Logistics and warehousing system optimization method based on multi-agent system

Abstract

The invention relates to the technical field of logistics and warehousing, in particular to a logistics and warehousing system optimization method based on a multi-agent system, which is specifically applied to collaborative optimization and dynamic task allocation among logistics nodes and can be widely applied to full-automatic logistics and warehousing systems. By integrating a full-automatic packaging line, an intelligent logistics management system (LMS), an automatic warehousing system (AS / RS) and a collaborative optimization algorithm based on a multi-agent system (MAS), an efficient solution is provided for logistics and warehousing links in the production process of the medicinal capsules.

Description

Technical Field

[0001] The present invention relates to the field of logistics and warehousing technology, and in particular to a logistics and warehousing system optimization method based on a multi-agent system, which is specifically applied to collaborative optimization and dynamic task allocation between logistics nodes and can be widely used in fully automated logistics and warehousing systems. Background Art

[0002] In the production and manufacturing process of medicinal capsules, the individual automation technology of each process is relatively mature. For example, key processes such as gelatin dissolution, capsule mold forming, drying and shearing can all be efficiently automated. However, the transportation of materials between nodes involved in the entire production process, the transmission of production information, and the final packaging and warehousing still require a lot of manual intervention. This model not only increases the complexity of manual operations, but may also lead to an increase in error rates, while also limiting the improvement of production efficiency.

[0003] Taking the packaging and storage of medicinal capsules as an example, the traditional method usually uses manual transportation or semi-automatic equipment to complete the transfer of capsules from the production line to the packaging process. This operation method is easily affected by human factors, and there are problems such as unreasonable logistics paths, low packaging efficiency, and waste of resources. In addition, in the warehousing process, the storage, sorting, and delivery scheduling of capsule products often rely on manual operations, which not only consumes a lot of human resources, but also easily leads to overload of logistics nodes due to improper task scheduling, thereby reducing the overall system efficiency.

[0004] Although the existing logistics and warehousing systems have tried to introduce some automated equipment, such as shuttles, automated conveying devices, and warehousing robots, the following problems still exist in terms of resource optimization and dynamic scheduling between nodes: 1. Insufficient coordination between nodes: Task allocation between logistics nodes usually adopts static rules and cannot respond to dynamic changes in the production line in real time.

[0005] 2. Low resource utilization: Some nodes are overloaded due to task backlogs, while other nodes have idle resources, resulting in uneven equipment utilization.

[0006] Lack of optimization in logistics route planning: During the packaging and storage process of multi-batch and multi-category medicinal capsules, there is a lack of effective route planning mechanism, which increases logistics time and cost.

[0007] In view of the above problems, a fully automatic logistics and warehousing solution that can integrate intelligent logistics management system (LMS) and automated warehousing system (AS / RS) in the production line of medicinal capsules, especially the use of intelligent algorithms for dynamic task allocation and resource optimization, will significantly improve the automation level and operation efficiency of the system. This is of great significance to reducing labor costs, improving the production efficiency of medicinal capsules, and improving the reliability of logistics and warehousing. Summary of the invention

[0008] The present invention aims to provide a fully automatic logistics and warehousing system suitable for a medicinal capsule production line, in particular, a logistics node collaborative optimization method based on a multi-agent system, which dynamically allocates tasks and optimizes resources in the logistics and warehousing links through intelligent algorithms, thereby solving the problems of excessive manual intervention, inflexible scheduling, and low resource utilization in the prior art.

[0009] The technical solution of the logistics and warehousing system optimization method based on a multi-agent system of the present invention is as follows: Establishing a multi-agent system for logistics nodes: the multi-agent system realizes resource monitoring and task allocation among logistics nodes through agent modules; Build a resource sharing model between logistics nodes: Build a resource sharing model between logistics nodes based on real-time data; Use collaborative optimization algorithm to dynamically adjust task allocation: Use collaborative optimization algorithm to dynamically adjust the allocation and execution order of logistics tasks according to the resource sharing model; The optimization plan is transmitted to the system and executed: the optimized task allocation plan is fed back to the intelligent logistics management system LMS and the automated warehousing system AS / RS, through which logistics path optimization and task scheduling are performed.

[0010] Preferably, the agent module of the multi-agent system includes a task monitoring module and a resource allocation module, which are used to monitor the task load and available resources of the logistics node in real time.

[0011] Preferably, the resource sharing model is constructed by the following steps: Collect resource status data of logistics nodes, including storage capacity, equipment utilization, and task queue length; Normalize resource status data to generate a standardized node resource matrix; An initial optimization model for the task allocation scheme is established based on the node resource matrix.

