Lean production management method and system
By integrating the effective production capacity of bearings into the value flow chart analysis, lean production management methods and systems are built, and problems such as low production efficiency and chaotic inventory management in traditional bearing production management methods are solved, and high-quality and efficient bearing production is achieved.
Patent Information
- Application Number
- CN202510156839.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-12
- Publication Date
- 2025-06-06
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The traditional bearing production management methods have problems such as low production efficiency, chaotic inventory management, excessive resource waste, and difficulty in quality control, which leads to increased production costs, product quality and delivery time being affected, and customer satisfaction is reduced.
A lean production management method and system is proposed to integrate the effective production capacity of bearings into the value flow chart analysis, and obtain the quality standard bias in the order demand data through data collection and processing, building production plan sets and optimization algorithms to obtain real-time optimal production plans, identify and rectify non-value-added activities in the production process, and achieve lean production.
It improves the lean level of bearing production, enhances the production management effect of value flow chart analysis on bearing products, improves product quality and production efficiency, and reduces production costs and inventory management complexity.
Smart Images
Figure CN120106653A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of data processing technology, and more specifically to a lean production management method and system. Background Art
[0002] Bearings are key components in mechanical equipment, and their production quality directly affects the performance and life of the equipment; traditional bearing production management methods have problems such as low production efficiency, chaotic inventory management, high resource waste, and difficult quality control. These problems not only increase production costs, but also affect product quality and delivery time, and reduce customer satisfaction; lean production, as a management concept centered on eliminating waste and improving efficiency, has been widely used in the manufacturing industry; lean production is a production technology and management technology that thoroughly pursues the rationality and efficiency of production, and can flexibly produce high-quality products that meet various needs; the core of lean production is production planning, control, and inventory management.
[0003] As an important tool in lean production management, the core concept of value stream mapping analysis is to comprehensively and systematically identify and eliminate waste in the production process through graphical methods, thereby improving the overall operational efficiency and competitiveness of the enterprise. Therefore, applying value stream mapping analysis to bearing production can effectively improve the lean level of bearing production; however, in the bearing production process, the production quality of bearings is of vital importance. Bearings with different functions have different quality standards. Therefore, the effective production capacity of bearings, that is, the production capacity of bearing products that meet the standards, plays an indispensable role in the lean production management of bearing products.
[0004] In view of this, the present invention proposes a lean production management method and system, which integrates the effective production capacity of bearings into the value stream map analysis, thereby enhancing the production management effect of the value stream map analysis on bearing products. Summary of the invention
[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides a lean production management method and system to solve the problems existing in the above-mentioned background technology.
[0006] The present invention provides the following technical solution: a lean production management method, comprising the following steps:
[0007] Step S01: Data collection and processing: collecting bearing production data and quality data, and pre-processing;
[0008] Step S02: Obtain the quality standard bias corresponding to various bearings in the order demand data, classify the produced bearing products into qualified products, products to be repaired, and unrepairable products, and calculate the first-time qualified rate and production qualified rate of the bearings;
[0009] Step S03: construct a production plan set, predict the first pass rate and production pass rate in combination with the constructed prediction model, and use the optimization algorithm to obtain the real-time optimal production plan;
[0010] Step S04: Draw a VSM diagram based on the raw material data, order demand data, real-time optimal production plan and resource data of the bearing products for analysis, identify various non-value-added activities in the production process and make rectifications;
[0011] Step S05: loop step S03-step S04 until the production process is completely completed.
[0012] Preferably, the production data includes raw material data, order demand data and resource data, the raw material data is the type, quantity and price of raw materials required for bearing production, and the order demand data is the required order quantity and order type; the resource data includes human resources, equipment resources and energy consumption resources; the human resources include the number of employees, employee working hours and employee salaries, the equipment resources include the number of equipment, equipment working hours and equipment performance, and the energy consumption resources are the energy consumed in bearing production; the quality data includes bearing size, shape, vibration frequency, noise value, cleanliness and material type; the preprocessing operation includes cleaning, denoising and normalizing the data.
[0013] Preferably, the specific method of obtaining the quality standard deviation corresponding to various bearings in the order demand data is:
[0014] Construct a bearing knowledge graph, the knowledge graph includes entities, entity attribute information, relationships between entities, and entity types; the entity is a bearing, and the entity attribute information is a label describing the entity attributes, including bearing functions and special quality standards; the relationship between entities represents the semantic connection between different entities, and describes various associations and interactions between entities in the knowledge graph, that is, the relationship between various bearings; the entity type defines the category to which the entity belongs, that is, the type of bearing, and the special quality standard is all quality standard items that are different from the basic quality standard;
[0015] Match the name of the i-th bearing in the order demand data with the entity in the knowledge graph. The matching methods include string matching and semantic matching. String matching includes exact matching and approximate matching. Exact matching is to compare the string corresponding to the name of the i-th bearing in the order demand data with the string corresponding to the entity in the knowledge graph, and match them if they are exactly the same. Approximate matching is to use a string similarity algorithm to calculate the similarity between the string corresponding to the name of the i-th bearing in the order demand data and the string corresponding to the entity in the knowledge graph, and match the entity in the knowledge graph with the highest similarity with the name of the i-th bearing in the order demand data. Semantic matching is to use a word embedding method. The word embedding method converts the name of the i-th bearing in the order demand data and the entity in the knowledge graph into vector form through a word embedding model, and matches them by calculating the similarity between the vectors.
[0016] After the match is successful, the successfully matched entity in the knowledge graph is marked as a matched entity, that is, a successfully matched bearing. The matched entities obtained by the two methods are merged, and the entity attribute information corresponding to the matched entity is obtained from the knowledge graph to obtain the function and special quality standard of the bearing corresponding to the matched entity. The quality standard item corresponding to the special quality standard is used as the quality standard bias of the i-th bearing in the order demand data.
[0017] In this way, the quality standard deviation corresponding to each bearing in the order demand data is obtained. If there are n types of bearings in total, then i = 1, 2, 3, ..., n.
