Flexible intelligent manufacturing system for multi-variety small-batch production

Abstract

The invention relates to the technical field of intelligent manufacturing, in particular to a flexible intelligent manufacturing system for multi-variety small-batch production, which comprises a flexibility enhancement module, a cost efficiency improvement module, an information interaction enhancement module, a technology depth fusion module and a decision ability optimization module, rapid switching and optimization of the production process can be achieved, so that production efficiency and product quality are improved, technological parameters are intelligently adjusted according to real-time production requirements, production requirements of diversified products are met, production preparation time is effectively shortened, the production cycle is shortened, meanwhile, by means of the Internet of Things technology, the system can monitor the state of production equipment in real time, and production efficiency is improved. In addition, through data analysis and a prediction model, the system can optimize inventory management, reduce the production cost, significantly improve the production flexibility and efficiency in a multi-variety small-batch production environment, and bring significant technical effects and economic benefits to the manufacturing industry.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing technology, and in particular to a flexible intelligent manufacturing system for multi-variety small batch production. Background Art

[0002] The integration of the global market has led to increasingly fierce competition in various industries. It is difficult for companies to maintain their competitive advantage by simply relying on low-cost, large-scale production of standard products. In order to stand out in the market, companies need to respond quickly to market changes and be able to flexibly launch a variety of product styles and specifications to attract more customer groups. The multi-variety small-batch production model allows companies to achieve product upgrades in a relatively short period of time, better cope with competition in the segmented market, seize market opportunities in a timely manner, and meet the needs of consumers at different levels and with different preferences;

[0003] Once the equipment layout and process flow of traditional manufacturing production lines are determined, it is difficult to make quick adjustments. When faced with the production of new products or process changes, it often takes a lot of time and cost to re-layout and transform equipment, resulting in low production efficiency and failure to respond to the needs of multi-variety and small-batch production in a timely manner. The production units are relatively isolated, information transmission is not smooth, there are many product types, and production tasks are complex and changeable. It is difficult to make accurate and timely decisions based solely on manual experience. Therefore, a flexible intelligent manufacturing system for multi-variety and small-batch production is proposed. Summary of the invention

[0004] In view of this, the present invention provides a flexible intelligent manufacturing system for multi-variety small batch production to solve or alleviate the technical problems existing in the prior art and at least provide a beneficial option.

[0005] The technical solution of the present invention is implemented as follows: a flexible intelligent manufacturing system for multi-variety small batch production includes a flexibility enhancement module, a cost efficiency improvement module, an information interaction enhancement module, a technology deep integration module and a decision-making ability optimization module;

[0006] The flexibility enhancement module improves the ability of the manufacturing system to cope with product and process changes. Through modular design, the production line can quickly adapt to new products or process changes. With the help of the intelligent scheduling system, the equipment can efficiently switch between different tasks. It relies on universal interface standards for cross-device and cross-platform integration to enhance the flexibility of the entire system.

[0007] The cost efficiency improvement module and the on-demand production module formulate reasonable production plans based on market forecasts, reduce inventory and excess production, and control costs;

[0008] The information interaction enhancement module breaks down information silos and enhances data interaction and collaboration among manufacturing units;

[0009] The technical deep integration module collects past production process data and uses machine learning algorithms to build a correlation model between process parameters and key indicators such as product quality and production efficiency;

[0010] The decision-making capability optimization module and dynamic programming algorithm facilitate efficient scheduling and switching of multi-variety production lines.

[0011] Further preferably, the flexibility enhancement module includes a modular design module and an intelligent scheduling system module. The modular design module splits the equipment and production units into multiple standardized modules according to functions and processes. Each module has an independent and clear interface. When new products are produced or the process is changed, these modules are recombined and replaced according to new requirements. The intelligent scheduling system module uses an artificial intelligence-driven scheduling algorithm to collect attribute information of production tasks and monitor the operating status and idle conditions of the equipment in real time. Based on the above data, intelligent analysis is performed through the algorithm to dynamically allocate production tasks to equipment and production units.

[0012] Further preferably, the cost efficiency improvement module includes an on-demand production module and a shared equipment platform module. The on-demand production module uses big data analysis and market research to predict market demand, and at the same time builds a production planning model based on the product's own inventory situation, production cycle and cost structure. The shared equipment platform module builds a shared platform for centrally managing production resources, and integrates idle mobile robots and general equipment resources within an enterprise or between multiple enterprises.

