Apparatus and method for determining optimal cleaning cycle of screen printing machine, and recording medium recording instructions

Abstract

A technique for determining an optimal cleaning cycle of a screen printer is disclosed. An apparatus of an aspect of the present disclosure may include: one or more processors; and one or more memories, the one or more memories storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations; wherein the one or more processors may determine a plurality of defect scores indicative of a printing quality of each of a plurality of circuit boards based on at least one indicator indicative of a state of a solder paste printed on the plurality of circuit boards, generate prediction data based on the plurality of defect scores, and transmit the prediction data to the plurality of circuit boards, the prediction data is used for indicating the reduction degree of the printing quality of the Nth circuit board when N circuit boards are sequentially printed without cleaning the screen printing machine, and determining a cleaning period based on the prediction data and the resource amount consumed when cleaning of the screen printing machine is executed, the cleaning period can correspond to the number of circuit boards printed by the screen printing machine after the screen printing machine is cleaned and before the screen printing machine is cleaned next time.

Description

Technical Field

[0001] The present disclosure relates to techniques for determining an optimal cleaning cycle for a screen printing machine. Background Art

[0002] The process of printing solder paste on a printed circuit board (PCB) can be performed by a screen printer. In the process of printing solder paste, there is a problem that the template is not cleaned in an efficient cycle because the amount of residual solder paste accumulated on the template is not taken into account. Summary of the invention

[0003] Technical issues

[0004] The present disclosure provides a technique capable of determining an optimal cleaning cycle.

[0005] Technical Solution

[0006] An apparatus according to one aspect of the present disclosure may include: one or more processors; and one or more memories, wherein the one or more memories store instructions, which, when executed with the aid of the one or more processors, enable the one or more processors to perform operations; wherein the one or more processors may determine a plurality of anomaly scores indicating the printing quality of each of the plurality of circuit boards based on at least one indicator indicating the state of solder paste printed on the plurality of circuit boards, generate prediction data based on the plurality of anomaly scores, the prediction data being used to indicate the degree of degradation in the printing quality of the Nth circuit board when the screen printer prints N circuit boards in sequence without cleaning the screen printer, and determine a cleaning cycle based on the prediction data and the amount of resources consumed when performing cleaning of the screen printer, the cleaning cycle may correspond to the number of circuit boards printed by the screen printer after cleaning the screen printer and before the next cleaning.

[0007] In one embodiment, the one or more processors can determine a corresponding weighted value for each of the at least one indicator, determine a predicted range for the at least one indicator, and determine the multiple defective scores for each of the multiple circuit boards based on the degree to which each of the at least one indicator exceeds the predicted range and the weighted value.

[0008] In one embodiment, the one or more processors may determine a bad score for each pad group according to a plurality of pad groups differentiated according to pad shapes, and determine the plurality of bad scores for each of the plurality of circuit boards based on the bad scores for the pad groups.

[0009] In one embodiment, the one or more processors may determine a weighted value of each pad group based on the number of pads included in each of the multiple pad groups, and determine the multiple bad scores for each of the multiple circuit boards based on the bad scores of the pad groups and the weighted values ​​of the pad groups.

[0010] In one embodiment, the one or more processors may determine the weight value of a pad group including a pad number less than a predetermined number as 0 among the plurality of pad groups.

[0011] In one embodiment, the at least one indicator may include at least one selected from the group consisting of an over-soldering degree of the printed solder paste, an under-soldering degree of the printed solder paste, a standard deviation of the amount of printed solder paste, and a degree of deviation from the alignment position of the circuit board and the template.

[0012] In one embodiment, it may also include a communication circuit connected to the screen printing machine in a communicative manner, and the communication circuit can transmit the cleaning cycle to the screen printing machine. The screen printing machine can perform predetermined actions in response to the reception of the cleaning cycle so as to perform cleaning according to the cleaning cycle.

[0013] In one embodiment, the one or more processors may generate the prediction data using survival analysis.

[0014] In one embodiment, the resources consumed when performing the cleaning may include the time required for cleaning the screen printing machine and the cost of purchasing items.

[0015] In one embodiment, the one or more processors can determine the cleaning cycle that can minimize the amount of resources consumed and the reduction in printing quality when cleaning the screen printing machine based on the prediction data, the time required for cleaning the screen printing machine, and the cost of purchasing items.