[0018] Preferably, the specific method of determining whether the bearing products to be produced are qualified products, products to be repaired, and products that cannot be repaired is:
[0019] If there are n types of bearings in total, and the i-th type of bearing has a total of P bearing products, the bearings that meet the basic quality standards are marked as candidate products, and the bearings that do not meet the basic quality standards are marked as scrapped products. The candidate products corresponding to each bearing are obtained, and the bearing products are further classified into qualified products, products to be repaired, and non-repairable products among the candidate products based on the quality standard bias corresponding to each bearing;
[0020] The quality standard deviation corresponding to the i-th bearing is denoted as U i , U i = {u i1 ,u i2 ,u i3 ,...,u im}, where u ijis the jth quality standard bias element in the quality standard bias corresponding to the i-th bearing, that is, the jth quality standard item in the special quality standard corresponding to the quality standard bias. If the quality standard bias includes the accuracy standard and the high temperature resistance standard, the accuracy standard and the high temperature resistance standard are both quality standard items, j = 1, 2, 3, ..., m; each quality standard bias element corresponds to a quality standard formula; if there is a bearing product in the candidate product set that meets all the corresponding quality standard bias elements, the bearing is marked as a qualified product; if there is a bearing product in the candidate product set that does not meet all the corresponding quality standard bias elements, a threshold judgment is performed;
[0021] The specific method of determining the threshold is as follows:
[0022] Set a repairable threshold for all quality standard bias elements in each quality standard bias, subtract the actual value corresponding to the produced bearing product from the formula value of each quality standard bias element, and make an absolute value. If there is a difference whose absolute value exceeds the repairable threshold, the bearing product is marked as an unrepairable product; if all the absolute values of the difference do not exceed the repairable threshold, the bearing product is marked as a product to be repaired;
[0023] Classify P bearing products until all the bearing products produced for the i-th bearing are classified into qualified products, products to be repaired, and unrepairable products; classify n bearings in turn until all the bearing products produced for the n bearings are classified into qualified products, products to be repaired, and unrepairable products.
[0024] Preferably, the basic quality standards all correspond to quality standard formulas, and the quality standard formulas are any one of inequalities and equations, that is, numerical inequalities and numerical equations of quality standard items. If the quality standard formula is an inequality, the formula value is the value of the inequality; if the quality standard formula is an equation, the formula value is the value of the equation; the quality standard bias is the quality standard that the bearing needs to pay special attention to in addition to the basic quality standard, and the basic quality standard is the quality standard that all bearings need to meet.
[0025] Preferably, the calculation formula for the first pass rate and production pass rate of the bearing production is expressed as:
[0026] Among them, HG i _one is the first pass rate of the i-th bearing, SL i _hg is the number of qualified products of the i-th type of bearing, SL i _z is the total number of products of the i-th type of bearings, i.e., the sum of the number of qualified products, products to be repaired, unrepairable products, and scrapped products;
[0027] Among them, HGi _sc is the production qualification rate of the i-th bearing, SL i _dx is the number of bearings of the i-th type to be repaired, i = 1, 2, 3, ..., n, and n is the total number of bearing types.
[0028] Preferably, the production plan set is a collection of production plans, and the production plans are obtained in the following manner:
[0029] Obtain order demand data and generate multiple different production plans based on the order product quantity, current time, and delivery time. Each bearing corresponds to a production plan set, so there are n production plan sets corresponding to n types of bearings.
[0030] The use of an optimization algorithm to obtain a real-time optimal production plan includes the following steps:
[0031] Step S11: automatically generate n production plan sets;
[0032] Step S12: Encode each production plan in the i-th production plan set, obtain chromosomes, and construct an initial population;
[0033] Step S13: determining the fitness function;
[0034] Step S14: performing natural selection on chromosomes in the population;
[0035] Step S15: performing crossover recombination on chromosomes in the population;
[0036] Step S16: mutating the chromosomes in the population;
[0037] Step S17: Obtain a new population, preset the population generation number to be L, the fitness threshold to be Q, L is an integer greater than 0, and Q is a real number greater than 0; loop through steps S14 to S16 until the generation number corresponding to the new population is L or the fitness corresponding to a chromosome in the new population is greater than or equal to the fitness threshold Q, the loop ends, and the production plan corresponding to the chromosome with the maximum fitness in the new population is taken as the real-time optimal production plan of the i-th production plan set;
[0038] Step S18: loop steps S12 to S17 until all n production plan sets obtain corresponding real-time optimal production plans.
[0039] Preferably, in step S11, each production plan is encoded as U, where U is a chromosome, and R chromosomes are randomly generated to form an initial population C i , C i = {U i1 ,U i2 ,U i3 ,...,U iR}; Among them, UiR is the Rth chromosome in the production plan of the i-th bearing;
[0040] The expression of the fitness function is: Among them, SY ir is the fitness of the production plan corresponding to the rth chromosome in the i-th production plan set, HG ir _one y HG is the predicted value of the first pass rate of the production plan corresponding to the rth chromosome in the i-th production plan set, ir _sc y is the predicted value of the production qualification rate of the production plan corresponding to the rth chromosome in the i-th production plan set;
[0041] The method for obtaining the first pass rate prediction value and the production pass rate prediction value of the production plan corresponding to the rth chromosome in the i-th production plan set is:
[0042] Input the production plan, resource data and quality data corresponding to the rth chromosome into the first pass rate prediction model of the ith production plan set, output the first pass rate prediction value of the daily production in the production plan corresponding to the rth chromosome in the ith production plan set, and obtain the final first pass rate prediction value after weighted average;
[0043] The production plan corresponding to the rth chromosome is input into the prediction model of the proportion of products to be repaired in the i-th production plan set, and the proportion of products to be repaired produced daily in the production plan corresponding to the rth chromosome in the i-th production plan set is output. The proportion of products to be repaired produced daily and the predicted value of the first qualified rate of daily production are added to obtain the daily production qualified rate prediction value, and the final production qualified rate prediction value is obtained after weighted average.
