Intelligent injection molding workshop scheduling method and system

Abstract

The present invention proposes an intelligent injection molding workshop scheduling method and system, which aims to improve the efficiency and adaptability of injection molding production by combining a particle swarm optimization algorithm and a fuzzy logic controller. First, a set of production scheduling schemes including production task scheduling sequence and resource allocation are initialized using a particle swarm optimization algorithm, and each scheme is regarded as a particle. Then, a fitness score based on production efficiency and task completion timeliness is calculated for each particle to evaluate its performance. By examining the individual optimal solution and the group optimal solution, the particle swarm optimization algorithm updates the position and speed of the particles and iteratively optimizes the production scheduling scheme. At the same time, real-time production data of the injection molding workshop, including environmental parameters, machine status, and production task information, is collected to deal with uncertainty and ambiguity problems in the production process. Finally, the fuzzy logic controller is used to refine and generate the final production scheduling scheme by combining the optimized production scheduling scheme set and real-time production data.

Description

Technical Field

[0001] The present invention relates to the technical field of workshop production scheduling, and in particular to an intelligent injection molding workshop scheduling method and system. Background Art

[0002] Among the existing injection molding workshop production scheduling methods, conventional scheduling strategies mainly rely on static planning and empirical judgment, and lack the ability to respond quickly to dynamic changes in the production process. These methods often show shortcomings when dealing with complex production tasks, coping with machine failures, material shortages, and changes in production priorities. In addition, traditional scheduling methods find it difficult to reflect changes in the workshop environment in real time, such as temperature fluctuations or operator changes, which directly affect product quality and production efficiency. Due to the lack of flexibility and adaptability, these methods cannot maximize resource utilization, resulting in low production efficiency and increased production costs.

[0003] On the other hand, although some advanced scheduling technologies try to use computer-aided design and artificial intelligence algorithms to improve production scheduling, they often require complex algorithm adjustments and a large amount of computing resources, which may encounter problems of complex operations and high costs in practical applications.

[0004] Therefore, it is very necessary to develop a new type of intelligent injection molding workshop scheduling method and system. Summary of the invention

[0005] The present application provides an intelligent injection molding workshop scheduling method and system to improve the efficiency of injection molding production.

[0006] The present application provides an intelligent injection molding workshop scheduling method, comprising:

[0007] A set of production scheduling schemes are initialized using a particle swarm optimization algorithm, wherein each production scheduling scheme is a particle, and the particle includes the scheduling sequence and resource allocation of each production task in the injection molding workshop;

[0008] Calculating a fitness score for each particle based on production efficiency and timeliness of task completion to obtain a fitness score result for each particle, wherein the fitness score result is used to evaluate the performance of each particle;

[0009] According to the fitness score of each particle, the position and speed of each particle are updated through the particle swarm optimization algorithm to explore possible individual optimal solutions and group optimal solutions; by comparing and updating possible individual optimal solutions and group optimal solutions, a set of production scheduling solutions is obtained;

[0010] Collect real-time production data of the injection molding workshop, wherein the real-time production data includes environmental parameters, machine status, and production task information, which is used to deal with uncertainty and ambiguity problems in the production process;

[0011] According to the production scheduling plan set and real-time production data, a final production scheduling plan is generated through a fuzzy logic controller.

[0012] Furthermore, the particle swarm optimization algorithm is used to initialize a set of production scheduling schemes, including randomizing the initial position and speed of each particle to enhance the coverage of the search space and ensure that the optimal solution is searched from a variety of potential solutions.

[0013] Furthermore, the fitness score of each particle is calculated based on production efficiency and timeliness of task completion to obtain the fitness score result of each particle, including:

[0014] The fitness score F of each particle is calculated using the following formula 1:

[0015]

[0016] Among them, w1, w2 and w3 are weight parameters, which are used to balance the influence of various indicators in the fitness score; T avg Represents the average time required to complete all production tasks; T idle Indicates the total idle time of the machine; T total is the total production time; D late is the number of delayed production tasks; D total is the number of all production tasks.

[0017] Furthermore, according to the fitness score result of each particle, the position and speed of each particle are updated by the particle swarm optimization algorithm to explore possible individual optimal solutions and group optimal solutions, including:

[0018] The position and velocity of each particle are updated through the following formula 2 and formula 3 to explore possible individual optimal solutions and group optimal solutions:

[0019]

[0020]

[0021] in, is the velocity of particle i at the next time t+1; is the velocity of particle i at the current time t; is the position of particle i at the next time t+1; is the position of particle i at the current time t; ω is the inertia weight, which is used to control the contribution of the current velocity of the particle to its new velocity; φ1, φ2 and φ3 are positive coefficients, representing the influence of individual cognition, social cognition and fitness score results on velocity update; f(i) is the fitness score result of particle i; r1 and r2 are random numbers between the interval [0,1]; p best,i is the best position found by particle i so far, i.e., the individual optimal solution; g best It is the best position found by all particles in the current group, that is, the group optimal solution.

[0022] Furthermore, the collection of real-time production data of the injection molding workshop includes automatically collecting data using Internet of Things technology to achieve high-efficiency and high-precision data collection.

[0023] Furthermore, the generating of the final production scheduling plan through the fuzzy logic controller according to the production scheduling plan set and the real-time production data includes:

[0024] Constructing a fuzzy logic control system, wherein the fuzzy logic control system is based on a predefined fuzzy rule set, wherein the fuzzy rule set is used to describe the relationship between the production scheduling solution set and the real-time production data;

[0025] The fuzzy logic control system uses the following formula 4 to score the candidate scheduling solutions in the production scheduling solution set:

[0026]

[0027] Among them, Score represents the comprehensive score of the candidate scheduling scheme; w n is the weight parameter, which represents the importance of each fuzzy rule in the decision-making process; is a fuzzy set A n The corresponding membership function value indicates the degree to which the real-time production data meets the conditions of a specific fuzzy rule; x is the real-time production data, including environmental parameters, machine status, and production task information; n is the number of fuzzy rules;

[0028] According to the scoring results, the final production scheduling plan is obtained.

[0029] Furthermore, the fuzzy logic control system includes a mechanism for dynamically updating the fuzzy rule set, and the mechanism adjusts the fuzzy rule set to reflect the latest production scheduling requirements and constraints according to new situations and data emerging in the production process.

