Multi-station quality problem tracing method and system in detection device assembling process

Abstract

The invention provides a multi-station quality problem tracing method and system in the assembly process of a detection device, and aims to solve the problem that the key control point problem in the assembly process is difficult to trace due to numerous links and complex problem tracing range and boundary in the assembly process of parts of a certain detection device. According to the method, a quality problem analysis model is established by collecting multi-station quality data in the assembly process and based on an offset absolute value, a mean square error and a quality loss method, and the key control point problem tracing method for the multi-station quality in the assembly process of the detection device is designed and used for accurately positioning and analyzing reasons and influences of the quality problems. And the assembly process quality and the assembly efficiency of a certain detection device are improved.

Description

Technical Field

[0001] The present application relates to the technical field of quality control, and particularly to a multi-station quality problem tracing method and system for the assembly process of a detection device. Background Art

[0002] A certain detection device is an angle tracking system. As a key part, its performance directly determines the final accuracy of the device. This detection device has the characteristics of complex structure, high precision, and numerous components, including hundreds of ultra-precision parts, with system integration and complexity characteristics. It is necessary to ensure not only the geometric assembly performance of the detection device, such as assembly accuracy, but also its physical assembly performance, such as pneumatic performance. The complexity of the assembly problem makes quality control and quality problem tracing difficult. On the one hand, although technologies such as data collection, big data analysis, and the Internet of Things can be used to achieve real-time monitoring and tracing of quality control points, quality problems in the assembly process of the detection device may involve multiple links and multiple participants. Different from the general processing process, there are numerous quality control points in the assembly of the detection device, and multiple control points at a single station form a control surface. On the other hand, the scope of influence of quality problems may expand over time, resulting in a complex and difficult task of tracing the scope and boundary of the problem, which requires comprehensive consideration of various factors and possibilities. Limited by time and cost, sufficient resources may not be invested to deeply trace the quality problems in the assembly of the detection device, thereby affecting the quality of the assembly process of the detection device. The existing domestic theoretical research on assembly technology mainly focuses on tolerance analysis or optimization of assembly processes, but lacks research on tracing quality problems in the assembly process.

[0003] Therefore, the present invention proposes a multi-station quality problem tracing method for the assembly process of a detection device, which can be used to trace key control point problems at multiple stations in the assembly process of a certain detection device. Summary of the Invention

[0004] To solve the problems that there are numerous links in the assembly process of parts of a certain detection device, and the scope and boundary of problem tracing are complex, resulting in difficulties in tracing key control point problems in the assembly process. The present invention proposes to collect multi-station quality data in the assembly process, establish a quality problem analysis model based on the absolute value of deviation, mean square error, and quality loss method, and design a key control point problem tracing method for multi-station quality in the assembly process of a detection device, which is used to accurately locate and analyze the causes and impacts of quality problems, and improve the quality and assembly efficiency of the assembly process of a certain detection device. The specific technical solutions are as follows:

[0005] In the first aspect, the present invention provides a multi-station quality problem tracing method for the assembly process of a detection device, and the method includes:

[0006] Obtain the quality characteristic indicators of each control point through measurement to obtain the control point quality data; the control points are data acquisition points set on the quality control surfaces of each product in the same batch of products; the quality characteristic indicators are data characteristics of the control points under different quality standards;

[0007] Based on the gap between the control point quality data obtained from actual measurement and its corresponding target value, determine the offset data, MSE data, and quality loss data corresponding to each control point;

[0008] Based on the offset data, determine the aspects where quality problems occur; based on the MES data, determine the assembly links where quality problems occur; based on the quality loss data, determine the losses caused by quality problems; the aspects where quality problems occur include products, processes, and personnel.

[0009] Further, the obtaining the quality characteristic indicators of each control point through measurement to obtain the control point quality data includes:

[0010] Probe the control surfaces of the products on each workstation through a preset plurality of sensors to obtain the detection signals of the sensors;

[0011] Based on the detection signals of the sensors, determine the quality characteristic indicators of each control point;

[0012] Collect the quality characteristic indicators corresponding to each control point to form the control point quality data.