[0013] Further preferably, the information interaction enhancement module includes an industrial Internet of Things integration module, an edge computing and cloud computing combination module, and a real-time monitoring and feedback module. The industrial Internet of Things integration module deploys an Internet of Things communication module on each production unit, and uniformly adopts a suitable Internet of Things communication protocol to build a network for interconnection between devices and systems. The edge computing and cloud computing combination module deploys edge computing nodes at the edge close to the production equipment, which can receive and process local data collected from sensors and equipment in real time. The real-time monitoring and feedback module arranges various sensors on the production line to form a sensor network.

[0014] Further preferably, the technology deep integration module includes an artificial intelligence application module and a collaborative robot module. The artificial intelligence application module collects production process data and uses a machine learning algorithm to construct a correlation model between process parameters and product quality and production efficiency indicators. The collaborative robot module introduces a collaborative robot equipped with force sensors, vision sensors and intelligent control systems.

[0015] Further preferably, the decision-making capability optimization module includes a real-time dynamic programming algorithm module and a multi-objective optimization module. The real-time dynamic programming algorithm module constructs the scheduling and switching problem of multi-variety production lines into a dynamic optimization problem. The multi-objective optimization module establishes a mathematical optimization model that comprehensively considers cost, efficiency, and quality, and uses a multi-objective optimization algorithm to weigh and optimize the decision variables in the production process.

[0016] Further preferably, the task allocation dynamic programming formula of the device switching cost is as follows:

[0017]

[0018] Let n be the number of production tasks, m be the number of equipment, C ij represents the cost of assigning task i to device j, T ik represents the processing time of task i on equipment j, T ik represents the minimum total cost of the first i tasks assigned to device set j, where c jk represents the switching cost coefficient between device j and device k.

[0019] The embodiment of the present invention has the following advantages due to the adoption of the above technical solution:

[0020] The present invention can realize rapid switching and optimization of the production process, thereby improving production efficiency and product quality, intelligently adjusting process parameters according to real-time production needs, adapting to the production requirements of diversified products, effectively reducing production preparation time, and shortening the production cycle. At the same time, with the help of Internet of Things technology, the system can monitor the status of production equipment in real time to ensure production continuity and stability. In addition, through data analysis and prediction models, the system can optimize inventory management, reduce production costs, and significantly improve production flexibility and efficiency in a multi-variety small batch production environment, bringing significant technical effects and economic benefits to the manufacturing industry.

[0021] The above summary is for illustrative purposes only and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments and features described above, further aspects, embodiments and features of the present invention will be readily apparent by reference to the accompanying drawings and the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0024] In the following, only some exemplary embodiments are briefly described. As those skilled in the art will appreciate, the described embodiments may be modified in various ways without departing from the spirit or scope of the present invention. Therefore, the drawings and descriptions are considered to be exemplary and non-restrictive in nature.

[0025] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0026] like Figure 1 As shown, the embodiment of the present invention provides a flexible intelligent manufacturing system for multi-variety small batch production, including a flexibility enhancement module, a cost efficiency improvement module, an information interaction enhancement module, a technology deep integration module and a decision-making ability optimization module;

[0027] The flexibility enhancement module improves the manufacturing system's ability to cope with product and process changes. Through modular design, the production line can quickly adapt to new products or process changes. With the help of the intelligent scheduling system, the equipment can efficiently switch between different tasks. It relies on universal interface standards for cross-device and cross-platform integration to enhance the flexibility of the entire system.

[0028] The cost efficiency improvement module and the on-demand production module formulate reasonable production plans based on market forecasts, reduce inventory and excess production, and control costs. Market forecasts rely on the market information and intuitive feelings accumulated by the company's internal sales staff in daily contact with customers, comprehensively analyze the demand for different products in various regions and customer groups, and report forecast data to superiors. Through the design of questionnaires, field interviews, telephone surveys, and online surveys, they directly collect data information on product preferences, purchase intentions, and consumption trends from consumers, dealers, and end users, and then conduct statistical analysis to obtain market demand forecast results. Using the support vector machine (SVM) algorithm, SVM combines different categories of data (in market forecasts, different markets can be The SVM classifies products into categories based on demand level or product popularity, such as high demand, medium demand, and low demand categories, while ensuring that the classification interval is maximized to improve the generalization ability of the model. For nonlinearly separable data, SVM maps the original data to a high-dimensional space through a kernel function (such as a linear kernel, a polynomial kernel, a Gaussian kernel, etc.), making it linearly separable in the high-dimensional space, and then finds the optimal hyperplane in the high-dimensional space for classification prediction. When predicting the market, the various attributes of the product (price, function, appearance score, etc.) and external market environment factors (number of competitors, industry development trends, etc.) are used as input features. The trained SVM model is used to determine the category to which the product's future market demand belongs, and then the market situation is estimated;