[0044] Preferably, the one-time pass rate prediction model and the product proportion prediction model for repair are both deep neural network models, and the model training processes of the two are the same;
[0045] The specific training process of the first pass rate prediction model is as follows:
[0046] Pre-collecting d groups of analysis data of the i-th type of bearing, the analysis data being resource data, quality data and bearing production quantity, collecting d groups of analysis data, d being an integer greater than 1, converting the analysis data and the corresponding first-pass rate into a corresponding set of feature vectors, the first-pass rate being obtained by dividing the number of qualified products in the bearing production quantity by the number of bearings produced continuously, in the same way as calculated in step S02;
[0047] Each set of feature vectors is used as the input of the first-pass rate prediction model. The first-pass rate prediction model uses a set of first-pass rate prediction values corresponding to each set of analysis data as output, and uses the actual first-pass rate corresponding to each set of analysis data as the prediction target. The actual first-pass rate is the first-pass rate corresponding to the analysis data collected in advance; the training target is to minimize the sum of the prediction errors of all analysis data; the formula of the prediction error is expressed as: ε q =θ q -μ q , where ε q is the prediction error, q is the group number of the eigenvector corresponding to the analyzed data, θ q is the predicted value of the first pass rate corresponding to the qth group of analysis data, μ q is the actual first-time qualified rate corresponding to the qth group of analysis data, and the first-time qualified rate prediction model is trained until the sum of the prediction errors reaches convergence, and the training is stopped; q = 1, 2, 3, ..., d;
[0048] The prediction model for the proportion of products to be repaired is trained in the same way as the one-time pass rate prediction model. The proportion of products to be repaired is obtained by dividing the number of products to be repaired in the number of bearings produced by the number of bearings produced.
[0049] A lean production management system, comprising a data acquisition module, a data preprocessing module, a data analysis module, a data prediction module, a lean production analysis module and a human-computer interaction module;
[0050] The data acquisition module is used to collect bearing production data and quality data, and transmit them to the data preprocessing module for preprocessing;
[0051] The data preprocessing module is used to receive data from the data acquisition module, preprocess the data and then transmit it to the data analysis module and the data prediction module;
[0052] The data analysis module obtains the quality standard deviations corresponding to various bearings in the order demand data, classifies the produced bearing products into qualified products, products to be repaired, and unrepairable products, and calculates the first-time qualified rate and production qualified rate of the bearings and transmits them to the data prediction module;
[0053] The data prediction module is used to construct a production plan set, predict the first-time qualified rate and the production qualified rate in combination with the constructed prediction model, and use the optimization algorithm to obtain the real-time optimal production plan and transmit it to the lean production analysis module;
[0054] The lean production analysis module draws a VSM diagram based on the raw material data, order demand data, real-time optimal production plan and resource data of the bearing products for analysis, identifies various non-value-added activities in the production process and makes rectifications;
[0055] The human-computer interaction module is used to perform human-computer interaction display on the data.
[0056] Technical effects and advantages of the present invention:
[0057] (1) The present invention is provided with step S02, which is conducive to arranging the production plan of bearing products by obtaining the production status of bearings, and laying a foundation for the subsequent acquisition of real-time optimal production plans; different bearing products have different functions, different quality standard inclinations, and different materials used. Bearings used for precision machinery have higher accuracy requirements than ordinary bearings, and the quality standard tends to be bearing accuracy; bearings used to bear larger loads have higher wear resistance requirements than ordinary bearings, and the quality standard tends to be wear resistance; bearings used for high-speed operation, such as motor bearings, have higher noise and vibration requirements than ordinary bearings, and the materials need to have wear resistance and high temperature resistance, and the quality standard tends to be wear resistance and high temperature resistance; based on the use scenarios of bearings with different functions, the quality of bearing products is judged, which improves the accuracy of bearing product quality judgment, avoids the phenomenon of lax product quality inspection caused by a one-size-fits-all approach, improves the accuracy of daily production data of bearing products, and thereby improves the accuracy of production rhythm.
[0058] (2) The present invention is provided with step S03, which is beneficial to obtaining the real-time optimal production plan, thereby improving the flexibility of value stream map analysis and avoiding lag. At the same time, according to the changes in the daily production quantity of bearing products, the production plan and improvement measures can be flexibly adjusted, and the product production plan can be updated in real time to continuously improve the production process and improve product quality and production efficiency. For different production plans, the number of bearing products produced daily is different, so different equipment configurations and personnel configurations need to be allocated. When conducting value stream map analysis, the specific details corresponding to the production plan provide a clear analysis scope and boundary for the analysis. Analysts can determine which equipment and personnel activities need to be included in the value stream map based on the production plan; at the same time, the neural network model is integrated, and the trained neural network model is integrated into the fitness calculation of the optimization algorithm, giving full play to the nonlinear and multimodal data modeling capabilities of the neural network, and effectively capturing complex data relationships; based on the parallel computing processing mechanism, the computing efficiency is improved, thereby ensuring the comprehensiveness and accuracy of the analysis. BRIEF DESCRIPTION OF THE DRAWINGS
[0059] Figure 1 The present invention is a flow chart of a lean production management method.
[0060] Figure 2 This is a structural diagram of a lean production management system of the present invention. DETAILED DESCRIPTION
[0061] The technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the present invention. In addition, the forms of the various structures recorded in the following embodiments are merely illustrative. The lean production management method and system involved in the present invention are not limited to the various structures recorded in the following embodiments. All other implementations obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0062] like Figure 1 As shown, the present invention provides a lean production management method, comprising the following steps:
[0063] Step S01: Data collection and processing: The bearing production data and quality data are collected and preprocessed; the production data includes but is not limited to raw material data, order demand data and resource data, the raw material data is the type, quantity and price of raw materials required for bearing production, the order demand data is the required order quantity and order type, specifically the order type and corresponding order quantity of bearings; the resource data includes but is not limited to human resources, equipment resources and energy consumption resources; the human resources include but are not limited to the number of employees, employee working hours and employee salaries, the equipment resources include but are not limited to the number of equipment, equipment working hours and equipment performance, and the energy consumption resources are the energy consumed by bearing production; the quality data includes but is not limited to bearing size, shape, vibration frequency, noise value, cleanliness and material type; the preprocessing operation includes but is not limited to cleaning, denoising and normalizing the data to obtain data that can be directly used;