[0013] Further, the determining the offset data, MSE data, and quality loss data corresponding to each control point based on the gap between the control point quality data obtained from actual measurement and its corresponding target value includes:

[0014] Determine the offset data based on the gap between the expected value and the target value of the control point quality data;

[0015] Determine the MSE data based on the average value of the gap between the control point quality data and the target value;

[0016] Determine the quality loss data based on the losses caused by quality problems in the control point quality data.

[0017] Further, the determining the offset data based on the gap between the expected value and the target value of the control point quality data includes:

[0018] The expression of bias is Then the absolute value of the bias is expressed as:

[0019]

[0020] Wherein, Cumulative proportion A of absolute value of bias j Expressed as:

[0021]

[0022] where the quality characteristic index is x ijl , i = 1, 2, ..., m; j = 1, 2, ..., n; l = 1, 2, ..., L; representing the i-th quality control surface, the j-th control point, and the l-th batch of products. There are L products in one batch, and the quality characteristic target value of each quality control surface is T i .

[0023] Furthermore, determining the MSE data based on the average of the gaps between the quality data of the control points and the target values includes:

[0024]

[0025] Cumulative proportion B of MSE j Expressed as:

[0026]

[0027] where Var() is the variance operation, and x ij is the quality characteristic index of the i-th quality control surface and the j-th control point of a single product.

[0028] Furthermore, determining the quality loss data based on the losses caused by quality problems in the quality data of the control points includes:

[0029] The expression adopted by the quality loss function is:

[0030]

[0031] where L i (x) is the quality loss, and K is a constant.

[0032] The cumulative proportion of quality loss is expressed as:

[0033]

[0034] where m indicates the number of quality control surfaces, and j indicates the number of quality control points.

[0035] In a second aspect, the present invention also provides a multi-station quality problem traceability method for the assembly process of a detection device, and the method includes:

[0036] Obtain the quality characteristic indicators of each control point through measurement to obtain the control point quality data; the control point is a data acquisition point set on the quality control surface of each product in the same batch of products; the quality characteristic indicator is the data characteristic of the control point under different quality standards.

[0037] Based on the gap between the control point quality data obtained from actual measurement and its corresponding target value, determine the offset data, MSE data, and quality loss data corresponding to each control point.

[0038] Based on the offset data, determine the aspects where quality problems occur; based on the MES data, determine the assembly links where quality problems occur; based on the quality loss data, determine the losses caused by quality problems; the aspects where quality problems occur include products, processes, and personnel.

[0039] In another embodiment of the second aspect, the obtaining the quality characteristic indicators of each control point through measurement to obtain the control point quality data includes:

[0040] Use a preset number of sensors to detect the control surface of the product on each workstation to obtain the detection signals of the sensors.

[0041] Based on the detection signals of the sensors, determine the quality characteristic indicators of each control point.

[0042] Collect the quality characteristic indicators corresponding to each control point to form the control point quality data.

[0043] In another embodiment of the second aspect, the determining the offset data, MSE data, and quality loss data corresponding to each control point based on the gap between the control point quality data obtained from actual measurement and its corresponding target value includes:

[0044] Determine the offset data based on the gap between the expected value and the target value of the control point quality data.

[0045] Determine the MSE data based on the average value of the gap between the control point quality data and the target value.

[0046] Determine the quality loss data based on the losses caused by quality problems in the control point quality data.

[0047] The beneficial effects of the present invention are as follows:

[0048] The present invention provides a multi-station quality problem traceability method and system for the assembly process of a detection device. The present invention establishes a quality model for multiple stations in the assembly process of a certain detection device through the absolute value of bias and the cumulative percentage of the absolute value of bias, MSE and the cumulative percentage of MSE, quality loss and the cumulative percentage of quality loss. Based on this quality model, combined with the Pareto criterion, the quality problems that need to be improved first in the assembly process of a certain detection device are determined. Taking the control surface as a unit, the quality key control surfaces are first screened out to quickly locate quality problems, and a multi-station quality problem traceability method for the assembly process of a detection device is proposed, which simplifies the traceability process of simultaneously controlling the quality problems of multi-station control points and effectively solves quality problems. Brief Description of the Drawings

[0049] Figure 1 is a flowchart of the method of the present invention;

[0050] Figure 2 is a schematic diagram of multi-station quality control surfaces in the assembly process of a certain detection device;

[0051] Figure 3 is a schematic diagram of the distribution of middle control points of a control surface at the lens assembly station;

[0052] Figure 4 is a statistical chart of bias and cumulative percentage of bias of different control surfaces;

[0053] Figure 5 is the MSE and cumulative percentage of MSE of different control surfaces. Detailed Description of the Embodiments

[0054] To make the objectives, technical solutions, and advantages of this specification clearer, the technical solutions of this application will be clearly and completely described below in conjunction with the specific embodiments of this application and their corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, rather than all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope protected by this document.