[0029] The information interaction strengthening module breaks down information silos and strengthens data interaction and collaboration between manufacturing units. Information silos refer to the situation where data cannot flow and share smoothly between production units due to various reasons, forming relatively independent "islands". The data within each "island" is difficult to effectively interact with the outside world, resulting in the inability of the entire production system to coordinate operations and optimize decisions based on comprehensive and real-time data. Each production unit equipment is equipped with an Internet of Things communication module, and a suitable communication protocol is used to connect it to the Internet. Real-time data such as equipment operation, process parameters, and material flow are collected and aggregated to a unified Internet of Things platform to change the isolated status of equipment data. For example, machine tools are connected to the Internet to transmit processing parameters and other data for use by multiple departments. Data formats and standards covering coding, naming, units, etc. are formulated to ensure that data from different devices can be accurately identified and processed, so that data can flow and share smoothly on the platform, eliminating information blockages caused by format differences, such as unifying various data centers within the workshop. Data specifications for such equipment are used to aggregate and integrate data from various manufacturing units, different business systems, and IoT platforms through ETL technology, and then stored after cleaning and sorting to form a "data asset library", breaking the data limitations between departments. Through API and other means, unified data services and shared interfaces are provided to various application systems and production links, making it convenient to obtain shared data on demand and promote the collaboration of different business links. For example, the production planning and quality control departments call on the data of the data center to carry out work. From the planning level, ensure that each business system has integration and interoperability, clarify the data interaction interface and business process connection points, and realize seamless information transmission. For example, new orders can trigger multi-system linkage to adjust plans and arrange production tasks. Middleware is used to shield the underlying differences and serve as a "bridge" to help communication and collaboration between systems. For example, message middleware realizes asynchronous data transmission and notification, ensures timely and accurate information transmission, and breaks the information islands caused by technical differences between systems.

[0030] The deep technology integration module collects past production process data and uses machine learning algorithms to build a correlation model between process parameters and key indicators such as product quality and production efficiency;

[0031] The decision-making capability optimization module and dynamic programming algorithm facilitate efficient scheduling and switching of multi-variety production lines.

[0032] In one embodiment, the flexibility enhancement module includes a modular design module and an intelligent scheduling system module. The modular design module splits the equipment and production units into multiple standardized modules according to functions and processes. Each module has an independent and clear interface. When new products are produced or the process is changed, these modules are recombined and replaced according to the new requirements. The intelligent scheduling system module uses an artificial intelligence-driven scheduling algorithm to collect attribute information of production tasks and real-time monitor the operating status and idle status of the equipment on the production site. Based on the above data, the algorithm is used for intelligent analysis to dynamically allocate production tasks to equipment and production units; the production line can quickly adapt to new products and new processes, greatly shorten the adjustment time, reduce production stagnation caused by process changes, ensure that production activities can be quickly connected and carried out efficiently, significantly improve production efficiency, and make equipment switch between multiple tasks more efficient and convenient, so that equipment utilization can be improved, idle time is greatly reduced, and the entire production process is smoother, effectively responding to the task switching needs of small batch production of different products, and enhancing the overall flexibility and adaptability of the system.

[0033] In one embodiment, the cost efficiency improvement module includes an on-demand production module and a shared equipment platform module. The on-demand production module uses big data analysis and market research to predict market demand, and at the same time combines the product's own inventory situation, production cycle and cost structure to build a production planning model. The shared equipment platform module builds a shared platform for centralized management of production resources to integrate idle mobile robots and general equipment resources within the enterprise or between multiple enterprises. The production plan is more in line with market demand, reducing idle resources and waste caused by overproduction or underproduction. Resource sharing allows idle resources to play a role again, and energy optimization ensures the efficient operation of the production process. The overall production efficiency is steadily improved, achieving a benign balance between cost and efficiency.