[0016] In one embodiment, the one or more processors may continuously obtain the at least one indicator and store it in the memory, and after performing a predetermined number of cleanings, update the cleaning cycle based on the data stored in the memory.

[0017] A method according to another aspect of the present disclosure may include: a step of determining a plurality of anomaly scores indicating the printing quality of each of the plurality of circuit boards based on at least one indicator indicating the state of solder paste printed on the plurality of circuit boards; a step of generating prediction data based on the plurality of anomaly scores, the prediction data being used to indicate the degree of quality degradation of the Nth circuit board when the screen printer prints N circuit boards in sequence without cleaning the screen printer; and a step of determining a cleaning cycle based on the prediction data and the amount of resources consumed when performing cleaning of the screen printer; wherein the cleaning cycle may correspond to the number of circuit boards printed by the screen printer after cleaning the screen printer and before the next cleaning.

[0018] In one embodiment, the step of determining the defective score may include: a step of determining a weighted value corresponding to each of the at least one indicator; a step of determining a predicted range of the at least one indicator; and a step of determining the multiple defective scores for each of the multiple circuit boards based on the degree to which each of the at least one indicator exceeds the predicted range and the weighted value.

[0019] In one embodiment, the step of determining the bad score may include: a step of determining the bad score of each pad group according to a plurality of pad groups distinguished according to pad shapes; and a step of determining the plurality of bad scores for each of the plurality of circuit boards based on the bad scores of the respective pad groups.

[0020] In one embodiment, the step of determining the bad score may include: the step of determining the weighted value of each pad group based on the number of pads included in each group of the multiple pad groups, and the step of determining the multiple bad scores for each of the multiple circuit boards based on the bad scores of each pad group and the weighted values ​​of each pad group.

[0021] In one embodiment, the step of determining weighted values ​​according to the plurality of pad groups may further include: determining the weighted value of a pad group including a pad number less than a predetermined number as 0 among the plurality of pad groups.

[0022] In one embodiment, the at least one indicator may include at least one selected from the group consisting of an over-soldering degree of the printed solder paste, an under-soldering degree of the printed solder paste, a standard deviation of the amount of printed solder paste, and a degree of deviation from the alignment position of the circuit board and the template.

[0023] In one embodiment, the step of generating the prediction data may include the step of generating the prediction data using survival analysis.

[0024] In one embodiment, the step of determining the length of the cleaning cycle may include: continuously obtaining the at least one indicator and storing it in a memory; and updating the cleaning cycle based on the data stored in the memory after performing a predetermined number of cleanings.

[0025] According to another aspect of the present disclosure, a computer-readable recording medium is provided, which is a non-temporary computer-readable recording medium having instructions recorded thereon, and when the instructions are executed by one or more processors, the one or more processors are caused to perform operations, wherein the instructions include: a step in which the one or more processors determine a plurality of anomaly scores indicating the printing quality of each of the plurality of circuit boards based on at least one indicator indicating the state of solder paste printed on the plurality of circuit boards; a step in which prediction data is generated based on the plurality of anomaly scores, the prediction data being used to indicate the degree of quality degradation of the Nth circuit board when the screen printer is not cleaned and the N circuit boards are printed in sequence; and a step in which a cleaning cycle is determined based on the prediction data and the amount of resources consumed when performing cleaning of the screen printer; wherein the cleaning cycle may correspond to the number of circuit boards printed by the screen printer after cleaning the screen printer and before the next cleaning.

[0026] Effects of the Invention

[0027] According to various embodiments of the present disclosure, a printing cycle of a screen printer can be determined without delaying the printing process while appropriately maintaining the printing quality of a printed circuit board. Therefore, the cleaning cycle can be determined more efficiently than a technique that determines a cleaning cycle based on a value manually input by a user. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 The block diagram briefly shows the structure of the device for determining the cleaning cycle according to one embodiment of the present disclosure.

[0029] Figure 2 An example of a template showing residual solder paste accumulation according to an embodiment of the present disclosure is shown.

[0030] Figure 3 A graph showing measurement data and a predicted range of an indicator indicating a solder paste state according to an embodiment of the present disclosure is shown.

[0031] Figure 4 An embodiment is shown in which a device according to an embodiment of the present disclosure determines a bad score for each pad group, and determines a bad score of a circuit board based on the bad scores of each pad group.

[0032] Figure 5 An embodiment of the present disclosure shows that a device according to an embodiment of the present disclosure determines a defect score of a printed circuit board based on measurement data of an index indicating a state of solder paste.