[0055] The following will be combined with Figures 1-5 to detail the technical solutions provided by each embodiment of this specification. Specific Embodiment 1:

[0057] In order to solve the problem that there are many links in the assembly process of parts of a detection device, and the scope and boundary of problem traceability are complex, resulting in difficulties in tracing key control points in the assembly process. The present invention proposes to collect multi-station quality data in the assembly process, and based on the methods of absolute deviation, mean square error and quality loss, establish a quality problem analysis model, and design a method for tracing key control points of multi-station quality in the assembly process of a detection device, which is used to accurately locate and analyze the causes and impacts of quality problems, and improve the quality and assembly efficiency of the assembly process of a detection device.

[0058] The present invention proposes a method for tracing key control points of multi-station quality in the assembly process of a detection device, and the specific implementation steps are as follows:

[0059] Step 1: Define the multi-station quality measurement indicators for the assembly process of the detection device and collect quality data of control points

[0060] In order to trace the multi-station quality problems in the assembly process of the detection device, first define the quality measurement indicators, and then collect the quality data of the control points.

[0061] As Figure 2 shown, the assembly process of a detection device is multi-station, and the quality characteristic index concerned by each control point is x ijl , i = 1, 2,..., m; j = 1, 2,..., n; l = 1, 2,..., L;, representing the i-th quality control surface, the j-th control point, and there are L products in a batch. The quality characteristic target value of each quality control surface is T i .

[0062] Step 2: Establish a quality problem analysis model based on data of different control points

[0063] Based on the quality data of the control points collected, analyze the data, use the absolute value of bias to analyze the direction of the quality problem and the degree of deviation from the true value, clarify the size of the quality fluctuation according to the fluctuation MSE index, and obtain the severity of the quality problem according to the quality loss. According to the quality data x ijl and the target value of T i , the quality problem analysis models established by bias, MSE and quality loss are respectively shown as follows.

[0064] (1) Absolute value of bias and cumulative proportion of absolute value of bias

[0065] Bias can reflect the gap between the expected value of the measurement result and the target value. A larger deviation indicates a greater degree of deviation of the assembly process from the target value. The expression of bias is Then the absolute value of bias is expressed as:

[0066]

[0067] Among them, Cumulative proportion A of absolute bias j Expressed as:

[0068]

[0069] (2) MSE and cumulative proportion of MSE

[0070] It is the average value of the gap between the measurement result and the target value. A higher MSE indicates a higher degree of dispersion between the production process and the target value. The expression of MSE is:

[0071]

[0072] Cumulative proportion B of MSE j Expressed as:

[0073]

[0074] (3) Quality loss and cumulative proportion of quality loss:

[0075] Quality loss refers to the losses caused by product quality problems during production, operation processes and activities, that is, internal and external losses due to poor quality. The expression used for the quality loss function is

[0076]

[0077] Among them, L i (x) is the quality loss and K is a constant.

[0078] The cumulative proportion of quality loss is expressed as:

[0079]

[0080] Therefore, according to the bias, MSE and quality loss models, the offset, MSE and quality loss information of each control point are obtained.

[0081] Step 3: Trace the problems of key control points based on the quality model

[0082] According to Step 2, the bias, MSE and quality loss information of each control point can be obtained. Since there are many multi-station control points in the assembly process of a certain detection device, if quality control and problem analysis are carried out for each control point, it will not only waste manpower and material resources but also be inefficient.

[0083] Therefore, to effectively screen key control points, taking the control surface as a unit, first, it is necessary to analyze the accuracy of each control surface in the assembly process according to the absolute value of the bias, and analyze whether the quality problem is caused by the product, process or personnel.