[0034] In one embodiment, the information interaction enhancement module includes an industrial Internet of Things integration module, an edge computing and cloud computing combination module, and a real-time monitoring and feedback module. The industrial Internet of Things integration module deploys an Internet of Things communication module on each production unit, and uniformly adopts a suitable Internet of Things communication protocol to build a network for interconnection between devices and systems. The edge computing and cloud computing combination module deploys edge computing nodes at the edge close to the production equipment, which can receive and process local data collected from sensors and equipment in real time. The real-time monitoring and feedback module arranges various sensors on the production line to form a sensor network; real-time monitoring can capture any abnormal situation in the production process in time, and relevant personnel can quickly take measures to make adjustments to avoid the expansion of problems, thereby ensuring stable and continuous production and reducing the risk of production interruption due to emergencies such as failures.

[0035] In one embodiment, the technology deep integration module includes an artificial intelligence application module and a collaborative robot module. The artificial intelligence application module collects production process data and uses machine learning algorithms to build a correlation model between process parameters and product quality and production efficiency indicators. The collaborative robot module introduces a collaborative robot equipped with force sensors, visual sensors and intelligent control systems; improves production efficiency by optimizing process parameters to make the production process more efficient; improves product quality to ensure that products meet quality standards; reduces equipment failure rates and reduces production delays caused by equipment failures, allowing the entire production system to run more stably and efficiently, and giving full play to the advantages of artificial intelligence in intelligent manufacturing.

[0036] In one embodiment, the decision-making capability optimization module includes a real-time dynamic programming algorithm module and a multi-objective optimization module. The real-time dynamic programming algorithm module constructs the scheduling and switching problem of multi-variety production lines into a dynamic optimization problem. The multi-objective optimization module establishes a mathematical optimization model that comprehensively considers cost, efficiency, and quality, and uses a multi-objective optimization algorithm to weigh and optimize the decision variables in the production process. The real-time dynamic programming algorithm effectively improves the scheduling and switching efficiency of the production line. The production line can quickly respond to various dynamic production needs, reduce production delays caused by untimely and unreasonable scheduling, ensure the smoothness and efficiency of production, and enhance the system's adaptability to complex production situations.

[0037] In one embodiment, the dynamic programming formula for task allocation of device switching cost is as follows:

[0038]

[0039] Let n be the number of production tasks, m be the number of equipment, c ij represents the cost of assigning task i to device j, T ik represents the processing time of task i on equipment j, T ik represents the minimum total cost of the first i tasks assigned to device set j, where c jk represents the switching cost coefficient between device j and device k;

[0040] Construction of dynamic programming model:

[0041] Introduce the state variable F(i,S), which represents the first i tasks assigned to the device set The minimum total cost on the task, i represents the number of tasks that have been considered for allocation, which increases gradually from 1 to n, and S represents the subset of devices available for task allocation, which changes continuously with the task allocation process;

[0042] Consider that when assigning the i-th task, we need to select a device j from the currently available device set S to assign the task. The minimum total cost F(i,S) of the first i tasks assigned to the device set S depends on the minimum total cost F(i-1,S-{j}) of the first i-1 tasks assigned to the device set S-{j}, plus the cost C generated by assigning task i to device j. ij , the possible device switching cost due to the selection of device j;

[0043] Boundary conditions

[0044] When i = 1 and S = {j}, the boundary conditions are:

[0045] F(1,{j})=C 1j

[0046] Algorithm solution process

[0047] Based on the above dynamic programming model, a bottom-up approach is used to solve the problem. The specific steps are as follows:

[0048] initialization

[0049] According to the boundary conditions, first calculate the value of F(1,{j})(j=1,2,…,m), which is the cost of assigning the first task to each device;

[0050] Iterative Calculation

[0051] For i = 2 to N, and all possible device subsets Calculate the value of F(i,S) according to the state transfer equation. During the calculation process, it is necessary to traverse each device j in the device subset S and calculate the current F(i,S) using the calculated F(i-1,S-{j}) and other related values ​​according to the calculation method of the state transfer equation.