[0012] Preferably, the normalized formula of the resource sharing model is: , in: is the normalized value of the i-th node on the j-th resource; is the original resource value; and are the maximum and minimum values ​​of resource j respectively.

[0013] Preferably, the collaborative optimization algorithm comprises the following steps: Construct the objective function: , in is the task completion time, is the resource utilization rate, is the transmission delay; , , is the weight coefficient of the objective function.

[0014] Preferably, the objective function is solved by a genetic algorithm, which specifically includes the following steps: Initialize the population and randomly generate multiple task allocation schemes; Calculate the population fitness value according to the objective function; The optimization is iteratively performed through selection, crossover, and mutation operations until the convergence condition is met.

[0015] Preferably, the population fitness value calculation formula is: , Where f(x) is the objective function value.

[0016] Preferably, the weight coefficient of the optimization algorithm is dynamically adjusted, and the dynamic adjustment formula is: , = constant, in and are the maximum and minimum task completion time respectively.

[0017] Preferably, the dynamic priority of the resource sharing model is determined by the following formula: , in: represents the priority of node i; , are the available resources and total resources of node i respectively; Indicates the task completion time.

[0018] Preferably, the optimized task allocation plan is directly transmitted to the logistics node through the control module to guide the shuttle vehicle or sorting robot to perform the task.

[0019] The present invention proposes an efficient solution for the logistics and warehousing links in the production process of medicinal capsules by integrating a fully automatic packaging line, an intelligent logistics management system (LMS), an automated warehousing system (AS / RS) and a collaborative optimization algorithm based on a multi-agent system (MAS). The technical effects are specifically reflected in the following aspects: Improve the automation of logistics operations: The present invention introduces an agent module between logistics nodes and combines it with an intelligent algorithm to realize automated logistics task allocation and dynamic scheduling for the entire process from packaging to warehousing, significantly reducing the need for manual intervention; the automated packaging line completes operations such as unpacking, packing, labeling and packaging of medicinal capsules, transforming traditional manual processes into a fully automated production mode, thereby improving production efficiency.

[0020] Optimize resource utilization: Through the multi-agent system, the task status and resource usage of logistics nodes are dynamically monitored, a resource sharing model is built, and the collaborative optimization algorithm is used to reasonably distribute the task load to avoid node overload or resource idleness. In the warehousing link, the storage layout is optimized through intelligent scheduling, the utilization rate of shuttle vehicles is improved, the length of logistics routes is reduced, and the comprehensive utilization rate of equipment and storage capacity is improved.

[0021] Improve logistics efficiency and accuracy: The task allocation and path planning mechanism based on collaborative optimization algorithm can reduce unnecessary waiting time and logistics bottleneck problems in the logistics process and shorten the task completion cycle; through real-time data analysis and scheduling optimization, the logistics error rate caused by task conflicts or improper path planning can be significantly reduced.

[0022] Improve system adaptability and flexibility: The optimization algorithm of the present invention can respond to the dynamic changes of the medicinal capsule production line in real time, and quickly adapt to the production needs of different batches and categories by dynamically adjusting the weight coefficient and priority allocation; the system has strong fault tolerance and can effectively deal with logistics node anomalies or equipment failures to ensure the stable operation of the logistics system.

[0023] Reduce logistics and warehousing costs: Replacing manual operations with automated equipment significantly reduces labor costs; optimized logistics route planning and dynamic task scheduling mechanisms reduce energy consumption and logistics operating costs; integrating LMS and AS / RS systems to achieve logistics information management and reduce error costs caused by manual records.

[0024] Wide applicability: The present invention is not only applicable to the packaging and warehousing links of the medicinal capsule production line, but can also be extended to other production and manufacturing fields with high automation requirements, such as food processing, pharmaceutical packaging, and electronic product assembly. The present invention solves the problems of insufficient coordination of logistics nodes, uneven task distribution, and low efficiency in the production process of medicinal capsules by systematically integrating automated equipment, intelligent algorithms, and logistics management systems, significantly improving the overall performance of the logistics and warehousing system, and bringing greater economic benefits and technical competitiveness to medicinal capsule manufacturers. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a structural block diagram of the present invention. DETAILED DESCRIPTION

[0026] Embodiment 1: The technical solution of the logistics and warehousing system optimization method based on a multi-agent system in this embodiment is as follows: The following steps are involved: Establishing a multi-agent system for logistics nodes, wherein the multi-agent system realizes resource monitoring and task allocation among logistics nodes through agent modules; Build a resource sharing model between logistics nodes based on real-time data; Adopt collaborative optimization algorithms to dynamically adjust the allocation and execution order of logistics tasks based on the resource sharing model; The optimized task allocation plan is fed back to the intelligent logistics management system (LMS) and automated warehousing system (AS / RS), through which logistics path optimization and task scheduling are performed.