[0084] Secondly, according to the MSE analysis, the stability and consistency of each control surface in each assembly link are analyzed. When the MSE is large, it indicates that there are large fluctuations in this control surface, which may lead to unstable quality in the assembly process. Then, according to the quality loss analysis, the losses caused by quality problems of each control surface in the assembly process are analyzed, so as to improve and optimize the assembly process targeted.

[0085] Finally, the Pareto principle, also known as the 80 / 20 rule, is a classic efficiency principle that can help find the root cause of problems more accurately, optimize resource allocation and improve efficiency. Specifically, it means that in many cases, about 20% of the causes will lead to 80% of the results. According to the absolute value of the bias of each control surface and the cumulative percentage of the absolute value of the bias, MSE and the cumulative percentage of MSE, combined with the Pareto principle, key control points are determined, which are specifically divided into the following situations:

[0086] (1) Situation 1, when the biases (or MSEs) are arranged from large to small, if a bias (or MSE) ranks the i-th and the difference from the (i + 1)-th bias (or MSE) is significantly large, the larger points before the i-th bias (or MSE) can be selected as key control points.

[0087] (2) Situation 2, according to the cumulative proportion of the bias (or MSE), the first a% of the points are selected as key control points.

[0088] In summary, according to the absolute value of the bias of each control surface and the cumulative percentage of the absolute value of the bias, MSE and the cumulative percentage of MSE, combined with the Pareto principle, the quality problems that need to be improved first in the multi-station assembly process are determined, so as to screen out the key quality control points.

[0089] (3) Advantages

[0090] This invention corresponds to the technical problems to be solved and the technical solutions, and specifically describes the solution ideas and the effects produced by this invention; the specific advantages of this invention are as follows:

[0091] This invention establishes a multi-station quality model for the assembly process of a certain detection device through the absolute value of the bias and the cumulative percentage of the absolute value of the bias, MSE and the cumulative percentage of MSE, quality loss and the cumulative percentage of quality loss. Based on this quality model, combined with the Pareto principle, the quality problems that need to be improved first in the assembly process of a certain detection device are determined. Taking the control surface as the unit, the key quality control surfaces are screened out first to quickly locate the quality problems, and a multi-station quality problem tracing method for the assembly process of a detection device is proposed, which simplifies the process of tracing the quality problems of multi-station control points simultaneously and effectively solves the quality problems. Specific Embodiment 2:

[0093] The present invention proposes a multi-station quality problem traceability method for the assembly process of a detection device, and its flowchart is as Figure 1 shown, and it includes the following steps:

[0094] Step 1: Define the multi-station quality measurement indicators for the assembly process of the detection device and collect the quality data of the control points

[0095] The assembly process of a certain detection device is a multi-station processing process. The quality characteristics of the components concerned at each station are different. Therefore, the number of control surfaces set at each station is different. There are a total of 50 control surfaces distributed in the part assembly process of a certain detection device, and each control surface has n = 7 control points. There are a total of L = 156 parts in one batch, and a total of 50×7×156 data are collected. Table 1 lists the quality data of 7 control points of 20 control surfaces in one batch of parts in the assembly process of a certain detection device.

[0096] Table 1 Quality data of 20 control surfaces at multiple stations in the assembly process of a certain detection device

[0097]

[0098]

[0099] Step 2: Establish a quality problem analysis model based on the data distribution characteristics of different points

[0100] According to the absolute value of deviation, MSE, and quality loss model, calculate the absolute value of bias, MSE, and quality loss for each control surface, and sort them from small to large, and record and summarize them as shown in Tables 2-4.

[0101] Table 2 Sorting of absolute values of bias

[0102]

[0103] Table 3 Sorting of MSE

[0104]

[0105] Table 4 Sorting of quality loss

[0106] No. Loss No. Loss No. Loss No. Loss No. Loss 28 0.000629 6 0.000791 38 0.001024 44 0.00128 14 0.001606 48 0.000659 7 0.000792 5 0.001097 10 0.001286 20 0.001644 8 0.000691 47 0.000805 13 0.001109 4 0.001292 12 0.001656 37 0.000702 29 0.000837 46 0.001141 31 0.001385 40 0.002003 15 0.000716 36 0.000868 18 0.001182 19 0.001463 2 0.033719 35 0.000718 32 0.000893 30 0.001184 41 0.001474 3 0.035786 34 0.000761 17 0.000907 25 0.001205 33 0.001494 1 0.035831 45 0.000764 49 0.000933 9 0.001247 43 0.001542 23 0.036169 26 0.000773 16 0.000935 39 0.001268 42 0.001545 22 0.037466 27 0.000781 50 0.001009 11 0.001279 21 0.001602 24 0.039457