[0030] Furthermore, the method of generating a final production scheduling plan based on the production scheduling plan set and real-time production data through a fuzzy logic controller also includes comparing and analyzing the final scheduling plan with historical data, evaluating the degree of improvement of the final scheduling plan relative to the historical plan, and providing suggestions for continuous improvement based on this.

[0031] The present application provides an intelligent injection molding workshop scheduling system, comprising:

[0032] An initialization unit, used to initialize a set of production scheduling schemes using a particle swarm optimization algorithm, wherein each production scheduling scheme is a particle, and the particle includes a scheduling sequence and resource allocation of each production task in the injection molding workshop;

[0033] A calculation unit, used to calculate a fitness score for each particle based on production efficiency and timeliness of task completion, to obtain a fitness score result for each particle, wherein the fitness score result is used to evaluate the performance of each particle;

[0034] The acquisition unit is used to update the position and speed of each particle through the particle swarm optimization algorithm according to the fitness score result of each particle, so as to explore the possible individual optimal solution and group optimal solution; by comparing and updating the possible individual optimal solution and group optimal solution, a production scheduling solution set is obtained;

[0035] A collection unit, used to collect real-time production data of the injection molding workshop, wherein the real-time production data includes environmental parameters, machine status, and production task information, and is used to handle uncertainty and ambiguity problems in the production process;

[0036] The generating unit is used to generate a final production scheduling plan through a fuzzy logic controller according to the production scheduling plan set and real-time production data.

[0037] Furthermore, the computing unit is specifically used for:

[0038] The fitness score F of each particle is calculated using the following formula 1:

[0039]

[0040] Among them, w1, w2 and w3 are weight parameters, which are used to balance the influence of various indicators in the fitness score; T avg Represents the average time required to complete all production tasks; T idle Indicates the total idle time of the machine; T total is the total production time; D late is the number of delayed production tasks; D total is the number of all production tasks.

[0041] The beneficial effects of the present application include: (1) By initializing and optimizing the production scheduling plan through the particle swarm optimization algorithm, the present application can effectively match production tasks with available resources, thereby reducing idle time and resource waste in production and improving overall production efficiency. (2) By using real-time production data and combining the fuzzy logic controller to dynamically adjust the production scheduling plan, the present application can quickly respond to various emergencies that occur during the production process (such as machine failures, emergency orders, etc.) to ensure the continuity and flexibility of the production process. (3) By accurately calculating the fitness score of each production scheduling plan and optimizing the scheduling plan based on the fitness results, the present application can more reasonably allocate production resources, avoid excessive concentration or idleness of resources, and thus reduce production costs. (4) By real-time monitoring and adjusting the production plan, it is ensured that the production tasks can be completed on time according to the optimized scheduling plan, which helps to improve product quality and meet delivery requirements, and enhance the market competitiveness of the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a flow chart of an intelligent injection molding workshop scheduling method provided in the first embodiment of the present application.

[0043] Figure 2 It is a schematic diagram of an intelligent injection molding workshop scheduling system provided in the second embodiment of the present application. DETAILED DESCRIPTION

[0044] Many specific details are described in the following description to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present application, so the present application is not limited by the specific implementation disclosed below.

[0045] The first embodiment of the present application provides a method for scheduling an intelligent injection molding workshop. Figure 1 , which is a schematic diagram of the first embodiment of the present application. Figure 1 A first embodiment of the present application provides a method for scheduling an intelligent injection molding workshop and is described in detail.

[0046] Step S101: Initialize a set of production scheduling schemes using a particle swarm optimization algorithm, wherein each production scheduling scheme is a particle, and the particle includes the scheduling sequence and resource allocation of each production task in the injection molding workshop.

[0047] In step S101, a particle swarm optimization algorithm is used to initialize a set of production scheduling solutions for the injection molding workshop. This process first involves defining a solution space, where each solution represents a possible production scheduling solution. Each solution is regarded as a "particle", and the entire solution space constitutes a "swarm".

[0048] Each particle specifically contains the scheduling order of all production tasks to be completed in the injection molding workshop and the corresponding resource allocation plan. The resource allocation here mainly involves key factors such as machine allocation, personnel arrangement, and raw material usage. In order to initialize these particles, the algorithm first randomly generates a set of production scheduling plans. This means that each particle has a random position at the beginning, representing a specific production scheduling order and resource allocation strategy.

[0049] Next, the particle swarm optimization algorithm uses predetermined criteria to simulate the social behavior of particles, where each particle adjusts its "position" and "speed" based on its own experience and the experience of other particles in the group. In this process, "position" refers to the specific production scheduling plan represented by the particle, and "speed" reflects the trend and speed of change of the plan. The movement of particles means exploring the solution space of the production scheduling plan to find a better scheduling plan that can improve production efficiency and timeliness of task completion.

[0050] Furthermore, the particle swarm optimization algorithm is used to initialize a set of production scheduling schemes, including randomizing the initial position and speed of each particle to enhance the coverage of the search space and ensure that the optimal solution is searched from a variety of potential solutions.

[0051] Randomizing the initial position and velocity of particles means that the starting state of each particle will be set based on a certain randomness before the algorithm starts executing. This not only helps to prevent the algorithm from focusing on a local area in the solution space too early, but also increases the possibility of discovering the global optimal solution.

[0052] The implementation steps are as follows:

[0053] (1) Randomization of initial position: For each particle, its initial position needs to be randomly selected from the entire solution space. This can be achieved by generating random values ​​for the parameters in each scheduling scheme, for example, randomly determining the execution order of production tasks or the initial allocation of resources. This step ensures that the algorithm's starting search points are distributed in different areas of the solution space, thereby enhancing the breadth of exploration.

[0054] (2) Randomization of initial velocity: Similarly, the initial velocity of each particle should also be randomly set. The randomization of velocity involves not only the magnitude of the velocity, but also the direction, that is, the randomness that determines the initial movement trend of the particle. This step allows the particles to explore different directions in the solution space at the initial stage, which helps to avoid the entire particle group from gathering near the local optimal solution in the early stage.