[0027] Embodiment 2: The difference between this embodiment and embodiment 1 is that this embodiment further includes: The resource sharing model is constructed by the following steps: Collect resource status data of logistics nodes, including storage capacity, equipment utilization, and task queue length; Normalize resource status data to generate a standardized node resource matrix; An initial optimization model for the task allocation scheme is established based on the node resource matrix.

[0028] Preferably, the normalized formula of the resource sharing model is: ,

[0029] in: is the normalized value of the i-th node on the j-th resource; is the original resource value; and are the maximum and minimum values ​​of resource j respectively.

[0030] Preferably, the collaborative optimization algorithm comprises the following steps: Construct the objective function:

[0031] , in is the task completion time, is the resource utilization rate, is the transmission delay; , , is the weight coefficient of the objective function.

[0032] Preferably, the objective function is solved by a genetic algorithm, which specifically includes the following steps: Initialize the population and randomly generate multiple task allocation schemes; Calculate the fitness value according to the objective function; The optimization is iteratively performed through selection, crossover, and mutation operations until the convergence condition is met.

[0033] Preferably, the population fitness calculation formula is: , Where f(x) is the objective function value.

[0034] Preferably, the weight coefficient of the optimization algorithm is dynamically adjusted, and the dynamic adjustment formula is: , = constant, in and are the maximum and minimum task completion time respectively.

[0035] Preferably, the dynamic priority of the resource sharing model is determined by the following formula: , in: represents the priority of node i; , are the available resources and total resources of node i respectively; Indicates the task completion time.

[0036] Preferably, the optimized task allocation plan is directly transmitted to the logistics node through the control module to guide the shuttle vehicle or sorting robot to perform the task.

[0037] Embodiment 3: This embodiment is an example of the application field of medicinal capsules: Background A pharmaceutical capsule manufacturer needs to process 100,000 packaging units per day, involving the following logistics nodes: Node 1: Packaging line: responsible for capsule packing, labeling and packaging tasks; Node 2: Logistics transfer line: transfer materials through conveyor belts; Node 3: Automated storage system (AS / RS): including shuttles and shelves; Node 4: Outbound site: completes sorting and delivery.

[0038] In the traditional model, the fixed task allocation rules and lack of dynamic adjustment lead to low resource utilization and unreasonable logistics path planning. By introducing the present invention, the dynamic priority calculation formula and collaborative optimization algorithm are adopted, combined with the multi-agent system, to achieve intelligent optimization of task allocation and path planning.

[0039] Implementation steps Step 1: Establish a multi-agent system for logistics nodes Configure agent modules for each logistics node, including: Task monitoring module: real-time monitoring of task queue length, device load and resource status; Resource allocation module: provides the collaborative optimization algorithm with the task and resource status of the current node.

[0040] Example: Packaging line (node ​​1): the current task queue length is 50 units, and the average task completion time is 20 minutes; Logistics transfer line (node ​​2): conveyor belt load rate is 70%; Warehousing system (Node 3): The shelf storage occupancy rate is 70%, and the shuttle task queue length is 15 times / hour.

[0041] Step 2: Build a resource sharing model between logistics nodes Data collection: Collect resource status data of all logistics nodes, including: Available resources and total resources ; Task completion time .

[0042] Dynamic priority calculation: Use the formula to calculate the task priority of each node: , Weight coefficient: =0.6, =0.4.

[0043] Node data: Node 1 (Packaging Line): =40%, =15 seconds; Node 2 (Logistics transfer line): =20%, =25 seconds; Node 3 (warehouse system): =70%, =40 seconds.

[0044] Calculation priority: Node 1: , Node 2: , Node 3: .

[0045] Prioritization: Node 3 (warehousing system) has the highest priority, followed by node 1 (packaging line), and node 2 (logistics transfer line) has the lowest priority.

[0046] Step 3: Use collaborative optimization algorithms to dynamically adjust task allocation Objective function construction: Use the following formula to optimize task allocation: ,

[0047] α=0.4,β=0.4,γ=0.2; : Task completion time, : Resource utilization, : Length of logistics path.