[0107] Step 3: Carry out problem traceability of key control points based on the quality model

[0108] According to the quality model established in Step 2 and the obtained absolute value of bias, MSE, and quality loss, calculate the cumulative proportion of bias, MSE, and quality loss of each control surface, and draw an image as Figure 4 shown. Among them, Figure 4 shows the bias and cumulative proportion of bias of different control surfaces.

[0109] (1) Figure 4 shows the images of the biases of different control surfaces arranged from large to small. From Table 2 and Figure 4 it can be seen that the obvious biases of the 24th, 22nd, 23rd, 1st, 3rd, and 2nd control surfaces are relatively large, and the bias gap between the 2nd control surface and the subsequent 40th control surface is relatively large; from the cumulative proportion of biases, it can be seen that the cumulative proportion of the 24th control surface is 13.17%, the cumulative proportion of the 22nd control surface is 25.99%, the cumulative proportion of the 23rd control surface is 38.54%, the cumulative proportion of the 1st control surface is 51.07%, the cumulative proportion of the 3rd control surface is 63.60%, the cumulative proportion of the 2nd control surface is 75.70%, the cumulative proportion of the 40th control surface is 77.57%, and the cumulative proportion of the 12th control surface is 79.05%. Therefore, it is recommended to correct the biases of the 24th, 22nd, 23rd, 1st, 3rd, and 2nd control surfaces first. Correcting the biases of the 24th, 22nd, 23rd, 1st, 3rd, 2nd, 40th, and 12th control surfaces, the correction effect can reach 79.05%.

[0110] (2) Figure 5 shows the images of the MSE of different control surfaces arranged from large to small. Those with large fluctuations are: the 24th, 22nd, 23rd, 1st, 3rd, and 2nd control surfaces have relatively large fluctuations, and the fluctuation gap between the 2nd control surface and the subsequent 40th control surface is relatively large. From the cumulative proportion of fluctuations, it can be seen that the cumulative proportion of fluctuations of the 24th control surface is 14.73%, the cumulative proportion of fluctuations of the 22nd control surface is 28.73%, the cumulative proportion of fluctuations of the 23rd control surface is 42.24%, the cumulative proportion of fluctuations of the 1st control surface is 55.62%, the cumulative proportion of fluctuations of the 3rd control surface is 81.59%, and the cumulative proportion of fluctuations of the 2nd control surface is 82.34%.

[0111] In summary, through comprehensive analysis of biases, fluctuations, and quality loss diagrams, it is recommended to correct the biases of the 24th, 22nd, 23rd, 1st, 3rd, 2nd, 40th, and 12th first to solve the bias quantity problem up to 79.0%. Prioritize controlling and reducing the fluctuations of the 24th, 22nd, 23rd, 1st, 3rd, and 2nd, which can solve the fluctuation problem up to 82.34%. Reducing the quality losses of the 24th, 22nd, 23rd, 1st, 3rd, and 2nd can solve the loss problem up to 81.63%.

[0112] Example embodiments have been disclosed herein, and although specific terms are employed, they are used only and should be interpreted only as general illustrative meanings and not for the purpose of limitation. In some embodiments, it will be apparent to those skilled in the art that, unless otherwise expressly stated, features, characteristics, and / or elements described in connection with a particular embodiment may be used singly or in combination with features, characteristics, and / or elements described in connection with other embodiments. Accordingly, those skilled in the art will understand that various forms and details may be changed without departing from the scope of the invention as set forth by the appended claims.

Claims

1. A method for tracing multi-station quality problems in the assembly process of a detection device, characterized in that: The method comprises: The quality characteristic index of each control point is obtained by measuring and obtaining the quality data of the control point; the control point is a data collection point set on the quality control surface of each product in the same batch of products; the quality characteristic index is the data feature of the control point under different quality standards; Determine the offset data, MSE data, and quality loss data corresponding to each control point based on the gap between the control point quality data obtained by actual measurement and its corresponding target value; Based on the offset data, the aspects where the quality problem occurs are determined; based on the MES data, the assembly links where the quality problem occurs are determined; based on the quality loss data, the losses caused by the quality problem are determined; the aspects where the quality problem occurs include products, processes, and personnel.