[0055] Through such randomization, the particle swarm optimization algorithm has stronger diversity and exploration capabilities in the process of searching for the optimal production scheduling solution. This method is particularly suitable for solving scheduling problems with complex search spaces and multiple local optimal solutions, such as intelligent injection molding workshop scheduling. It ensures that a wide search range can be covered from the beginning, greatly increasing the possibility of finding a better and more efficient scheduling solution.

[0056] In summary, step S101 provides an efficient and flexible initialization method for the production scheduling problem of the injection molding workshop through the intelligent search capability of the particle swarm optimization algorithm. This method can provide workshop managers with a variety of scheduling solutions and provide strong support for maximizing production efficiency and minimizing costs.

[0057] Step S102: Calculate the fitness score of each particle based on production efficiency and timeliness of task completion to obtain a fitness score result of each particle, wherein the fitness score result is used to evaluate the performance of each particle.

[0058] In step S102, the performance of each production scheduling scheme in the injection molding workshop is evaluated. The core of this process is to calculate the fitness score. The fitness score is a key concept in the particle swarm optimization algorithm. It provides a quantitative performance indicator for each particle, that is, each production scheduling scheme. This score is based on two main factors: production efficiency and timeliness of task completion. Production efficiency reflects the execution speed and resource utilization of production tasks under given resource conditions, while timeliness of task completion measures whether the production task can be completed on schedule.

[0059] In order to calculate the fitness score, we first need to define a fitness function. This function needs to be able to accept detailed information about the production scheduling plan represented by each particle, such as the scheduling order of tasks, the resource allocation required for each task, etc., and evaluate the production efficiency of the plan and the timeliness of task completion. For example, the fitness function may consider factors such as the start time and end time of each task, the idle time of the required machines, and the effectiveness of personnel allocation.

[0060] During the calculation process, the fitness function will generate a score for each production scheduling plan. A high score means that the plan can use resources more efficiently and complete production tasks faster, while a low score may indicate improper resource allocation or delays in the production process. These scores not only provide a direct comparison basis for the performance of each scheduling plan, but also provide important feedback information for the subsequent optimization process.

[0061] When implementing this step, you need to design and adjust the fitness function according to the specific production environment and task requirements. This may include determining which specific indicators of production efficiency and timeliness best reflect the performance of the scheduling plan, and how to combine these indicators to obtain a fitness score. In addition, you need to be able to process and analyze the relevant data of the production scheduling plan to ensure that the calculation of the fitness score is both accurate and efficient.

[0062] Furthermore, the fitness score of each particle is calculated based on production efficiency and timeliness of task completion to obtain the fitness score result of each particle, including:

[0063] The fitness score F of each particle is calculated using the following formula 1:

[0064]

[0065] Among them, w1, w2 and w3 are weight parameters, which are used to balance the influence of various indicators in the fitness score; T avg Represents the average time required to complete all production tasks; T idle Indicates the total idle time of the machine; T total Total production time; D late is the number of delayed production tasks; D total is the number of all production tasks.

[0066] The fitness score F is a comprehensive evaluation index of each particle (production scheduling plan), which is used to measure the pros and cons of the plan in meeting production efficiency and completing tasks in a timely manner.

[0067] w1, w2, and w3 are weight parameters used to balance the influence of different items in the formula. They can be adjusted based on actual production conditions and management preferences. For example, if timely completion of tasks is more important than improving production efficiency, this preference can be reflected by increasing the value of w3. These parameters can be determined through discussions with the production management team and historical data analysis to reflect the relative importance of different production goals.

[0068] T avg Represents the average time required to complete all production tasks, which indicates the average time required to complete all production tasks. A shorter average completion time means higher production efficiency, so this indicator is inversely proportional to the fitness score. This value can be obtained by analyzing historical production data.

[0069] T idle Represents the total idle time of the machine, which means the total time that the machine is not used during the entire production cycle. High idle time of the machine means low resource utilization and has a negative impact on the fitness score. This value can also be calculated from production data.

[0070] T total is the total production time, which is the total time required to complete all production tasks, including machine operation and idle time. This value reflects the length of the production cycle and serves as the basis for calculating the proportion of machine idle time.

[0071] D late is the number of delayed production tasks, which indicates the number of production tasks that were not completed within the scheduled completion time. The delayed completion of tasks directly affects the delivery time and therefore has a negative impact on the fitness score.

[0072] D total is the number of all production tasks, which is the total number of tasks that need to be completed. This value provides a measure of the size of the production tasks and is used to calculate the proportion of delayed tasks.

[0073] The production tasks of the injection molding workshop involve multiple steps, from raw material preparation to the molding and post-processing of the final product, each step is a key component of the entire production process. The following are some typical production task examples:

[0074] 1. Raw material preparation: This includes selecting the right plastic pellets or powder and ensuring that the quality of the material meets production requirements. This may involve cleaning, drying and preheating the material.

[0075] 2. Mold preparation and installation: Select the appropriate mold, clean and lubricate it, and then install it on the injection molding machine. This step is crucial to ensure product quality and reduce production failures.

[0076] 3. Injection molding process: preheated plastic material is injected into the mold and molded at appropriate temperature and pressure. This process requires precise control to ensure that the size and shape of the product meet the design requirements.

[0077] 4. Cooling and solidification: Before the plastic cools and solidifies into its final shape, it needs to remain in the mold for a period of time. The cooling speed and method will directly affect the quality and production efficiency of the product.

[0078] 5. Product demoulding: Remove the molded plastic part from the mold. This may require the use of specialized demoulding equipment to prevent damage to the product.

[0079] 6. Post-processing: including removing excess materials from the product, grinding and cleaning the surface, and performing necessary assembly or painting. The post-processing steps depend on the specific requirements of the product.

[0080] 7. Quality inspection: Check whether each product meets the quality standards, including size, weight, surface quality, etc. Unqualified products will be rejected or sent back for re-processing.

[0081] 8. Packaging and warehousing: Qualified products will be properly packaged and ready for shipment or transferred to warehouse for storage. The packaging process needs to take into account product protection and transportation efficiency.

[0082] 9. Equipment maintenance and cleaning: Regularly maintain and clean the injection molding machine and related equipment to ensure the smooth progress of the production process and extend the life of the equipment.

[0083] Together, these tasks constitute the production process of the injection molding workshop, and each step needs to be carefully planned and managed to ensure production efficiency and product quality.