[0048] Optimization process: Initialize the population: generate multiple sets of task allocation schemes; Calculate fitness: .

[0049] Through iterative optimization of genetic algorithms, the task allocation plan is finally output.

[0050] Step 4: Transfer the optimization plan to the system and execute it Packaging line: tasks are evenly distributed, each packaging machine completes the same amount of tasks, and the task completion time is synchronized; Logistics transfer line: optimize route planning and shorten the single transport route to 15 meters; Warehousing system: Prioritize high-priority tasks and reduce shuttle task queue length by 30%.

[0051] Implementation effect: project Traditional Model Implementation of the present invention Improvement rate Total task processing time 6 hours 3 hours 50% Warehouse utilization 60% 90% 30% Task assignment error rate 5% 1% 80% Manual operation requirements 10 people 2 people 80%

[0052] Improved efficiency: Task completion time reduced from 6 hours to 3 hours; Resource optimization: warehouse system utilization increased by 30%, and shuttle task queues reduced by 20%; Labor saving: operators are reduced by 80%; Improved accuracy: Task error rate reduced from 5% to 1%.

[0053] Summarize In this comprehensive embodiment, dynamic priority calculation and collaborative optimization algorithm are fully combined to form a closed-loop optimization system from node monitoring to task execution. All steps are reflected in a unified process, and their practical application effects are demonstrated through numerical examples.

Claims

1. A logistics and warehousing system optimization method based on a multi-agent system, characterized in that: The following steps are involved: Establishing a multi-agent system for logistics nodes: the multi-agent system realizes resource monitoring and task allocation among logistics nodes through agent modules; Build a resource sharing model between logistics nodes: Build a resource sharing model between logistics nodes based on real-time data; Use collaborative optimization algorithm to dynamically adjust task allocation: Use collaborative optimization algorithm to dynamically adjust the allocation and execution order of logistics tasks according to the resource sharing model; The optimization plan is transmitted to the system and executed: the optimized task allocation plan is fed back to the intelligent logistics management system LMS and the automated warehousing system AS / RS, through which logistics path optimization and task scheduling are performed.

2. The method for optimizing logistics and warehousing system based on multi-agent system according to claim 1, characterized in that: The agent module of the multi-agent system includes a task monitoring module and a resource allocation module, which are used to monitor the task load and available resources of the logistics node in real time.

3. The method for optimizing logistics and warehousing system based on multi-agent system according to claim 2, characterized in that: The resource sharing model is constructed by the following steps: Collect resource status data of logistics nodes, including storage capacity, equipment utilization, and task queue length; Normalize resource status data to generate a standardized node resource matrix; An initial optimization model for the task allocation scheme is established based on the node resource matrix.

4. The method for optimizing logistics and warehousing system based on multi-agent system according to claim 3 is characterized in that: The normalized processing formula of the resource sharing model is: ， in: is the normalized value of the i-th node on the j-th resource; is the original resource value; and are the maximum and minimum values ​​of resource j respectively.

5. The method for optimizing logistics and warehousing system based on multi-agent system according to claim 1, characterized in that: The collaborative optimization algorithm comprises the following steps: Construct the objective function: ， in is the task completion time, is the resource utilization rate, is the transmission delay; , , is the weight coefficient of the objective function.

6. The method for optimizing logistics and warehousing system based on multi-agent system according to claim 5, characterized in that: The objective function is solved by a genetic algorithm, which specifically includes the following steps: Initialize the population and randomly generate more than one task allocation scheme; Calculate the population fitness value according to the objective function; The optimization is iteratively performed through selection, crossover, and mutation operations until the convergence condition is met.

7. The method for optimizing logistics and warehousing system based on multi-agent system according to claim 6, characterized in that: The population fitness value calculation formula is: ， Where f(x) is the objective function value.

8. The method for optimizing logistics and warehousing system based on multi-agent system according to claim 1, characterized in that: The weight coefficient of the optimization algorithm is dynamically adjusted, and the dynamic adjustment formula is: ， ， = constant, in and are the maximum and minimum task completion time respectively.

9. The method for optimizing logistics and warehousing system based on multi-agent system according to claim 1, characterized in that: The dynamic priority of the resource sharing model is determined by the following formula: , in: represents the priority of node i; , are the available resources and total resources of node i respectively; Indicates the task completion time.

10. The method for optimizing logistics and warehousing system based on multi-agent system according to claim 1, characterized in that: The optimized task allocation plan is directly transmitted to the logistics node through the control module to guide the shuttle vehicle or sorting robot to perform tasks.

Images