2. A method for tracing multi-station quality problems in the assembly process of a detection device as claimed in claim 1, characterized in that: The quality characteristic index of each control point is obtained by measuring to obtain the control point quality data, including: Detect the control surface of the product at each workstation through multiple preset sensors to obtain the detection signal of the sensor; Determining a quality characteristic index of each control point based on a detection signal of the sensor; The quality characteristic indicators corresponding to each control point are collected to form the control point quality data.

3. A method for tracing multi-station quality problems in the assembly process of a detection device as claimed in claim 1, characterized in that: The step of determining the offset data, MSE data, and quality loss data corresponding to each control point based on the gap between the control point quality data obtained by actual measurement and the corresponding target value includes: determining the offset data based on a difference between an expected value and a target value of the control point quality data; Determine the MSE data based on an average value of the difference between the control point quality data and the target value; The quality loss data is determined based on a loss caused by a quality problem in the control point quality data.

4. A method for tracing multi-station quality problems in the assembly process of a detection device as claimed in claim 3, characterized in that: The determining the offset data based on the gap between the expected value and the target value of the control point quality data comprises: The expression of bias is The absolute value of the bias is expressed as: in, Cumulative percentage of absolute value of bias A j It is expressed as: Among them, the quality characteristic index is x ijl ,i=1,2,...,m;j=1,2,...,n;l=1,2,...,L;represents the i-th quality control surface, the j-th control point, the l-th batch of products, a batch has L products, and the target value of each quality control surface quality characteristic is T i .

5. A method for tracing multi-station quality problems in the assembly process of a detection device as claimed in claim 3, characterized in that: The determining the MSE data based on the average value of the gap between the control point quality data and the target value comprises: MSE cumulative proportion B j It is expressed as: Among them, Var() is the variance operation, x ij is the quality characteristic index of the i-th quality control surface and the j-th control point of a single product.

6. A method for tracing multi-station quality problems in the assembly process of a detection device as claimed in claim 3, characterized in that: The determining of the quality loss data based on the loss caused by the quality problem in the control point quality data comprises: The expression used for the mass loss function is: Among them, L i (x) is the mass loss and K is a constant. The cumulative percentage of quality loss is expressed as: Among them, m indicates the number of quality control surfaces, and j indicates the number of quality control points.

7. A method for tracing multi-station quality problems in the assembly process of a detection device, characterized in that: The method comprises: The quality characteristic index of each control point is obtained by measuring and obtaining the quality data of the control point; the control point is a data collection point set on the quality control surface of each product in the same batch of products; the quality characteristic index is the data feature of the control point under different quality standards; Determine the offset data, MSE data, and quality loss data corresponding to each control point based on the gap between the control point quality data obtained by actual measurement and its corresponding target value; Based on the offset data, the aspects where the quality problem occurs are determined; based on the MES data, the assembly links where the quality problem occurs are determined; based on the quality loss data, the losses caused by the quality problem are determined; the aspects where the quality problem occurs include products, processes, and personnel.

8. A method for tracing multi-station quality problems in the assembly process of a detection device as claimed in claim 1, characterized in that: The quality characteristic index of each control point is obtained by measuring to obtain the control point quality data, including: Detect the control surface of the product at each workstation through multiple preset sensors to obtain the detection signal of the sensor; Determining a quality characteristic index of each control point based on a detection signal of the sensor; The quality characteristic indicators corresponding to each control point are collected to form the control point quality data.

9. A method for tracing multi-station quality problems in the assembly process of a detection device as claimed in claim 1, characterized in that: The step of determining the offset data, MSE data, and quality loss data corresponding to each control point based on the gap between the control point quality data obtained by actual measurement and the corresponding target value includes: determining the offset data based on a difference between an expected value and a target value of the control point quality data; Determine the MSE data based on an average value of the difference between the control point quality data and the target value; The quality loss data is determined based on a loss caused by a quality problem in the control point quality data.

10. A production line equipment, characterized in that: The device performs quality problem tracing by the method described in any one of claims 1 to 6.

Images