[0084] By accurately calculating the fitness score of each scheduling plan, step S102 not only provides the injection molding workshop with a scientific and quantitative method for evaluating production scheduling plans, but also lays the foundation for further optimizing these plans through subsequent steps, with the ultimate goal of achieving higher production efficiency and better timeliness of task completion.

[0085] Step S103: According to the fitness score result of each particle, the position and speed of each particle are updated through the particle swarm optimization algorithm to explore possible individual optimal solutions and group optimal solutions; by comparing and updating the possible individual optimal solutions and group optimal solutions, a production scheduling solution set is obtained.

[0086] In step S103, we focus on how to use the particle swarm optimization algorithm (PSO) to update the position and speed of each particle, i.e., the representative of each production scheduling scheme, according to the fitness score result. This process is the core of finding the optimal production scheduling scheme, which directly affects the production efficiency of the injection molding workshop and the timeliness of task completion.

[0087] First of all, it should be clear that the "position" of each particle here represents a specific production scheduling plan, while the "speed" indicates the direction and magnitude of the change in the plan. The fitness score results reflect the performance of the current plan, including key indicators such as production efficiency and task completion speed. Based on these scores, the algorithm will calculate new speeds and positions in order to explore better production scheduling plans.

[0088] The update process follows the rules of the particle swarm optimization algorithm, which includes two key social behavior models: individual learning and group learning. Individual learning considers the best position of each particle in its past experience (i.e., the best solution found by the particle), while group learning refers to the best solution in the entire particle group. These two aspects of information jointly affect the update of particle speed, which in turn determines the new position of the particle, i.e., the production scheduling solution.

[0089] Specifically, the velocity update takes into account the particle's current velocity, the relative distance from the particle to its individual optimal position, and the relative distance from the particle to the group's optimal position. The position update is based on the new velocity value, in this way, the particle will move in the solution space to find a better scheduling solution.

[0090] In order to implement this step, it is necessary to set the parameters of the PSO algorithm according to the actual production tasks and resource configuration, including the particle speed update rules, learning factors, and the calculation method of the fitness score. In addition, a mechanism needs to be designed to record the individual best position of each particle and the group best position of the entire particle group, which are indispensable information in the speed and position update process.

[0091] Suppose you are managing an injection molding shop, and your goal is to optimize production scheduling to improve efficiency and reduce costs. Using the particle swarm optimization (PSO) algorithm, a group of particles is simulated, where each particle represents a potential production scheduling solution. Each solution involves factors such as the order of production tasks, machine allocation, and staff scheduling.

[0092] First, a fitness score is calculated for each particle (i.e., each potential scheduling solution). This score is based on several key performance indicators, such as timeliness of task completion, resource utilization, and production efficiency. For example, if a particle represents a scheduling solution that ensures that all tasks are completed on time while minimizing machine idle time, then the fitness score of this solution will be high.

[0093] Based on the fitness score results, the PSO algorithm then updates the position and velocity of each particle. This update process is designed to simulate the collective behavior of bird flocks or fish schools in nature, so that each particle moves towards its individual optimal solution (that is, the best scheduling solution it has found) and the group optimal solution (that is, the best scheduling solution among all particles).

[0094] Assume there are three particles A, B and C, which represent three different production scheduling schemes. After one round of calculation:

[0095] Particle A finds a good scheduling solution, and its fitness score is higher than that of B and C;

[0096] Particles B and C have lower solution fitness scores because they do not fully utilize the available resources.

[0097] In the next iteration, particles B and C will update their speed and position according to the PSO algorithm, that is, adjust the scheduling solutions they represent to get closer to the efficient solution of particle A. This may involve changing the execution order of production tasks, reallocating machines, or adjusting work shifts.

[0098] After multiple rounds of iterations, the particle swarm will gradually gather in the optimal area in the solution space. At this time, the group optimal solution represents the best production scheduling solution. By comparing and updating the individual optimal solution and the group optimal solution, a series of optimized production scheduling solutions are obtained, from which the solution that best suits the current production demand and resource status can be selected and implemented.

[0099] This process not only enhances the overall performance of the scheduling solution, but also provides the injection molding workshop with a flexible and efficient production scheduling method to cope with various challenges and changes that may arise in the production process.

[0100] Furthermore, according to the fitness score result of each particle, the position and speed of each particle are updated by the particle swarm optimization algorithm to explore possible individual optimal solutions and group optimal solutions, including:

[0101] The position and velocity of each particle are updated through the following formula 2 and formula 3 to explore possible individual optimal solutions and group optimal solutions:

[0102]

[0103]

[0104] in, It is the speed of particle i at the next moment t+1; the concept of speed is abstract in the PSO algorithm, referring to the "speed" or rate of change of the particle moving in the solution space.

[0105] is the velocity of particle i at the current time t;

[0106] is the position of particle i at the next time t+1;

[0107] is the position of particle i at the current time t;

[0108] ω is the inertia weight that controls the contribution of the particle's current velocity to its new velocity; ω can be adjusted based on previous studies or through experiments to achieve optimal search efficiency and accuracy.

[0109] φ1, φ2, and φ3 are positive coefficients, representing the influence of individual cognition, social cognition, and fitness score results on speed updating, respectively; φ1, φ2, and φ3 can be adjusted based on previous studies or through experiments to achieve optimal search efficiency and accuracy.

[0110] f(i) is the fitness score of particle i; it can be calculated according to formula 1, taking into account factors such as production efficiency and timeliness of task completion. r1 and r2 are random numbers between the interval [0,1], which introduce randomness to increase the efficiency of exploring the solution space. r1 and r2 are generated by a random number generator to ensure randomness and diversity in the search process.

[0111] p best,i is the best position found by particle i so far, i.e., the individual optimal solution; g best is the best position found by all particles in the current group, that is, the group optimal solution. best,i and g best During the search, these values ​​are updated whenever a particle finds a better position.

[0112] By implementing step S103, the injection molding workshop manager can continuously optimize the production scheduling plan to cope with various changes and challenges in the production process. This step ensures that the scheduling plan always develops in the direction of improving production efficiency and accelerating task completion, providing an effective method for achieving efficient and flexible production scheduling.

[0113] Step S104: collecting real-time production data of the injection molding workshop, wherein the real-time production data includes environmental parameters, machine status, and production task information, which are used to handle uncertainty and ambiguity problems in the production process.