[0033] Figure 6 An embodiment showing an apparatus according to an embodiment of the present disclosure utilizes survival analysis to determine an optimal cleaning cycle based on measurement data and the amount of resources consumed by cleaning of a screen printing machine.

[0034] Figure 7 The present invention shows an embodiment in which a device according to an embodiment of the present disclosure uses measurement data and survival analysis to determine an optimal cleaning cycle.

[0035] Figure 8 A sequence diagram of a method for determining an optimal cleaning cycle for a device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0036] The various embodiments described in this document are provided for the purpose of clearly describing the technical concept of the present disclosure, and are not intended to be limited to a specific implementation form. The technical concept of the present disclosure includes various modifications, equivalents, alternatives, and embodiments of selectively combining all or part of the embodiments described in this document. In addition, the scope of rights of the technical concept of the present disclosure is not limited to the various embodiments or specific descriptions thereof suggested below.

[0037] Including technical or scientific terms, as long as they are not defined differently, the terms used in this document may have the meaning generally understood by one of ordinary skill in the art to which the present disclosure belongs.

[0038] Expressions such as "including", "may include", "have", "may have", "have", "may have", etc. used in this document mean that the features that become the object (e.g., function, action or component, etc.) exist, and do not exclude the existence of other additional features. In other words, such expressions should be understood as open-ended terms that include the possibility of including other embodiments.

[0039] Singular expressions used in this document may include plural expressions unless the context indicates otherwise. This also applies to singular expressions recorded in the claims.

[0040] As used in this document, the expressions "first", "second", or "the first", "the second", etc. are used to distinguish one object from other objects when referring to multiple objects of the same kind, unless the context indicates otherwise, and do not limit the order or importance of the corresponding objects.

[0041] As used in this document, expressions such as "A, B, and C", "A, B or C", "A, B and / or C", or "at least one of A, B, and C", "at least one of A, B, or C", or "at least one of A, B, and / or C" may refer to each listed item or all possible combinations of the listed items. For example, "at least one of A or B" may refer to all of (1) at least one A, (2) at least one B, or (3) at least one A and at least one B.

[0042] The expression "based on..." used in this document is used to describe one or more factors that influence the decision, judgment, behavior or action described in the sentence or article containing the corresponding expression, and the expression does not exclude additional factors that influence the corresponding decision, judgment, behavior or action.

[0043] When used in this document, the expression that a certain component (for example, the first component) is “connected” or “connected to” another component (for example, the second component) means not only that the certain component is directly connected or connected to the other component, but also that the connection or connection is carried out with another new component (for example, the third component) as the medium.

[0044] The expression "configured to" used in this document may have the meaning of "set to", "having the ability to", "changed to", "formed to", "capable of performing", etc., depending on the context. The corresponding expression is not limited to the meaning of "specially designed in hardware". For example, a processor configured to perform a specific operation may mean a generic-purpose processor that can perform its specific action by running software.

[0045] The following describes various embodiments of the present disclosure with reference to the accompanying drawings. In the drawings and descriptions of the drawings, the same or substantially equivalent components may be given the same reference numerals. In addition, in the following descriptions of various embodiments, repeated descriptions of the same or corresponding components may be omitted, but this does not mean that the corresponding components are not included in the embodiment.

[0046] Figure 1A block diagram briefly illustrating the composition of an apparatus for determining a cleaning cycle according to an embodiment of the present disclosure. According to various embodiments of the present disclosure, the apparatus 110 may include a processor 111, a memory 112, and a communication circuit 113. In addition, the apparatus 110 may further include an output device 114 for outputting a result processed by the processor 111. In some embodiments, at least one of these components of the apparatus 110 may be omitted, or other components may be added to the apparatus 110. In a certain embodiment, a portion of the components may be implemented in an integrated manner or in a single or multiple individuals in addition or alternatively.

[0047] At least some of the components inside and outside the electronic device 10 can be interconnected through a bus, GPIO (general purpose input / output), SPI (serial peripheral interface) or MIPI (mobile industry processor interface) to send and receive data and / or signals.

[0048] The device 110 (e.g., electronic device) may be a device of various forms. For example, the electronic device may be a portable communication device, a computer device, a portable multimedia device, a wearable device, a home appliance device, or a device that is one of the above devices or a combination of the above devices. The device disclosed in the present invention is not limited to the above devices.

