to pick up goods. Out-factory joint optimization uses a vehicle routing problem (VRP) algorithm to plan the shortest path from suppliers to the factory.

[0061] By adopting a two-stage robust optimization approach, we can suppress the bullwhip effect caused by demand fluctuations and improve supply chain resilience. The first stage determines the baseline plan, and the second stage adjusts to cope with fluctuations. Through a spatiotemporal coupled resource scheduling model, we can reduce transportation costs and improve inventory turnover.

[0062] Based on the above technical solution, the cross-enterprise collaboration module blockchain records performance data, triggers exception processing, and realizes data sharing and exception processing automation in the upstream and downstream of the supply chain, including blockchain evidence storage, smart contracts and real-time collaboration protocols;

[0063] Blockchain evidence storage records logistics fulfillment data through blockchain to ensure that the data cannot be tampered with, and supports supply chain traceability and responsibility tracing. Logistics fulfillment data includes cargo delivery time and quality inspection results;

[0064] Smart contracts automatically trigger exception handling, including alternative plans and compensation clauses when suppliers delay delivery;

[0065] The real-time collaboration protocol uses the MQTT protocol to achieve real-time data exchange between systems inside and outside the factory, eliminating information silos and reducing data synchronization delays.

[0066] Based on the above technical solutions, the visualization and decision support module previews instructions, provides real-time data dashboards and intelligent early warnings, and assists managers in making decisions;

[0067] The digital twin large screen displays the real-time logistics status inside and outside the factory, including AGV location, vehicle trajectory, and inventory levels. It also supports multi-dimensional data analysis, including equipment utilization and route transportation efficiency.

[0068] Intelligent early warning is to prompt potential risks by setting threshold alarms. Potential risks include inventory below the safety value and AGV failure rate exceeding the limit. When the alarm is triggered, the early warning information is transmitted to the responsible personnel through remote notification to trigger the emergency plan.

[0069] like Figure 2 As shown, a real-time scheduling method for supply chain logistics includes the following steps:

[0070] Step 1: Real-time perception of production environment changes and dynamic adjustment of production schedules;

[0071] Step 2: Build a digital simulation environment that combines virtual and real elements to rehearse the probability of command conflicts;

[0072] Step 3: Dynamically adjust logistics instructions to adapt to real-time changes in production plans;

[0073] Step 4: Jointly optimize the allocation of storage space within the factory and the transportation routes outside the factory to reduce the bullwhip effect;

[0074] Step 5: Realize data sharing and exception handling automation across the supply chain;

[0075] Step six: Provide three-dimensional visualization of real-time data and intelligent early warning to assist decision-making.

[0076] Based on the above technical solution, step one dynamically adjusts the production plan based on order equipment failures, perceives changes in the production environment in real time, dynamically adjusts the production schedule, and provides precise instructions to downstream logistics. It also accesses real-time data on market demand changes, equipment status, and raw material supply fluctuations. It uses machine learning models to predict short-term demand trends, combines constraints to generate a schedulable production plan, and issues updated production scheduling instructions to downstream logistics modules.

[0077] Step 2: Use the Unity3D engine to build an in-plant logistics supply chain warehousing model. Combined with AutoNavi Maps API road network simulation, the probability of order conflicts is predicted. The Unity3D engine is used to build an in-plant warehousing layout model. The AutoNavi Maps API is integrated to simulate the off-plant road network, traffic conditions, and the geographic location of supply customers. This simulation is used to deduce potential conflicts between different logistics orders.

[0078] Step three: Based on the rolling horizon optimization algorithm (RHO) and the Markov decision process (MDP), pull logistics timing instructions (PLTI) are dynamically generated every 15 minutes. Logistics instructions are adjusted in real time to balance resources inside and outside the factory, suppress demand fluctuations, predict future logistics demand through state transition probabilities, and optimize the decision sequence.

[0079] Based on the above technical solution, in step four, by jointly optimizing the allocation of in-factory storage space and the vehicle routing problem (VRP) of transportation routes outside the factory, a clustering algorithm is used to classify and store goods, reducing the distance for AGVs to pick up goods. The vehicle routing problem (VRP) algorithm is used to plan the shortest path from suppliers to the factory. A two-stage robust optimization is used to suppress the bullwhip effect caused by demand fluctuations and improve supply chain resilience. A spatiotemporal coupled resource scheduling model is also used to reduce transportation costs and improve inventory turnover.