[0114] In step S104, the real-time monitoring link of the injection molding workshop is deeply involved, which is crucial for adjusting and optimizing the production scheduling plan. The core of this step is to collect key production data in the workshop in real time, including but not limited to environmental parameters, machine operation status, and the latest information of various production tasks. These data constitute a comprehensive reflection of the current production status of the workshop and are the basis for understanding and responding to uncertainty and ambiguity problems in the production process.

[0115] First, environmental parameters may include factors such as temperature and humidity, which may directly affect product quality and production efficiency. For example, under certain temperature and humidity conditions, some materials may require longer cooling or heating time, which is directly related to the adjustment of production rhythm.

[0116] Next, the real-time data of the machine status includes but is not limited to the machine's operating efficiency, fault records, maintenance logs, etc. This information is extremely important for preventive maintenance, rapid response to faults, and ensuring stable operation of the production line.

[0117] Finally, real-time updates of production task information provide detailed information on the progress, priority, required resources, etc. of each task. This is essential for adjusting production plans, optimizing resource allocation, and responding to urgent orders.

[0118] In order to implement step S104, an efficient data collection and processing system needs to be established. This usually involves installing various sensors to collect environmental and machine status data, and developing software tools to integrate and analyze this data. In addition, a real-time production management system is required to track the status and progress of production tasks. The design and implementation of these systems and tools need to ensure the accuracy, real-time and reliability of the data in order to provide solid information support for the optimization of production scheduling.

[0119] Furthermore, the collection of real-time production data of the injection molding workshop includes automatically collecting data using Internet of Things technology to achieve high-efficiency and high-precision data collection.

[0120] By utilizing the Internet of Things (IoT) technology to automatically collect real-time production data from the injection molding workshop, the introduction of this method aims to improve the efficiency and accuracy of data collection, thereby providing a more accurate and comprehensive information basis for the optimization of production scheduling.

[0121] IoT technology allows for the automatic collection and exchange of data through networked devices and sensors. In an injection molding shop, these devices may include, but are not limited to:

[0122] Temperature sensor: monitors the workshop ambient temperature and injection molding machine temperature to ensure that production conditions meet standards.

[0123] Humidity sensor: records the humidity level in the workshop, which is particularly important for the processing of certain materials.

[0124] Machine Condition Monitor: Tracks the operating status, efficiency and possible fault indications of the injection molding machine.

[0125] Production Progress Tracker: Track the progress of each production task in real time, including start time, end time and current status.

[0126] Raw Materials Monitoring System: Monitors the usage of raw materials, ensures adequate supply and tracks material consumption rates.

[0127] Data collection and processing include:

[0128] Automated data collection: Data is automatically collected through various sensors and devices installed in the workshop. These data are then transmitted to the central data processing system through wireless or wired networks.

[0129] Data processing and analysis: The collected data is sent to a central server or cloud platform for processing and analysis. Using data analysis tools, useful information can be extracted from this data, such as identifying production bottlenecks, predicting equipment failures, etc.

[0130] Real-time feedback loop: Processed data can be used to instantly adjust production plans and optimize resource allocation. For example, if a machine’s efficiency is detected to be declining, the system can automatically reallocate tasks to other machines to avoid production delays.

[0131] It is necessary to ensure that all sensors and monitoring devices are properly connected to the network and can transmit data stably. When designing the system, special attention should be paid to data security and privacy protection to ensure that sensitive information is not accessed by unauthorized persons. IoT devices and systems need to be compatible with existing production management software and hardware to achieve seamless integration and operation.

[0132] Through this IoT technology-driven approach, the injection molding workshop is able to achieve real-time monitoring and management of the production process, optimize production scheduling, improve production efficiency, and reduce resource waste.

[0133] By carefully and comprehensively collecting and analyzing real-time production data, step S104 provides the necessary input for the subsequent generation of the final production scheduling plan through the fuzzy logic controller. This data-based approach not only improves the adaptability and flexibility of the production scheduling plan, but also provides a solid information foundation for injection molding workshop management and decision-making.

[0134] Step S105: Generate a final production scheduling plan based on the production scheduling plan set and real-time production data through a fuzzy logic controller.

[0135] In step S105, the core task is to use the real-time production data collected previously and the production scheduling solution set refined by the particle swarm optimization algorithm, and comprehensively process this information through the fuzzy logic controller (FLC) to generate the final production scheduling solution. This step is the key link in realizing the intelligent injection molding workshop scheduling method. It combines the advantages of algorithm optimization and real-time data processing to ensure that the production scheduling solution is both efficient and adaptable.

[0136] Fuzzy logic controllers play a vital role in this process. FLCs are able to process ambiguous and uncertain information and convert it into specific decisions and actions. This means that even fuzzy environmental parameters (such as "temperature is too high"), machine status (such as "slight wear"), or production task information (such as "higher priority") can be effectively used to guide the decision-making process of production scheduling.

[0137] To implement step S105, it is necessary to first define a set of fuzzy rules based on the actual experience and expertise of production management in the injection molding workshop. For example, a rule may be: "If the status of machine A is slightly worn and the production task X has a higher priority, then assign task X to machine B." The design of fuzzy rules needs to take into account various production conditions and possible scenarios to ensure that most of the situations that may be encountered in the production process can be covered.

[0138] Next, the fuzzy logic controller will use this set of fuzzy rules, combined with the real-time production data collected in step S104 and the optimized production scheduling plan set obtained in step S103, to make decisions. The FLC will conduct a comprehensive evaluation of each production task and resource allocation, and dynamically adjust the production scheduling plan based on the specific situation of the fuzzy rules and real-time data.

[0139] Finally, through the processing of the fuzzy logic controller, a final production scheduling plan can be generated that takes into account production efficiency and task completion timeliness, and can adapt to actual production changes. This plan will specify in detail the scheduling sequence and resource allocation plan for each production task in the injection molding workshop to ensure that the workshop operates in the optimal state.