[0064] Step S02: Obtain the quality standard bias corresponding to various bearings in the order demand data, classify the produced bearing products into qualified products, products to be repaired and non-repairable products, and calculate the first-time qualified rate and production qualified rate of the bearings; the quality standard bias is the quality standard that the bearing needs to pay special attention to in addition to the basic quality standard, and the basic quality standard is the quality standard that all bearings need to meet; its purpose is to arrange the production plan of the bearing products by obtaining the production situation of the bearings, laying the foundation for the subsequent acquisition of the real-time optimal production plan; different bearing products have different functions, different quality standard biases, and different materials. The requirements for precision of bearings used in precision machinery will be higher than those of ordinary bearings, so the quality standard bias Bearing precision; Bearings used to bear larger loads have higher requirements for wear resistance than ordinary bearings, so the quality standard tends to be wear resistance; Bearings used for high-speed operation, such as motor bearings, have higher requirements for noise and vibration than ordinary bearings, and the materials need to have wear resistance and high temperature resistance, so the quality standard tends to be wear resistance and high temperature resistance; Based on the use scenarios of bearings with different functions, the quality of bearing products is judged, which improves the accuracy of bearing product quality judgment, avoids the phenomenon of lax product quality inspection caused by a one-size-fits-all approach, improves the accuracy of daily production data of bearing products, and thus improves the accuracy of production rhythm. In value stream map analysis, production rhythm is a key indicator used to evaluate whether the efficiency of the production process meets customer needs;
[0065] Step S03: Construct a set of production plans, predict the first-time qualified rate and the production qualified rate in combination with the constructed prediction model, and use the optimization algorithm to obtain the real-time optimal production plan; the production plan is the number of bearing products produced daily; its purpose is to improve the flexibility of value stream map analysis and avoid lag by obtaining the real-time optimal production plan, and at the same time, flexibly adjust the production plan and improvement measures according to the changes in the daily production quantity of bearing products, and update the product production plan in real time to continuously improve the production process, improve product quality and production efficiency. For different production plans, the number of bearing products produced daily is different, so different equipment configurations and personnel configurations need to be allocated. When conducting value stream map analysis, the specific details corresponding to the production plan provide a clear analysis scope and boundary for the analysis. Analysts can determine which equipment and personnel activities need to be included in the value stream map based on the production plan, thereby ensuring the comprehensiveness and accuracy of the analysis;
[0066] Step S04: Based on the raw material data, order demand data, real-time optimal production plan and resource data of the bearing product, a VSM diagram is drawn for analysis to identify various waste phenomena in the production process and make rectifications. The waste phenomenon is non-value-added activities, which are production activities that cannot add value to the final product. The waste phenomenon includes but is not limited to waiting, handling, inventory backlog, over-processing and rework of defective products, and the waste phenomenon is eliminated or reduced; VSM is an important tool for lean production management, and the production plan has a significant impact on the value stream map analysis. Due to its use characteristics, bearing products have different functions and use occasions. The bearings have different quality requirements. Therefore, the quality inspection of the product cannot be generalized. Obtaining the corresponding real-time optimal production plan for bearing products with different functions can ensure the real-time and accuracy of the value stream map, and improve the effectiveness and comprehensiveness of the lean production management of bearing products; the drawing of the VSM diagram is a prior art, and this embodiment will not go into details;
[0067] Step S05: loop step S03-step S04 until the production process is completely completed; analyze the bearing production management in combination with the real-time production plan and the VSM diagram, and perform lean management for different production plans. Different production plans have different equipment, personnel and process arrangements, so the non-value-added activities in the production process are also different. The real-time optimal production plan ensures production efficiency and reduces production energy consumption. VSM ensures the lean level of production. Therefore, the combination of the two can effectively enhance the management effect of lean production.
[0068] In this embodiment, it should be specifically explained that the specific method of obtaining the quality standard deviations corresponding to various bearings in the order demand data is as follows:
[0069] Construct a bearing knowledge graph, the knowledge graph includes entities, entity attribute information, relationships between entities and entity types; the entity is a bearing, and the entity attribute information is a label that describes the entity attributes, including but not limited to bearing functions and special quality standards; the relationship between entities represents the semantic connection between different entities, and describes various associations and interactions between entities in the knowledge graph, that is, the relationship between various bearings; the entity type defines the category to which the entity belongs, that is, the type of bearing, such as rotary bearings, linear bearings, and high-temperature bearings; the special quality standard is all quality standard items that are different from the basic quality standard. For example, if the accuracy standard in the basic quality standard of the bearing is the inequality a<a1, then the bearings whose accuracy meets the inequality are all bearings that meet the quality standard, that is, bearings that meet the quality standards. At this time, the accuracy standard in the quality standard item of bearing A is the inequality a<b1, b1 is not equal to a1, then the accuracy standard in the quality standard item is the special quality standard, and if there are multiple quality standard items that are special quality standards at the same time, then the set composed of all quality standard items is the final special quality standard;
[0070] Match the name of the i-th bearing in the order demand data with the entity in the knowledge graph. The matching methods include string matching and semantic matching. String matching includes exact matching and approximate matching. Exact matching is to compare the string corresponding to the name of the i-th bearing in the order demand data with the string corresponding to the entity in the knowledge graph, and match them if they are exactly the same. Approximate matching is to use a string similarity algorithm to calculate the similarity between the string corresponding to the name of the i-th bearing in the order demand data and the string corresponding to the entity in the knowledge graph, and match the entity in the knowledge graph with the highest similarity with the name of the i-th bearing in the order demand data. Semantic matching is to use a word embedding method. The word embedding method converts the name of the i-th bearing in the order demand data and the entity in the knowledge graph into vector form through a word embedding model, and matches them by calculating the similarity between the vectors.
[0071] After the match is successful, the successfully matched entity in the knowledge graph is marked as a matched entity, that is, a successfully matched bearing. The matched entities obtained by the two methods are merged, and the entity attribute information corresponding to the matched entity is obtained from the knowledge graph to obtain the function and special quality standard of the bearing corresponding to the matched entity. The quality standard item corresponding to the special quality standard is used as the quality standard bias of the i-th bearing in the order demand data.
[0072] In this way, the quality standard deviation corresponding to each bearing in the order demand data is obtained. If there are n types of bearings in total, then i = 1, 2, 3, ..., n.