[0049] The processor 111 may drive software to control at least one component of the device 110 connected to the processor 111. In addition, the processor 111 may perform various operations related to the present disclosure, such as computing, processing, data generation, and processing. In addition, the processor 111 may load data from the memory 112 or store it in the memory 112.

[0050] The memory 112 may store various data. The data stored in the memory 112 may include software as data obtained, processed, or used by at least one component of the electronic device 110. The memory 112 may include volatile and / or nonvolatile memory.

[0051] The communication circuit 113 can perform wireless or wired communication between the device 110 and other external electronic devices. For example, the communication circuit 113 can perform wireless communication based on LTE (long-term evolution), LTE-A (LTE Advance), CDMA (code division multiple access), WCDMA (wideband CDMA), WiBro (Wireless Broadband), WiFi (wireless fidelity), Bluetooth, NFC (near field communication), GPS (Global Positioning System) or GNSS (global navigation satellite system). For example, the communication circuit 113 can perform wired communication based on USB (universal serial bus), HDMI (high definition multimedia interface), RS-232 (recommended standard 232) or POTS (plain old telephone service). The communication circuit 113 may be omitted in the device 110 according to an embodiment.

[0052] The following is a brief description of the process system 100 of surface mount technology (SMT) for printing printed circuit boards using a screen printer 130. The screen printer 130 may include an operating table, a template (or a metal mask), and a squeegee blade. After the printed circuit board enters the screen printer 130 and is fixed on the operating table, it can be configured so that the template is unfolded on the printed circuit board. The template may include a plurality of openings, and the plurality of openings allow solder paste to be printed only in predetermined areas of the printed circuit board. The screen printer 130 can print solder paste on the template, push the solder paste into the openings of the template using a squeegee blade, and perform soldering on the pads of the printed circuit board. The screen printer then separates the template from the printed circuit board.

[0053] In the process of printing solder paste, the residue of solder paste will accumulate in the opening of the template. When printing multiple printed circuit boards, the residual solder paste will accumulate on the template, and the accumulated residual solder paste will affect the subsequent printing process. For example, when the residual solder paste accumulates in the opening of the template, it will cause under-welding of the solder paste. The opening with a smaller area in the template will be blocked by the residual solder paste, and at the opening with a larger area, a gap will be generated between the printed circuit board and the template due to the residual solder paste, which will cause over-welding of the solder paste. Therefore, welding is not normally achieved on the printed circuit board, and the printing quality will be reduced. In order to prevent this problem, the screen printer 130 can be cleaned periodically. If the cleaning cycle for cleaning the residual solder paste is too short, the printing process will be delayed, and if the cleaning cycle is too long, the printing quality will have problems. Therefore, it is extremely important to determine a cleaning cycle of an appropriate length that does not cause problems with the printing quality and does not delay the printing process.

[0054] In this specification, cleaning the screen printer 130 may mean cleaning the screen printer 130 and / or the template. In the printing process of the circuit board, if the screen printer 130 is cleaned periodically, clogging can be prevented, but if it is cleaned too frequently, the printing process will be delayed, so it is necessary to clean the screen printer 130 at an appropriate period. The following describes a method of measuring the defective score of the printed circuit board and determining the optimal cleaning cycle using survival analysis. The cleaning cycle may mean the number of circuit boards printed after the screen printer 130 is cleaned once until the screen printer 130 is cleaned again. The length of the cleaning cycle may mean the number of printed circuit boards printed during one cleaning cycle.

[0055] On the other hand, after the printed circuit board printed with solder paste is transferred from the screen printer 130 to the inspection device (solder paste inspection; SPI) 120, the inspection device 120 can inspect the printing state of the solder paste on the transferred printed circuit board. For example, the inspection device 120 can irradiate light to the printed circuit board printed with solder paste, and receive light reflected from the printed circuit board to obtain two-dimensional and / or three-dimensional image data of the printed circuit board.

[0056] In addition, the inspection device 120 can compare the image data of the printed circuit board with the reference value or reference data for judging whether the solder paste printing state is good or bad, and inspect whether the solder paste printed on the printed circuit board is bad. For example, the inspection device 120 inspects whether the solder paste printed on the printed circuit board is quantitatively printed at the correct position, thereby inspecting whether the solder paste is bad and generating measurement data representing the inspection result.