[0080] Step 5: Record fulfillment data through blockchain, trigger exception handling, and achieve data sharing and automated exception handling across the supply chain. Blockchain records logistics fulfillment data to ensure that the data cannot be tampered with. Smart contracts automatically trigger exception handling, and use the MQTT protocol to achieve real-time data exchange between internal and external systems, eliminating information silos and reducing data synchronization delays.

[0081] Step six: Through rehearsal instructions, real-time data dashboards and intelligent early warnings are provided to assist managers in decision-making. Digital twin screens are used to display the logistics status inside and outside the factory in real time, support multi-dimensional data analysis, and set threshold alarms to prompt potential risks. Early warning information is transmitted to responsible personnel through remote notification to trigger emergency plans.

[0082] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A real-time scheduling system for supply chain logistics, characterized by: Realize real-time collaboration of in-plant and out-of-plant logistics in a dynamic environment, optimize logistics scheduling through intelligent means, reduce the risk of inventory backlogs and production line downtime, and solve the real-time scheduling problem of supply chain logistics in dynamic production planning and multi-source data collaboration; Specifically, it includes production plan dynamic adjustment module, digital twin simulation module, dynamic response and optimization module, space-time coupling resource scheduling model, cross-enterprise collaboration module and visualization and decision support module.

2. A supply chain logistics real-time scheduling system according to claim 1, characterized in that: The production plan dynamic adjustment module dynamically adjusts the production plan according to the order equipment failure, perceives the changes in the production environment in real time, dynamically adjusts the production schedule, and provides accurate instructions for downstream logistics. It includes an input interface, a dynamic adjustment engine, and an output interface; The input interface accesses data on market demand changes, equipment status, and raw material supply fluctuations in real time. Market demand changes include order increases and decreases, and urgent orders; equipment status includes equipment failures and maintenance warnings; and raw material supply fluctuations include raw material inventory and delivery delays. The dynamic adjustment engine uses a machine learning model to analyze historical order and equipment data to predict short-term demand trends and generate a schedulable production plan based on constraints, including production capacity, inventory capacity, and equipment availability. The output interface sends an updated production scheduling instruction to the downstream logistics module, and the production scheduling instruction is the updated product order production time.

3. A supply chain logistics real-time scheduling system according to claim 1, characterized in that: The digital twin simulation module uses the Unity3D engine to build a logistics supply chain warehousing model inside and outside the factory, combined with the Amap API road network simulation, to preview the probability of command conflicts, including a three-dimensional simulation environment and conflict preview functions; The three-dimensional simulation environment uses the Unity3D engine to build a warehouse layout model within the factory, including shelves, AGV paths, and loading and unloading areas; At the same time, the AutoNavi Map API is integrated to simulate the off-site road network, traffic conditions, and the geographic location of supply customers; The conflict rehearsal function simulates potential conflicts between different logistics instructions, where logistics instructions include AGV scheduling and transportation routes, and potential conflicts include path intersection and equipment occupancy.

4. A supply chain logistics real-time scheduling system according to claim 1, characterized in that: The dynamic response and optimization module is based on the rolling horizon optimization algorithm RHO and the Markov decision process MDP. It dynamically generates pull logistics timing instructions PLTI every 15 minutes, adjusts logistics instructions in real time, balances resources inside and outside the factory, and suppresses demand fluctuations. Specifically, the rolling horizon optimization algorithm (RHO) uses a 15-minute time window to dynamically generate pull-type logistics timing instructions, and dynamically adjusts the logistics instructions (PLTI) to adapt to real-time changes in production plans. Combined with the Markov decision process (MDP), it predicts future logistics demand through state transition probabilities and optimizes the decision sequence.

5. A supply chain logistics real-time scheduling system according to claim 1, characterized in that: The spatiotemporal coupled resource scheduling model jointly optimizes the in-factory storage allocation and the out-factory transportation path VRP. The in-factory joint optimization classifies and stores goods through a clustering algorithm to reduce the AGV pickup distance. The out-factory joint optimization uses the vehicle routing problem (VRP) algorithm to plan the shortest path from the supplier to the factory. By adopting a two-stage robust optimization approach, we can suppress the bullwhip effect caused by demand fluctuations and improve supply chain resilience. The first stage determines the baseline plan, and the second stage adjusts to cope with fluctuations. Through a spatiotemporal coupled resource scheduling model, we can reduce transportation costs and improve inventory turnover.