[0052] Determine the optimal allocation

[0053] After completing all the iterative calculations, the value of F(n,J) is the minimum total cost of assigning all n tasks to m devices. By backtracking the calculation process, we can determine which device each task is specifically assigned to, thereby obtaining the optimal task allocation plan. The specific backtracking method can start from F(n,J), and according to the device j that minimizes the cost when calculating each state F(i,S), we can gradually deduce forward to determine the assigned device for each task.

[0054] In another embodiment, when the enterprise receives a customer order, it first sorts out and analyzes key information such as the type, quantity, delivery time, and customization requirements of the product in the order, and combines market forecast information to comprehensively judge the scale and complexity of the production task. The on-demand production module preliminarily plans the production quantity, time schedule, and other rough production plan frameworks based on the order details and market forecast data. According to the results of the order analysis, the production plan is further refined to determine the sequence of each production link, the time nodes of each stage, and which production lines and equipment are specifically assigned to perform the production tasks. The real-time dynamic planning algorithm module uses reinforcement learning or genetic algorithms based on the current production tasks, equipment status and other information to generate a specific, real-time optimized production scheduling plan, determine the allocation and execution order of different production tasks on each device, and ensure efficient use of equipment resources. The shared equipment platform module coordinates and allocates the required shared equipment based on the scheduling plan and incorporates it into the corresponding production process. The general interface standard module ensures that the newly connected shared equipment or other newly introduced equipment can be smoothly integrated with the existing production line to achieve seamless docking, so as to facilitate subsequent production according to the production plan. The modular design module quickly adjusts the production line according to the plan, and adapts the production line layout to the specific product by adding, reducing or replacing the production unit module. Production process requirements, further optimize the allocation of production resources, each production equipment starts to operate according to the established production plan and process parameters, and processes and manufactures products. During the production process, various production information such as the equipment's operating status, process parameter data, and material flow are collected in real time. At the same time, operators and collaborative robots work together according to their respective division of labor. For example, collaborative robots are responsible for high-precision and repetitive operations, and operators perform more flexible and creative processes or monitor and intervene in the production process as necessary. The artificial intelligence application module continuously monitors production data and uses machine learning algorithms to optimize process parameters in real time to ensure product quality and production efficiency. At the same time, it predicts faults based on equipment operation data, warns of possible equipment failures in advance, and ensures stable production operation. The collaborative robot module works closely with the operator, and senses the surrounding environment and personnel movements through its equipped sensors to achieve safe and efficient human-machine collaboration and improve production efficiency. The industrial Internet of Things integration module collects and transmits data from each production unit in real time to provide comprehensive and accurate production status information for the entire system. On the one hand, these data are used by the local edge computing and cloud computing combination modules. The edge computing nodes quickly process real-time data close to the device end to achieve local real-time responses such as immediate fault alarms and local process parameter fine-tuning.On the other hand, the data is uploaded to the cloud computing platform for more complex global analysis, long-term trend forecasting, etc., to provide a decision-making basis for subsequent production optimization. Through the real-time monitoring and feedback mechanism composed of sensor networks and data acquisition systems, we continue to pay attention to various indicators in the production process, such as product quality, equipment operation status, production progress, etc. Once deviations from the expected plan or abnormal situations are found, such as an increase in product quality failure rate, sudden equipment failure, delayed production progress, etc., the adjustment mechanism is activated in time. The real-time monitoring and feedback module will promptly feedback the monitored abnormal information to relevant personnel and systems. The real-time dynamic planning algorithm module will re-schedule tasks and allocate resources according to the new production status, and adjust the production plan, such as transferring some tasks to other idle equipment, adjusting the order of subsequent processes, etc., to ensure that production returns to normal as soon as possible and on schedule. Plan advancement. When all products are completed and pass quality inspection, they are delivered to customers. Then the entire production process is reviewed and summarized, and various data in the production process are collected, including cost data, quality data, equipment operation data, etc. The advantages and disadvantages of the production process are analyzed to provide reference for subsequent production improvements. The multi-objective optimization module comprehensively considers data from multiple dimensions such as cost, efficiency, and quality to evaluate whether this production has achieved the expected comprehensive benefit goals, analyze the balance between various goals, and find out the links that can be further optimized. The artificial intelligence application module further trains the model based on the data of this production to improve the model's optimization ability for similar production tasks and the accuracy of fault prediction, accumulate experience for the next round of production activities, and continuously improve the performance of the entire flexible intelligent manufacturing system and its ability to cope with multi-variety small-batch production. ;