[0057] In an embodiment, the measurement data may include at least one of the volume, area, height, width, or gradient of the solder paste printed by means of the screen printer 130 .

[0058] The processor 111 can establish a communication connection with the inspection device 120 (hereinafter referred to as the inspection device) by using the communication circuit 113, and obtain measurement data from the inspection device 120. The measurement data may include at least one of the over-welding degree of the printed solder paste, the under-welding degree of the printed solder paste, the standard deviation of the amount of printed solder paste, the degree of deviation from the alignment position of the circuit board and the template, and the process capability index (Capability of Process, Katayori; CPK). For example, the processor 111 can obtain measurement data of the over-welding or under-welding degree of the solder paste printed on each of the multiple pads included in a printed circuit board, the standard deviation of the amount of printed solder paste, and the degree of deviation from the alignment position of the circuit board and the template. The processor 111 can sequentially obtain measurement data for each of the multiple printed circuit boards printed sequentially from the inspection device 120.

[0059] The processor 111 may determine the bad score of the printed circuit board based on the measurement data. The bad score of the printed circuit board may be a value indicating the printing quality of the printed circuit board. For example, when the bad score of the printed circuit board is high, there will be relatively more pads on the corresponding printed circuit board where solder paste is over-welded or under-welded, and the standard deviation of the amount of solder paste printed on multiple pads will be large. On the contrary, when the bad score of the printed circuit board is low, there will be fewer pads on the corresponding printed circuit board where solder paste is over-welded or under-welded, and the standard deviation of the amount of solder paste printed on multiple pads will be small. The processor 111 determines a reference value of the bad score, and a printed circuit board with a bad score greater than the reference value may be determined to be bad. According to one embodiment, if a printed circuit board with a bad score greater than the reference value is printed, the processor 111 may determine that the screen printer 130 needs to be cleaned. Therefore, the processor 111 may determine a cleaning cycle in which the screen printer 130 may be cleaned before printing a printed circuit board with a bad score greater than the reference value.

[0060] The following is a detailed description of the process of the processor 111 determining the bad score of the printed circuit board. In order to determine the bad score of the printed circuit board, the processor 111 may classify the pads of the printed circuit board into a plurality of pad groups according to the shape. For example, the processor 111 may classify the plurality of pads of the printed circuit board into a first pad group including at least one pad having a first shape and a second pad group including at least one pad having a second shape.

[0061] According to one embodiment, the processor 111 may determine the bad score by pad group, and determine the bad score of the printed circuit board based on the bad score of each pad group. According to one embodiment, the processor 111 may determine the sum of the bad scores of each pad group as the bad score of the printed circuit board. For example, the processor 111 may determine the bad score of the first pad group and the bad score of the second pad group, and determine the sum of the bad scores of the first pad group and the second pad group as the bad score of the printed circuit board. According to another embodiment, the processor 111 may determine the weighted values ​​of each pad group, and determine the bad score of the printed circuit board considering the weighted values ​​of each pad group. For example, the processor 111 may determine the weighted value of the first pad group and the weighted value of the second pad group, and determine the bad score of the printed circuit board by adding the product of the bad score of the first pad group and the weighted value to the product of the bad score of the second pad group and the weighted value. According to one embodiment, the processor 111 may determine the weighted value of each pad group based on the number of pads included in each pad group. The more pads a pad group includes, the higher the weighted value of the corresponding pad group can be set by the processor 111. For example, when the number of pads included in the second pad group is greater than the number of pads included in the first pad group, the processor 111 can set the weighted value of the second pad group to a value higher than the weighted value of the first pad group. For example, if the number of pads included in the pad group is less than a predetermined number, the processor 111 can determine the weighted value of the corresponding pad group to be 0.