[0073] In this embodiment, it should be specifically explained that the specific method of determining whether the bearing products to be produced are qualified products, products to be repaired, and products that cannot be repaired is as follows:
[0074] If there are n types of bearings in total, and the i-th type of bearing has a total of P bearing products produced, the bearings that meet the basic quality standards are marked as candidate products, and the bearings that do not meet the basic quality standards are marked as scrapped products. The candidate products corresponding to each bearing are obtained, and the bearing products are further classified into qualified products, products to be repaired, and non-repairable products among the candidate products based on the quality standard bias corresponding to each bearing; the basic quality standards all correspond to quality standard formulas, and the quality standard formulas can be inequalities or equations, that is, numerical inequalities or numerical equations of quality standard items. If the quality standard formula is an inequality, the formula value is the value of the inequality, and if the quality standard formula is an equation, the formula value is the value of the equation;
[0075] The quality standard deviation corresponding to the i-th bearing is denoted as U i , U i = {u i1 ,u i2 ,u i3 ,...,u im}, where u ij is the jth quality standard bias element in the quality standard bias corresponding to the i-th bearing, that is, the jth quality standard item in the special quality standard corresponding to the quality standard bias. If the quality standard bias includes the accuracy standard and the high temperature resistance standard, the accuracy standard and the high temperature resistance standard are both quality standard items, j = 1, 2, 3, ..., m; each quality standard bias element corresponds to a quality standard formula; if there is a bearing product in the candidate product set that meets all the corresponding quality standard bias elements, the bearing is marked as a qualified product; if there is a bearing product in the candidate product set that does not meet all the corresponding quality standard bias elements, a threshold judgment is performed;
[0076] The specific method of determining the threshold is as follows:
[0077] A repairable threshold is set for all quality standard bias elements in each quality standard bias, and the actual value corresponding to the produced bearing product is subtracted from the formula value of each quality standard bias element, and the absolute value is calculated. If there is a difference whose absolute value exceeds the repairable threshold, the bearing product is marked as an unrepairable product; if all the difference absolute values do not exceed the repairable threshold, the bearing product is marked as a product to be repaired; the repairable threshold can be set by a technician in this field according to the actual bearing product and combined with professional technical knowledge. If the absolute value of the difference between the formula value of the quality standard bias element and the actual value corresponding to the bearing product does not exceed the repairable threshold, the bearing product can be repaired by secondary processing to ensure the performance and life of the bearing. If the absolute value of the difference between the formula value of the quality standard bias element and the actual value corresponding to the bearing product exceeds the repairable threshold, the bearing product cannot be repaired. At this time, the bearing product cannot be repaired and may have serious defects. The repaired bearing cannot guarantee its normal performance, so it cannot be produced and used; for example, if the quality standard bias corresponding to the i-th bearing is U i = {u i1 ,u i2},u i1 Bias element for the first quality criterion: accuracy, u i2 is the second quality standard bias element: high temperature strength. If the actual values of the quality standard bias elements corresponding to the p-th bearing product in the candidate set i-th bearing are: accuracy B, high temperature strength C, u i1 The quality standard formula is expressed as u i1 ≤B′, then u i1 The formula value is B′, u i2 The quality standard formula is expressed as u i1 ≥C′, then u i2 The formula value is C′, and the repairable limit value of the first quality standard bias element is set to B * The second quality standard is the repairability limit of the element. * ; If B is less than or equal to B', and C is greater than or equal to C', then the pth bearing product is a qualified product, otherwise a threshold judgment is performed, if |B'-B|≤B * , |C′-C|≤C * , then the pth bearing product is recorded as a product to be repaired, otherwise it is recorded as an unrepairable product; p = 1, 2, 3, ..., P;
[0078] Classify P bearing products until all the bearing products produced for the i-th bearing are classified into qualified products, products to be repaired, and unrepairable products; classify n bearings in sequence until all the bearing products produced for n bearings are classified into qualified products, products to be repaired, and unrepairable products;
[0079] The quality standard bias element may be a feature representing the quality standard bias element. If there are multiple features representing the quality standard bias element, any one feature may be selected. By way of example, if there is a quality standard bias element which is crack degree, the feature representing the quality standard bias element may be any one of crack depth, crack length or crack width. Then the quality standard bias element may be any one of crack depth, crack length or crack width. The crack degree may also be subdivided into three types: crack depth, crack length and crack width, that is, crack depth, crack length and crack width are all quality standard bias elements. The specific setting method may be set by technicians in this field.
[0080] In this embodiment, it should be specifically explained that the calculation formula of the first pass rate and the production pass rate of the bearing production is expressed as:
[0081] Among them, HG i _one is the first pass rate of the i-th bearing, SL i _hg is the number of qualified products of the i-th type of bearing, SL i _z is the total number of products of the i-th type of bearings, i.e., the sum of the number of qualified products, products to be repaired, unrepairable products, and scrapped products;
[0082] Among them, HG i _sc is the production qualification rate of the i-th bearing, SL i _dx is the number of bearings of the i-th type to be repaired, i = 1, 2, 3, ..., n, and n is the total number of bearing types.
[0083] In this embodiment, it should be specifically explained that the production plan set is a collection of production plans, and the production plan is obtained by: obtaining order demand data, generating multiple different production plans based on the order product quantity, current time and delivery time, and each production plan can complete the order product quantity within the delivery time; the production plans are all production plans from the current time to the day before the delivery time; each bearing corresponds to a production plan set, so there are n production plan sets corresponding to n types of bearings; because different types of bearings have different quality standards during production, the process parameters during production are different, and thus the production plans cannot be universal, and setting corresponding production plans for different types of bearings can improve the accuracy of lean production management and lay the foundation for subsequent value stream map analysis;
[0084] The optimization algorithm for obtaining the real-time optimal production plan includes the following steps:
[0085] Step S11: automatically generate n production plan sets;
[0086] Step S12: Encode each production plan in the i-th production plan set, obtain chromosomes, and construct an initial population;
[0087] Step S13: determining the fitness function;
[0088] Step S14: performing natural selection on chromosomes in the population;
[0089] Step S15: performing crossover recombination on chromosomes in the population;
[0090] Step S16: mutating the chromosomes in the population;
[0091] Step S17: obtain a new population, preset the population generation number to be L, the fitness threshold to be Q, L is an integer greater than 0, and Q is a real number greater than 0; loop step S14-step S16 until the generation number corresponding to the new population is L or the fitness corresponding to the chromosome in the new population is greater than or equal to the fitness threshold Q, the loop ends, and the production plan corresponding to the chromosome with the maximum fitness in the new population is used as the real-time optimal production plan of the i-th production plan set; illustratively, if the preset population generation number is 1, the chromosomes in the initial population are subjected to natural selection, crossover recombination and a new population is obtained, at which time the generation number corresponding to the new population is 1, so the loop ends;
[0092] Step S18: loop steps S12 to S17 until all n production plan sets obtain corresponding real-time optimal production plans.