[0140] Furthermore, the generating of the final production scheduling plan through the fuzzy logic controller according to the production scheduling plan set and the real-time production data includes:

[0141] Constructing a fuzzy logic control system, wherein the fuzzy logic control system is based on a predefined fuzzy rule set, wherein the fuzzy rule set is used to describe the relationship between the production scheduling solution set and the real-time production data;

[0142] The fuzzy logic control system uses the following formula 4 to score the candidate scheduling solutions in the production scheduling solution set:

[0143]

[0144] Among them, Score represents the comprehensive score of the candidate scheduling scheme; w n is the weight parameter, which represents the importance of each fuzzy rule in the decision-making process; is a fuzzy set A n The corresponding membership function value indicates the degree to which the real-time production data meets the conditions of a specific fuzzy rule; x is the real-time production data, including environmental parameters, machine status, and production task information; n is the number of fuzzy rules;

[0145] According to the scoring results, the final production scheduling plan is obtained.

[0146] In this embodiment, an intelligent injection molding workshop scheduling method is used to generate a final production scheduling plan using a fuzzy logic controller. This method particularly emphasizes how to process and evaluate the relationship between the production scheduling plan set and the real-time production data by building a fuzzy logic control system based on a predefined fuzzy rule set. The following information is provided to ensure that those skilled in the art can implement the method according to these instructions.

[0147] First, the construction of the fuzzy logic control system begins with defining a set of fuzzy rules. These rules are based on the specific needs and experience of the injection molding workshop and describe how to evaluate the adaptability of each potential production scheduling scheme based on real-time production data. For example, a rule may state that if the idle time of the machine is short (i.e., the resource utilization is high), the corresponding production scheduling scheme should receive a higher score.

[0148] Each rule in the fuzzy rule set involves how to interpret the relationship between real-time production data (such as environmental parameters, machine status, production task information, etc.) and a specific production scheduling plan. By inputting real-time data into these rules, the fuzzy logic controller can evaluate the adaptability of each plan.

[0149] In order to implement this step, it is necessary to have the ability to design and implement fuzzy logic control systems, and to be able to integrate and process a large amount of real-time production data. Through carefully designed fuzzy rules and efficient data processing mechanisms, step S105 not only enhances the intelligent level of production scheduling, but also brings significant improvements in production efficiency and management benefits to the injection molding workshop.

[0150] The fuzzy logic control system uses the following formula 4 to score the candidate scheduling solutions in the production scheduling solution set:

[0151]

[0152] Among them, Score represents the comprehensive score of the candidate scheduling scheme, which integrates the evaluation results of all relevant fuzzy rules to determine the overall adaptability of the scheme.

[0153] w n is a weight parameter representing the importance of each fuzzy rule in the decision-making process. These weights reflect the priorities of different production goals. For example, reducing idle time may be more important than shortening the average completion time.

[0154] is a fuzzy set A n The corresponding membership function value indicates the degree to which the real-time production data satisfies the conditions of a specific fuzzy rule; for example, if a fuzzy set represents "low idle time", then the membership function defines the evaluation of the satisfaction of this condition by the given real-time data.

[0155] Membership function value is a core concept in fuzzy logic, used to represent the degree of membership of a specific input x in a fuzzy set A n . The range of the degree of membership is from 0 to 1, where 0 means not belonging to the set at all, 1 means completely belonging to the set, and values between 0 and 1 indicate the degree of partial belonging to the set. The following is a specific example to illustrate how to obtain the membership function value.

[0156] Suppose in a production scheduling system of an injection molding workshop, a real-time production data x of concern is the current operating temperature of the machine (in °C). Now define a fuzzy set A n , used to describe the state of "high temperature". To define this fuzzy set, a membership function needs to be specified, which maps the temperature value to the value of the degree of membership.

[0157] Let the range of "high temperature" be defined starting from 50 °C, and the higher the temperature, the higher the degree of belonging to the fuzzy set of "high temperature". A simple linear membership function can be chosen to represent this relationship, as follows:

[0158] When x ≤ 50, This means that if the machine operating temperature is lower than or equal to 50 °C, it is not considered to be in the "high temperature" state at all.

[0159] When 50 < x < 70, Within this range, the membership function value increases linearly, indicating the degree of transition of the machine operating state from "non-high temperature" to "fully high temperature" as the temperature ranges from 50 °C to 70 °C.

[0160] When x ≥ 50, This means that when the machine operating temperature reaches or exceeds 70 °C, it is considered to be in the "high temperature" state completely.

[0161] In this way, for a given machine operating temperature x, the defined membership function can be used to calculate its degree of membership in the fuzzy set A n .

[0162] The application of membership function values is very extensive and is particularly crucial in fuzzy logic control systems. By evaluating the degree of membership of each real-time production data point, the system can make more flexible and adaptable decisions in complex and uncertain environments. In the scenario of injection molding workshop scheduling, this enables the system to dynamically adjust the production scheduling plan according to the real-time machine operating state (such as temperature) to optimize production efficiency and ensure product quality.

[0163] Such membership functions can be adjusted according to actual needs and experience to suit different application scenarios and goals. Through well-designed membership functions, fuzzy logic control systems provide a powerful tool for dealing with uncertainty and ambiguity.

[0164] x is real-time production data, including environmental parameters, machine status, and production task information.

[0165] n is the number of fuzzy rules, which determines the total number of rules considered during the scoring process.

[0166] The following is a specific example to illustrate how to implement the intelligent injection molding workshop scheduling method through fuzzy logic controller.

[0167] Assume that an injection molding workshop needs to optimize production scheduling to improve production efficiency and reduce resource waste. There are multiple injection molding machines in the workshop, various types of product demands, and changing environmental conditions.

[0168] Step 1: Build a fuzzy logic control system

[0169] First, a fuzzy logic control system is established, which can evaluate the production scheduling scheme based on real-time production data and a predefined set of fuzzy rules. For example, the following fuzzy rules are defined:

[0170] - If the machine has a short idle time, the priority is high.

[0171] - If the ambient temperature is moderate, the production efficiency is high.

[0172] -If the production task is urgent, it will have the highest priority.

[0173] Step 2: Define fuzzy sets and membership functions

[0174] In order to apply these rules, it is necessary to define fuzzy sets and corresponding membership functions. For example, for the fuzzy set "machine idle time is short", a membership function can be set, when the idle time is less than 1 hour, the membership is 1 (completely consistent); when the idle time is between 1 and 2 hours, the membership decreases linearly; when it exceeds 2 hours, the membership is 0 (completely inconsistent).

[0175] Step 3: Collect real-time production data

[0176] Assume that at a certain moment, the following real-time data is collected from the workshop:

[0177] - Machine A has been idle for 30 minutes.