6. A supply chain logistics real-time scheduling system according to claim 1, characterized in that: The cross-enterprise collaboration module records contract performance data on the blockchain, triggers exception handling, and realizes data sharing and exception handling automation in the upstream and downstream of the supply chain, including blockchain evidence storage, smart contracts and real-time collaboration protocols; The blockchain evidence storage records logistics performance data through the blockchain to ensure that the data cannot be tampered with, and supports supply chain traceability and responsibility tracing. Logistics performance data includes cargo delivery time and quality inspection results; The smart contract automatically triggers exception handling, including activation of alternative plans and compensation clauses when suppliers delay delivery; The real-time collaboration protocol uses the MQTT protocol to achieve real-time data exchange between systems inside and outside the factory, eliminating information islands and reducing data synchronization delays.

7. A supply chain logistics real-time scheduling system according to claim 1, characterized in that: The visualization and decision support module previews instructions, provides real-time data dashboards and intelligent early warnings, and assists managers in making decisions; The digital twin large screen displays the real-time logistics status inside and outside the factory, including AGV location, vehicle trajectory, and inventory levels. It also supports multi-dimensional data analysis, including equipment utilization and route transportation efficiency. Intelligent early warning is to prompt potential risks by setting threshold alarms. Potential risks include inventory below the safety value and AGV failure rate exceeding the limit. When the alarm is triggered, the early warning information is transmitted to the responsible personnel through remote notification to trigger the emergency plan.

8. A real-time scheduling method for supply chain logistics, characterized by: The steps include: Step 1: Real-time perception of production environment changes and dynamic adjustment of production schedules; Step 2: Build a digital simulation environment that combines virtual and real elements to rehearse the probability of command conflicts; Step 3: Dynamically adjust logistics instructions to adapt to real-time changes in production plans; Step 4: Jointly optimize the allocation of storage space within the factory and the transportation routes outside the factory to reduce the bullwhip effect; Step 5: Realize data sharing and exception handling automation across the supply chain; Step six: Provide three-dimensional visualization of real-time data and intelligent early warning to assist decision-making.

9. A real-time scheduling method for supply chain logistics according to claim 8, characterized in that: In step 1, the production plan is dynamically adjusted based on order equipment failures, changes in the production environment are perceived in real time, and production scheduling is dynamically adjusted to provide accurate instructions for downstream logistics. Data on market demand changes, equipment status, and raw material supply fluctuations are accessed in real time. Short-term demand trends are predicted based on machine learning models, and a schedulable production plan is generated based on constraints. The updated production scheduling instructions are then issued to downstream logistics modules. In step 2, a warehousing model for the in-plant and out-plant logistics supply chain is constructed using the Unity3D engine. This model is combined with the AutoNavi Map API road network simulation to simulate the probability of order conflicts. The Unity3D engine is used to construct an in-plant warehousing layout model, which is integrated with the AutoNavi Map API to simulate the out-plant road network, traffic conditions, and geographic locations of supply customers. This simulation is used to deduce potential conflicts between different logistics orders. In step three, based on the rolling horizon optimization algorithm RHO and the Markov decision process MDP, pull logistics timing instructions PLTI are dynamically generated every 15 minutes, logistics instructions are adjusted in real time, resources inside and outside the factory are balanced, demand fluctuations are suppressed, future logistics demand is predicted through state transition probability, and the decision sequence is optimized.

10. A real-time scheduling method for supply chain logistics according to claim 8, characterized in that: In step 4, by jointly optimizing the in-plant storage allocation and the off-plant transportation route (VRP), a clustering algorithm is used to classify and store goods, reducing the AGV pickup distance. A vehicle routing problem (VRP) algorithm is used to plan the shortest path from the supplier to the factory. Two-stage robust optimization is used to suppress the bullwhip effect caused by demand fluctuations, thereby improving supply chain resilience. A spatiotemporal coupled resource scheduling model is also used to reduce transportation costs and improve inventory turnover. In step five, the blockchain records contract performance data, triggers exception handling, and implements data sharing and exception handling automation across the supply chain. The blockchain records logistics contract performance data to ensure that the data cannot be tampered with. The smart contract automatically triggers exception handling and implements real-time data exchange between internal and external systems through the MQTT protocol, eliminating information silos and reducing data synchronization delays. The aforementioned step six is ​​feasible through rehearsal instructions, providing real-time data dashboards and intelligent early warnings to assist managers in decision-making, using digital twin large screens to display the logistics status inside and outside the factory in real time, supporting multi-dimensional data analysis, and setting threshold alarms to prompt potential risks. The early warning information is transmitted to the responsible personnel through remote notification to trigger the emergency plan.

Images