[0093] The fitness threshold Q is preset by those skilled in the art according to the accuracy of the algorithm. The population algebra L is obtained by those skilled in the art under multiple sets of different production plan data conditions by using a genetic algorithm multiple times to obtain the real-time optimal production plan. In each process of using the genetic algorithm, when the fitness corresponding to the chromosome in the new population is greater than or equal to the fitness threshold Q, the cycle ends and the algebra corresponding to the new population is obtained; the largest algebra among the multiple algebras is taken as the population algebra L.
[0094] In this embodiment, it should be specifically explained that in step S11, each production plan is encoded as U, where U is a chromosome, and R chromosomes are randomly generated to form an initial population C. i , C i = {U i1 ,U i2 ,U i3 ,...,U iR}; Among them, U iR is the Rth chromosome in the production plan of the i-th bearing;
[0095] The expression of the fitness function is: Among them, SYir is the fitness of the production plan corresponding to the rth chromosome in the i-th production plan set, HG ir _one y HG is the predicted value of the first pass rate of the production plan corresponding to the rth chromosome in the i-th production plan set, ir _sc y is the predicted value of the production qualification rate of the production plan corresponding to the rth chromosome in the i-th production plan set;
[0096] The method for obtaining the first pass rate prediction value and the production pass rate prediction value of the production plan corresponding to the rth chromosome in the i-th production plan set is:
[0097] Input the production plan, resource data and quality data corresponding to the rth chromosome into the first pass rate prediction model of the ith production plan set, output the first pass rate prediction value of the daily production in the production plan corresponding to the rth chromosome in the ith production plan set, and obtain the final first pass rate prediction value after weighted average;
[0098] The production plan corresponding to the rth chromosome is input into the prediction model of the proportion of products to be repaired in the i-th production plan set, and the proportion of products to be repaired produced daily in the production plan corresponding to the rth chromosome in the i-th production plan set is output. The proportion of products to be repaired produced daily and the predicted value of the first qualified rate of daily production are added to obtain the daily production qualified rate prediction value, and the final production qualified rate prediction value is obtained after weighted average.
[0099] The first-pass rate prediction model and the product proportion prediction model for repair are both deep neural network models; they include an input layer, a hidden layer, and an output layer; each hidden layer includes multiple neurons, each neuron is connected to the neurons in the next layer, and the connection contains weights to determine the importance and influence of data transmission in the neural network; each neuron between the hidden layer and the output layer applies an activation function, which maps nonlinearity to allow the network to learn more complex patterns and features; and the model training process of the two is the same;
[0100] The specific training process of the first pass rate prediction model is as follows:
[0101] Pre-collecting d groups of analysis data of the i-th type of bearing, the analysis data being resource data, quality data and bearing production quantity, collecting d groups of analysis data, d being an integer greater than 1, converting the analysis data and the corresponding first-pass rate into a corresponding set of feature vectors, the first-pass rate being obtained by dividing the number of qualified products in the bearing production quantity by the number of bearings produced continuously, in the same way as calculated in step S02;
[0102] Each set of feature vectors is used as the input of the first-pass rate prediction model. The first-pass rate prediction model uses a set of first-pass rate prediction values corresponding to each set of analysis data as output, and uses the actual first-pass rate corresponding to each set of analysis data as the prediction target. The actual first-pass rate is the first-pass rate corresponding to the analysis data collected in advance; the training target is to minimize the sum of the prediction errors of all analysis data; the formula of the prediction error is expressed as: ε q =θ q -μ q , where ε q is the prediction error, q is the group number of the eigenvector corresponding to the analyzed data, θ q is the predicted value of the first pass rate corresponding to the qth group of analysis data, μ q is the actual first-time qualified rate corresponding to the qth group of analysis data, and the first-time qualified rate prediction model is trained until the sum of the prediction errors reaches convergence, and the training is stopped; q = 1, 2, 3, ..., d;
[0103] The training process of the prediction model for the proportion of products to be repaired is the same as that of the prediction model for the first qualified rate. The proportion of products to be repaired is obtained by dividing the number of products to be repaired in the number of bearings produced by the number of bearings produced. The rest will not be described in detail.
[0104] The weighted average of the daily first-time pass rate prediction value and the production pass rate prediction value in the production plan is taken to obtain the final first-time pass rate prediction value and the production pass rate prediction value, and used as the calculation of the final fitness function, so that the final real-time optimal production plan has the best product compliance, which improves production efficiency and production quality.
[0105] In this embodiment, it should be specifically explained that the elite method generates F1 offspring chromosomes. For a population with a capacity of R, the fitness corresponding to the R chromosomes is arranged from large to small, and each of the F1 chromosomes at the front generates a offspring chromosome; the rotation method generates F2 offspring chromosomes, that is, R chromosomes generate F2 offspring chromosomes according to the corresponding rotation probability; F1+F2=R, so as to keep the offspring population capacity R unchanged and the population increases algebraically;
[0106] The expression of the rotation probability is: Among them, ir is the rotation probability of the production plan corresponding to the rth chromosome in the i-th production plan set;
[0107] The crossover recombination adopts the PMX method, randomly selects E chromosomes in the population for crossover recombination, and obtains E new chromosomes; after the chromosome crossover recombination, calculates the fitness of the E new chromosomes, sorts the fitness of the E new chromosomes and the fitness of the E chromosomes from large to small and generates a sorting table, and replaces the E chromosomes in the population for crossover recombination with the E new chromosomes in the sorting table in positive order; in this embodiment, preferably E=0.7R, if the calculated E is not an integer, round E up to ensure that the calculated E is an integer;
[0108] In step S16, the mutation probability is preset as V, and the R chromosomes in the population are mutated according to the mutation probability. The mutation method is to randomly select the positions of two genes in the chromosome and exchange the values of the two genes. In this embodiment, V=0.02 is preferred. The mutation probability is preset by technicians in this field according to the algorithm efficiency and algorithm accuracy.
[0109] like Figure 2 As shown, the present invention provides a lean production management system, including a data acquisition module, a data preprocessing module, a data analysis module, a data prediction module, a lean production analysis module and a human-computer interaction module;
[0110] The data acquisition module is used to collect bearing production data and quality data, and transmit them to the data preprocessing module for preprocessing;
[0111] The data preprocessing module is used to receive data from the data acquisition module, preprocess the data and then transmit it to the data analysis module and the data prediction module;
[0112] The data analysis module obtains the quality standard deviations corresponding to various bearings in the order demand data, classifies the produced bearing products into qualified products, products to be repaired, and unrepairable products, and calculates the first-time qualified rate and production qualified rate of the bearings and transmits them to the data prediction module;
[0113] The data prediction module is used to construct a production plan set, predict the first-time qualified rate and the production qualified rate in combination with the constructed prediction model, and use the optimization algorithm to obtain the real-time optimal production plan and transmit it to the lean production analysis module;
[0114] The lean production analysis module draws a VSM diagram based on the raw material data, order demand data, real-time optimal production plan and resource data of the bearing products for analysis, identifies various non-value-added activities in the production process and makes rectifications;
[0115] The human-computer interaction module is used to perform human-computer interaction display on the data.