[0178] - The ambient temperature is 25°C (considered to be a moderate temperature for high production efficiency).

[0179] -There is a production task whose delivery deadline is approaching.

[0180] Step 4: Use fuzzy logic to control scoring

[0181] Based on the collected data, the fuzzy logic control system is used to score the current production scheduling plan. For each plan, its comprehensive score is calculated:

[0182] -Assume the weight parameter w n They are 0.5, 0.3 and 0.2 respectively.

[0183] - For machine A, the membership of its short idle time is 1 (because the idle time is only 30 minutes), and assume that the weight of this condition is 0.5.

[0184] - The suitability membership of ambient temperature is 1 (because 25°C is considered a moderate temperature) and the weight is 0.3.

[0185] -The priority of urgent tasks is 1 and the weight is 0.2.

[0186] Therefore, for a specific scheduling solution, its comprehensive score is:

[0187] Score=0.5·1+0.3·1+0.2·1=1

[0188] Step 5: Get the final production schedule

[0189] Based on the scores of all candidate solutions, the solution with the highest comprehensive score is selected as the final production scheduling solution. In this example, a solution that ensures efficient utilization of machine A, takes into account ambient temperature, and prioritizes urgent tasks will receive a higher score.

[0190] Through the above process, real-time production data and predefined fuzzy rule sets are effectively applied to the fuzzy logic control system to dynamically evaluate and optimize the production scheduling plan. This method not only improves the production efficiency and resource utilization of the injection molding workshop, but also increases the adaptability to changes in the production environment.

[0191] Furthermore, the fuzzy logic control system includes a mechanism for dynamically updating the fuzzy rule set, and the mechanism adjusts the fuzzy rule set to reflect the latest production scheduling requirements and constraints according to new situations and data emerging in the production process.

[0192] The dynamic update mechanism of fuzzy rule sets in fuzzy logic control systems is designed to ensure that the production scheduling method can adapt to the time-varying demands and constraints in the injection molding workshop, further enhancing the flexibility and adaptability of the system. The following is a detailed description of how to implement this mechanism.

[0193] The core of the dynamic update mechanism is to monitor and identify new situations and data changes in the production process so as to adjust the fuzzy rule set in a timely manner. This process involves several key steps:

[0194] 1. Data collection and analysis: The system continuously collects real-time data during the production process, including but not limited to the operating status of the machine, product quality data, environmental conditions, etc. Through data analysis, the system can identify new production patterns, potential efficiency bottlenecks or quality problems.

[0195] 2. Situation Identification: Based on the collected data, the system uses pre-defined algorithms or models to identify new situations that require special attention or adjustments. For example, if a certain raw material is detected to frequently cause product quality problems, the system will identify this situation as a point that needs adjustment.

[0196] 3. Rule Adjustment: Once a new situation is identified, the system will automatically or prompt the user to manually adjust the fuzzy rule set. Adjustments may include adding new rules, modifying the weights or membership functions of existing rules, or even deleting rules that are no longer applicable. For example, if it is found that the failure rate of a certain machine increases under high temperature conditions, a new rule can be added to reduce the priority of the machine to work at high temperatures.

[0197] 4. Verification and implementation: The adjusted fuzzy rule set will be verified in a simulation environment to ensure that the adjustment effectively improves the performance of production scheduling. After verification, the new rule set will be implemented into the fuzzy logic control system to guide subsequent production scheduling decisions.

[0198] By implementing this dynamic update mechanism, the production scheduling system of the injection molding workshop will be able to respond to changes in the production process more flexibly and accurately, improving production efficiency and product quality. The successful implementation of this mechanism depends on the detailed description and guidance of the above aspects to ensure that technicians can effectively implement the method according to this manual.

[0199] Furthermore, the method of generating a final production scheduling plan based on the production scheduling plan set and real-time production data through a fuzzy logic controller also includes comparing and analyzing the final scheduling plan with historical data, evaluating the degree of improvement of the final scheduling plan relative to the historical plan, and providing suggestions for continuous improvement based on this.

[0200] The selected solution is compared with the solutions implemented in the past. The purpose of this step is to evaluate the degree of improvement of the new solution compared with the previous solution and provide further optimization and improvement suggestions based on this evaluation.

[0201] The following is a detailed description of the implementation process of this step:

[0202] Once the fuzzy logic control system selects a production scheduling plan based on real-time production data, the plan will be automatically compared with the historical scheduling plans stored in the database. This comparison covers many aspects, including but not limited to production efficiency, resource utilization, product quality, machine idle time, and timeliness of task completion. By comparing the performance of the new and old plans on these key performance indicators, the system can accurately evaluate the specific improvements of the new plan over the historical plan.

[0203] The evaluation process involves comparing the performance indicators of the new solution with the average indicators of solutions in similar situations in historical data. For example, if the new solution can significantly reduce the idle time of the machine while maintaining product quality, then this is considered a significant improvement. In this way, the system can identify the specific improvements brought by the new solution and any possible shortcomings.

[0204] Based on the comparative analysis and the evaluation of the degree of improvement, the system then generates a series of specific recommendations aimed at guiding future production scheduling optimization. These recommendations may include adjusting certain production parameters, reconfiguring production lines, optimizing resource allocation, or introducing new workflows. These recommendations are recorded in detail and can be provided to shop floor managers as decision support information to help them make more information-based and data-driven decisions.

[0205] Through the above process, the management team of the injection molding workshop can not only achieve the optimal scheduling in the current production environment, but also continuously learn and improve to ensure that the production process is continuously optimized over time.

[0206] In the above-mentioned embodiment, a smart injection molding workshop scheduling method is provided. Correspondingly, the present application also provides a smart injection molding workshop scheduling system. Figure 2 , which is a schematic diagram of an embodiment of an intelligent injection molding workshop scheduling system of the present application. Since this embodiment, i.e., the second embodiment, is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The system embodiment described below is only illustrative.