[0055] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of various changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. Flexible intelligent manufacturing system for multi-variety and small batch production, characterized by: It includes flexibility enhancement module, cost efficiency improvement module, information interaction enhancement module, technology deep integration module and decision-making ability optimization module; The flexibility enhancement module improves the ability of the manufacturing system to cope with product and process changes. Through modular design, the production line can quickly adapt to new products or process changes. With the help of the intelligent scheduling system, the equipment can efficiently switch between different tasks. It relies on universal interface standards for cross-device and cross-platform integration to enhance the flexibility of the entire system. The cost efficiency improvement module and the on-demand production module formulate reasonable production plans based on market forecasts, reduce inventory and excess production, and control costs; The information interaction enhancement module breaks down information silos and enhances data interaction and collaboration among manufacturing units; The technical deep integration module collects past production process data and uses machine learning algorithms to build a correlation model between process parameters and key indicators such as product quality and production efficiency; The decision-making capability optimization module and dynamic programming algorithm facilitate efficient scheduling and switching of multi-variety production lines.

2. The flexible intelligent manufacturing system for multi-variety and small batch production according to claim 1 is characterized in that: The flexibility enhancement module includes a modular design module and an intelligent scheduling system module. The modular design module divides the equipment and production units into multiple standardized modules according to functions and processes. Each module has an independent and clear interface. When new products are produced or the process is changed, these modules are recombined and replaced according to new requirements. The intelligent scheduling system module uses an artificial intelligence-driven scheduling algorithm to collect attribute information of production tasks and monitor the operating status and idle conditions of the equipment in real time. Based on the above data, intelligent analysis is performed through the algorithm to dynamically allocate production tasks to equipment and production units.

3. The flexible intelligent manufacturing system for multi-variety and small batch production according to claim 1 is characterized by: The cost efficiency improvement module includes an on-demand production module and a shared equipment platform module. The on-demand production module uses big data analysis and market research to predict market demand, and at the same time combines the product's own inventory situation, production cycle and cost structure to build a production planning model. The shared equipment platform module builds a shared platform for centralized management of production resources, and integrates idle mobile robots and general equipment resources within an enterprise or between multiple enterprises.

4. The flexible intelligent manufacturing system for multi-variety and small batch production according to claim 1 is characterized in that: The information interaction enhancement module includes an industrial Internet of Things integration module, an edge computing and cloud computing combination module, and a real-time monitoring and feedback module. The industrial Internet of Things integration module deploys an Internet of Things communication module on each production unit, and uniformly adopts a suitable Internet of Things communication protocol to build a network for interconnection between devices and systems. The edge computing and cloud computing combination module deploys edge computing nodes at the edge end close to the production equipment, which can receive and process local data collected from sensors and equipment in real time. The real-time monitoring and feedback module arranges various sensors on the production line to form a sensor network.

5. The flexible intelligent manufacturing system for multi-variety and small batch production according to claim 1 is characterized by: The technology deep integration module includes an artificial intelligence application module and a collaborative robot module. The artificial intelligence application module collects production process data and uses machine learning algorithms to build a correlation model between process parameters and product quality and production efficiency indicators. The collaborative robot module introduces a collaborative robot equipped with force sensors, vision sensors and intelligent control systems.

6. The flexible intelligent manufacturing system for multi-variety and small batch production according to claim 1 is characterized by: The decision-making capability optimization module includes a real-time dynamic programming algorithm module and a multi-objective optimization module. The real-time dynamic programming algorithm module constructs the scheduling and switching problems of multi-variety production lines into a dynamic optimization problem. The multi-objective optimization module establishes a mathematical optimization model that comprehensively considers cost, efficiency, and quality, and uses a multi-objective optimization algorithm to weigh and optimize the decision variables in the production process.

7. The flexible intelligent manufacturing system for multi-variety and small batch production according to claim 1 is characterized by: The dynamic programming formula for task allocation of device switching cost is as follows: Let n be the number of production tasks, m be the number of equipment, C ij represents the cost of assigning task i to device j, T ik represents the processing time of task i on equipment j, T ik represents the minimum total cost of the first i tasks assigned to device set j, where c jk represents the switching cost coefficient between device j and device k.

Images