[0116] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
[0117] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A lean production management method, characterized in that: The following steps are involved: Step S01: Data collection and processing: collecting bearing production data and quality data, and pre-processing; Step S02: Obtain the quality standard bias corresponding to various bearings in the order demand data, classify the produced bearing products into qualified products, products to be repaired, and unrepairable products, and calculate the first-time qualified rate and production qualified rate of the bearings; Step S03: construct a production plan set, predict the first pass rate and production pass rate in combination with the constructed prediction model, and use the optimization algorithm to obtain the real-time optimal production plan; Step S04: Draw a VSM diagram based on the raw material data, order demand data, real-time optimal production plan and resource data of the bearing products for analysis, identify various non-value-added activities in the production process and make rectifications; Step S05: loop step S03-step S04 until the production process is completely completed.
2. A lean production management method according to claim 1, characterized in that: The production data includes raw material data, order demand data and resource data. The raw material data is the type, quantity and price of raw materials required for bearing production, and the order demand data is the required order quantity and order type; the resource data includes human resources, equipment resources and energy consumption resources; the human resources include the number of employees, employee working hours and employee salaries, the equipment resources include the number of equipment, equipment working hours and equipment performance, and the energy consumption resources are the energy consumed in bearing production; the quality data includes bearing size, shape, vibration frequency, noise value, cleanliness and material type; the preprocessing operation includes cleaning, denoising and normalizing the data.
3. A lean production management method according to claim 2, characterized in that: The specific method of obtaining the quality standard deviation corresponding to various bearings in the order demand data is: Construct a bearing knowledge graph, the knowledge graph includes entities, entity attribute information, relationships between entities, and entity types; the entity is a bearing, and the entity attribute information is a label describing the entity attributes, including bearing functions and special quality standards; the relationship between entities represents the semantic connection between different entities, and describes various associations and interactions between entities in the knowledge graph, that is, the relationship between various bearings; the entity type defines the category to which the entity belongs, that is, the type of bearing, and the special quality standard is all quality standard items that are different from the basic quality standard; Match the name of the i-th bearing in the order demand data with the entity in the knowledge graph. The matching methods include string matching and semantic matching. String matching includes exact matching and approximate matching. Exact matching is to compare the string corresponding to the name of the i-th bearing in the order demand data with the string corresponding to the entity in the knowledge graph, and match them if they are exactly the same. Approximate matching is to use a string similarity algorithm to calculate the similarity between the string corresponding to the name of the i-th bearing in the order demand data and the string corresponding to the entity in the knowledge graph, and match the entity in the knowledge graph with the highest similarity with the name of the i-th bearing in the order demand data. Semantic matching is to use a word embedding method. The word embedding method converts the name of the i-th bearing in the order demand data and the entity in the knowledge graph into vector form through a word embedding model, and matches them by calculating the similarity between the vectors. After the match is successful, the successfully matched entity in the knowledge graph is marked as a matched entity, that is, a successfully matched bearing. The matched entities obtained by the two methods are merged, and the entity attribute information corresponding to the matched entity is obtained from the knowledge graph to obtain the function and special quality standard of the bearing corresponding to the matched entity. The quality standard item corresponding to the special quality standard is used as the quality standard bias of the i-th bearing in the order demand data. In this way, the quality standard deviation corresponding to each bearing in the order demand data is obtained. If there are n types of bearings in total, then i = 1, 2, 3, ..., n.
4. A lean production management method according to claim 3, characterized in that: The specific methods for determining the bearing products to be produced as qualified products, products to be repaired, and products that cannot be repaired are as follows: If there are n types of bearings in total, and the i-th type of bearing has a total of P bearing products, the bearings that meet the basic quality standards are marked as candidate products, and the bearings that do not meet the basic quality standards are marked as scrapped products. The candidate products corresponding to each bearing are obtained, and the bearing products are further classified into qualified products, products to be repaired, and non-repairable products among the candidate products based on the quality standard bias corresponding to each bearing; The quality standard deviation corresponding to the i-th bearing is denoted as U i , U i = {u i1 ,u i2 ,u i3 ,...,u im }, where u ij is the jth quality standard bias element in the quality standard bias corresponding to the i-th bearing, that is, the jth quality standard item in the special quality standard corresponding to the quality standard bias. If the quality standard bias includes the accuracy standard and the high temperature resistance standard, the accuracy standard and the high temperature resistance standard are both quality standard items, j = 1, 2, 3, ..., m; each quality standard bias element corresponds to a quality standard formula; if there is a bearing product in the candidate product set that meets all the corresponding quality standard bias elements, the bearing is marked as a qualified product; if there is a bearing product in the candidate product set that does not meet all the corresponding quality standard bias elements, a threshold judgment is performed; The specific method of determining the threshold is as follows: Set a repairable threshold for all quality standard bias elements in each quality standard bias, subtract the actual value corresponding to the produced bearing product from the formula value of each quality standard bias element, and make an absolute value. If there is a difference whose absolute value exceeds the repairable threshold, the bearing product is marked as an unrepairable product; if all the absolute values of the difference do not exceed the repairable threshold, the bearing product is marked as a product to be repaired; Classify P bearing products until all the bearing products produced for the i-th bearing are classified into qualified products, products to be repaired, and unrepairable products; classify n bearings in turn until all the bearing products produced for the n bearings are classified into qualified products, products to be repaired, and unrepairable products.
5. A lean production management method according to claim 4, characterized in that: The basic quality standards all correspond to quality standard formulas, and the quality standard formulas are either inequalities or equations, that is, numerical inequalities and numerical equations of quality standard items. If the quality standard formula is an inequality, the formula value is the value of the inequality; if the quality standard formula is an equation, the formula value is the value of the equation; the quality standard bias is the quality standard that the bearing needs to pay special attention to in addition to the basic quality standard, and the basic quality standard is the quality standard that all bearings need to meet.