[0207] The second embodiment of the present application provides an intelligent injection molding workshop scheduling system, including:

[0208] An initialization unit 201 is used to initialize a set of production scheduling schemes using a particle swarm optimization algorithm, wherein each production scheduling scheme is a particle, and the particle includes a scheduling sequence and resource allocation of each production task in the injection molding workshop;

[0209] A calculation unit 202 is used to calculate the fitness score of each particle based on production efficiency and timeliness of task completion to obtain a fitness score result of each particle, wherein the fitness score result is used to evaluate the performance of each particle;

[0210] The obtaining unit 203 is used to update the position and speed of each particle through the particle swarm optimization algorithm according to the fitness score result of each particle, so as to explore possible individual optimal solutions and group optimal solutions; and obtain a production scheduling solution set by comparing and updating possible individual optimal solutions and group optimal solutions;

[0211] A collecting unit 204 is used to collect real-time production data of the injection molding workshop, wherein the real-time production data includes environmental parameters, machine status, and production task information, and is used to handle uncertainty and fuzziness problems in the production process;

[0212] The generating unit 205 is used to generate a final production scheduling plan through a fuzzy logic controller according to the production scheduling plan set and the real-time production data.

[0213] Furthermore, the computing unit is specifically used for:

[0214] The fitness score F of each particle is calculated using the following formula 1:

[0215]

[0216] Among them, w1, w2 and w3 are weight parameters, which are used to balance the influence of various indicators in the fitness score; T avg Represents the average time required to complete all production tasks; T idle Indicates the total idle time of the machine; T total is the total production time; D late is the number of delayed production tasks; D total is the number of all production tasks.

[0217] Although the present application is disclosed as above in the form of a preferred embodiment, it is not intended to limit the present application. Any technical personnel in this field may make possible changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims of the present application.

Claims

1. An intelligent injection molding workshop scheduling method, characterized in that: include: A set of production scheduling schemes are initialized using a particle swarm optimization algorithm, wherein each production scheduling scheme is a particle, and the particle includes the scheduling sequence and resource allocation of each production task in the injection molding workshop; Calculating a fitness score for each particle based on production efficiency and timeliness of task completion to obtain a fitness score result for each particle, wherein the fitness score result is used to evaluate the performance of each particle; According to the fitness score of each particle, the position and speed of each particle are updated through the particle swarm optimization algorithm to explore possible individual optimal solutions and group optimal solutions; by comparing and updating possible individual optimal solutions and group optimal solutions, a set of production scheduling solutions is obtained; Collect real-time production data of the injection molding workshop, wherein the real-time production data includes environmental parameters, machine status, and production task information, which is used to deal with uncertainty and ambiguity problems in the production process; According to the production scheduling plan set and the real-time production data, a final production scheduling plan is generated through a fuzzy logic controller; Wherein, generating a final production scheduling plan through a fuzzy logic controller according to the production scheduling plan set and real-time production data includes: Constructing a fuzzy logic control system, wherein the fuzzy logic control system is based on a predefined fuzzy rule set, wherein the fuzzy rule set is used to describe the relationship between the production scheduling solution set and the real-time production data; The fuzzy logic control system uses the following formula (4) to score the candidate scheduling solutions in the production scheduling solution set: in, represents the comprehensive score of the candidate scheduling scheme; is the weight parameter, which represents the importance of each fuzzy rule in the decision-making process; is a fuzzy set The corresponding membership function value indicates the degree to which the real-time production data satisfies the conditions of a specific fuzzy rule; It is real-time production data, including environmental parameters, machine status, and production task information; is the number of fuzzy rules; According to the scoring results, the final production scheduling plan is obtained; Wherein, the fuzzy logic control system includes a mechanism for dynamically updating the fuzzy rule set, and the mechanism adjusts the fuzzy rule set to reflect the latest production scheduling requirements and constraints according to new situations and data emerging in the production process; Wherein, generating a final production scheduling plan through a fuzzy logic controller according to the production scheduling plan set and real-time production data also includes comparing and analyzing the final scheduling plan with historical data, evaluating the degree of improvement of the final scheduling plan relative to the historical plan, and providing continuous improvement suggestions accordingly; Among them, the mechanism for dynamically updating the fuzzy rule set is realized through the following steps: Based on the real-time data collected during the production process, use pre-defined algorithms or models to identify new situations that require special attention or adjustments; Once a new situation is identified, the user is prompted to manually adjust the fuzzy rule set; wherein the adjustment of the fuzzy rule set includes adding new rules, modifying the weights or membership functions of existing rules, or even deleting rules that are no longer applicable; The adjusted fuzzy rule set will be verified in a simulation environment. Once verified, the new rule set will be implemented into the fuzzy logic control system to guide subsequent production scheduling decisions.

2. The intelligent injection molding workshop scheduling method according to claim 1, characterized in that: The particle swarm optimization algorithm is used to initialize a group of production scheduling schemes, including randomizing the initial position and speed of each particle to enhance the coverage of the search space and ensure that the optimal solution is searched from a variety of potential solutions.

3. The intelligent injection molding workshop scheduling method according to claim 1, characterized in that: The fitness score of each particle is calculated based on production efficiency and timeliness of task completion to obtain the fitness score result of each particle, including: The fitness score of each particle is calculated using the following formula (1): : in, , and is a weight parameter used to balance the impact of various indicators in the fitness score; Represents the average time required to complete all production tasks; Indicates the total idle time of the machine; is the total production time; is the number of production tasks that are completed late; is the number of all production tasks.

4. The intelligent injection molding workshop scheduling method according to claim 3 is characterized in that: According to the fitness score of each particle, the position and speed of each particle are updated by the particle swarm optimization algorithm to explore possible individual optimal solutions and group optimal solutions, including: The position and velocity of each particle are updated through the following formulas (2) and (3) to explore possible individual optimal solutions and group optimal solutions: in, It is a particle At the next moment speed; It is a particle At the present moment speed; It is a particle At the next moment location; It is a particle At the present moment location; is the inertia weight, which controls the contribution of the particle's current velocity to its new velocity; , and are positive coefficients, representing the influence of individual cognition, social cognition, and fitness score results on speed updating; It is a particle The fitness score result of and is a random number between the interval [0, 1]; It is a particle The best position found so far, i.e. the individual optimal solution; It is the best position found by all particles in the current group, that is, the group optimal solution.

5. The intelligent injection molding workshop scheduling method according to claim 1, characterized in that: The collecting of real-time production data of the injection molding workshop includes utilizing Internet of Things technology to automatically collect data to achieve high-efficiency and high-precision data collection.

Images