6. A lean production management method according to claim 5, characterized in that: The calculation formula for the first pass rate and production pass rate of the bearing production is expressed as: Among them, HG i _one is the first pass rate of the i-th bearing, SL i _hg is the number of qualified products of the i-th type of bearing, SL i _z is the total number of products of the i-th type of bearings, i.e., the sum of the number of qualified products, products to be repaired, unrepairable products, and scrapped products; Among them, HG i _sc is the production qualification rate of the i-th bearing, SL i _dx is the number of bearings of the i-th type to be repaired, i = 1, 2, 3, ..., n, and n is the total number of bearing types.
7. A lean production management method according to claim 1, characterized in that: The production plan set is a collection of production plans, and the production plan is obtained in the following manner: Obtain order demand data and generate multiple different production plans based on the order product quantity, current time, and delivery time. Each bearing corresponds to a production plan set, so there are n production plan sets corresponding to n types of bearings. The use of an optimization algorithm to obtain a real-time optimal production plan includes the following steps: Step S11: automatically generate n production plan sets; Step S12: Encode each production plan in the i-th production plan set, obtain chromosomes, and construct an initial population; Step S13: determining the fitness function; Step S14: performing natural selection on chromosomes in the population; Step S15: performing crossover recombination on chromosomes in the population; Step S16: mutating the chromosomes in the population; Step S17: Obtain a new population, preset the population generation number to be L, the fitness threshold to be Q, L is an integer greater than 0, and Q is a real number greater than 0; loop through steps S14 to S16 until the generation number corresponding to the new population is L or the fitness corresponding to a chromosome in the new population is greater than or equal to the fitness threshold Q, the loop ends, and the production plan corresponding to the chromosome with the maximum fitness in the new population is taken as the real-time optimal production plan of the i-th production plan set; Step S18: loop steps S12 to S17 until all n production plan sets obtain corresponding real-time optimal production plans.
8. A lean production management method according to claim 7, characterized in that: In step S11, each production plan is encoded as U, where U is a chromosome, and R chromosomes are randomly generated to form an initial population C. i , C i = {U i1 ,U i2 ,U i3 ,...,U iR }; Among them, Ui R is the Rth chromosome in the production plan of the i-th bearing; The expression of the fitness function is: Among them, SY ir is the fitness of the production plan corresponding to the rth chromosome in the i-th production plan set, HG ir _one y HG is the predicted value of the first pass rate of the production plan corresponding to the rth chromosome in the i-th production plan set, ir _sc y is the predicted value of the production qualification rate of the production plan corresponding to the rth chromosome in the i-th production plan set; The method for obtaining the first pass rate prediction value and the production pass rate prediction value of the production plan corresponding to the rth chromosome in the i-th production plan set is: Input the production plan, resource data and quality data corresponding to the rth chromosome into the first pass rate prediction model of the ith production plan set, output the first pass rate prediction value of the daily production in the production plan corresponding to the rth chromosome in the ith production plan set, and obtain the final first pass rate prediction value after weighted average; The production plan corresponding to the rth chromosome is input into the prediction model of the proportion of products to be repaired in the ith production plan set, and the proportion of products to be repaired produced daily in the production plan corresponding to the rth chromosome in the ith production plan set is output. The proportion of products to be repaired produced daily and the predicted value of the first qualified rate of daily production are added to obtain the daily production qualified rate prediction value, and the final production qualified rate prediction value is obtained after weighted average.
9. A lean production management method according to claim 8, characterized in that: The first pass rate prediction model and the product proportion prediction model for repair are both deep neural network models, and the model training processes of the two are the same; The specific training process of the first pass rate prediction model is as follows: Pre-collecting d groups of analysis data of the i-th type of bearing, the analysis data being resource data, quality data and bearing production quantity, collecting d groups of analysis data, d being an integer greater than 1, converting the analysis data and the corresponding first-pass rate into a corresponding set of feature vectors, the first-pass rate being obtained by dividing the number of qualified products in the bearing production quantity by the number of bearings produced continuously, in the same way as calculated in step S02; Each set of feature vectors is used as the input of the first-pass rate prediction model. The first-pass rate prediction model uses a set of first-pass rate prediction values corresponding to each set of analysis data as output, and uses the actual first-pass rate corresponding to each set of analysis data as the prediction target. The actual first-pass rate is the first-pass rate corresponding to the analysis data collected in advance; the training target is to minimize the sum of the prediction errors of all analysis data; the formula of the prediction error is expressed as: ε q =θ q -μ q , where ε q is the prediction error, q is the group number of the eigenvector corresponding to the analyzed data, θ q is the predicted value of the first pass rate corresponding to the qth group of analysis data, μ q is the actual first-time qualified rate corresponding to the qth group of analysis data, and the first-time qualified rate prediction model is trained until the sum of the prediction errors reaches convergence, and the training is stopped; q = 1, 2, 3, ..., d; The prediction model for the proportion of products to be repaired is trained in the same way as the one-time pass rate prediction model. The proportion of products to be repaired is obtained by dividing the number of products to be repaired in the number of bearings produced by the number of bearings produced.
10. A lean production management system, used to implement a lean production management method according to any one of claims 1 to 9, characterized in that: It includes data acquisition module, data preprocessing module, data analysis module, data prediction module, lean production analysis module and human-computer interaction module; The data acquisition module is used to collect bearing production data and quality data, and transmit them to the data preprocessing module for preprocessing; The data preprocessing module is used to receive data from the data acquisition module, preprocess the data and then transmit it to the data analysis module and the data prediction module; The data analysis module obtains the quality standard deviations corresponding to various bearings in the order demand data, classifies the produced bearing products into qualified products, products to be repaired, and unrepairable products, and calculates the first-time qualified rate and production qualified rate of the bearings and transmits them to the data prediction module; The data prediction module is used to construct a production plan set, predict the first-time qualified rate and the production qualified rate in combination with the constructed prediction model, and use the optimization algorithm to obtain the real-time optimal production plan and transmit it to the lean production analysis module; The lean production analysis module draws a VSM diagram based on the raw material data, order demand data, real-time optimal production plan and resource data of the bearing products for analysis, identifies various non-value-added activities in the production process and makes rectifications; The human-computer interaction module is used to perform human-computer interaction display on the data.
Citation Information
Cited By
Bearing production control system for data analysis
CN120891810A