[0062] According to one embodiment, the processor 111 can determine the bad score of the pad group based on the measurement data, determine the bad score of at least one indicator for the corresponding pad group, and determine the bad score of the pad group based on the bad score of at least one indicator. The bad score of each indicator is an indication of the over-welding degree, under-welding degree, standard deviation of the amount of printed solder paste, degree of deviation from the alignment position of the circuit board and the template, and the value of the process capability index. For example, the processor 111 can determine the first data indicating the over-welding degree of the solder paste printed on at least one pad included in the first pad group, the second data indicating the under-welding degree of the solder paste printed on at least one pad included in the first pad group, the third data indicating the standard deviation of the solder paste printed on at least one pad included in the first pad group, the fourth data indicating the degree of deviation from the alignment position of the printed circuit board and the template when printing solder paste on at least one pad included in the first pad group, and the fifth data indicating the process capability index of the process of printing solder paste on at least one pad included in the first pad group based on the measurement data. The processor 111 may determine the bad score of the first pad group based on the first data to the fifth data. For example, the processor 111 may determine the sum of the first data to the fifth data as the bad score of the first pad group. According to another embodiment, the processor 111 may determine the weighted values ​​of each indicator, and determine the bad score of the pad group considering the weighted values ​​of each indicator. For example, the weighted value of the first indicator and the weighted value of the second indicator may be determined, and the sum of the product of the first data corresponding to the first indicator and the weighted value of the first indicator and the product of the second data corresponding to the second indicator and the weighted value of the second indicator may be determined as the bad score of the pad group.

[0063] In order to determine the bad score of each indicator, the processor 111 may set a prediction range for each indicator and determine the bad score for each indicator based on the degree to which the measured data deviates from the prediction range. The processor 111 may analyze the originally stored data to set the prediction range for each indicator.

[0064] According to an embodiment, the processor 111 may use regression analysis methodology to set the prediction range of each indicator. For example, the processor 111 may use the following mathematical formula 1 for predicting the printed solder paste amount Y to set the prediction range of each indicator.

[0065]

Mathematical formula 1

[0066] Y=a*X1+b*X2+c*X3+d*X4+e*X5+f

[0067] In the mathematical formula 1, X1 may refer to the printing direction of the screen printer 130. For example, the screen printer 130 may print the circuit board in the forward direction 1 or the backward direction 0. X2 may refer to the number of circuit boards previously printed by the screen printer 130 after cleaning. For example, when the cleaning cycle is 3, X2 may have a value between 1 and 3. X3 to X5 may refer to various parameters related to the amount of solder paste Y printed on the circuit board (for example, pressure, speed when printing the circuit board, and separation speed of the circuit board, etc.). However, X1 to X5 are not limited to the aforementioned contents, and may include various parameters that can determine the amount of solder paste Y printed. a, b, c, d, e, and f may be regression coefficients (coefficients) related to X1 to X5, respectively. The processor 111 may determine the regression coefficients (a to f) using the originally stored data. For example, the printing direction of the screen printer 130, the number of circuit boards previously printed by the screen printer 130 after cleaning, the pressure and speed when printing the circuit board may be confirmed using the originally stored data, and the regression coefficients may be determined using the confirmed values.

[0068] According to one embodiment, the determined regression coefficient may have an error, and therefore, the processor 111 may determine the value Y predicted by the mathematical formula 1 as a credible prediction range (confidence interval) of each indicator. The prediction range of each indicator is a value of the range of changes in the measurement data for each indicator predicted according to the number of printed circuit boards when the screen printer 130 is not cleaned and the circuit board is printed. The processor 111 may determine the bad score of each indicator based on the degree to which the measurement data deviates from the prediction range. For example, the greater the degree to which the measurement data deviates from the prediction range, the higher the bad score of each indicator may be set by the processor 111. On the contrary, the closer the measurement data is to the prediction range, the lower the bad score of each indicator may be set by the processor 111.

[0069] To summarize the process of the processor 111 determining the bad score of the printed circuit board based on the measurement data described above, the processor 111 can determine the bad score of each indicator based on the degree to which the measurement data deviates from the prediction range, determine the bad score of each pad group based on the bad score of each indicator, and determine the bad score of the printed circuit board based on the bad score of each pad group. The method for the processor 111 to determine the bad score of the printed circuit board based on the measurement data is not limited to the above-mentioned embodiments, and the bad score of the printed circuit board can be determined by various methods that can measure the printing quality of the printed circuit board.

[0070] The processor 111 may generate prediction data based on the bad score. According to one embodiment, the processor 111 may generate prediction data using survival analysis. The prediction data is data indicating the average quality loss when each printed circuit board is printed when the length of the cleaning cycle is N. Survival analysis is a method for analyzing the time taken until a specific phenomenon occurs. When the screen printer 130 is not cleaned and the printed circuit boards are printed in sequence, the processor 111 may use survival analysis to determine the time point when the printed circuit boards with a bad score greater than the reference value are printed. According to one embodiment, when multiple printed circuit boards are printed in sequence without cleaning, the last printed circuit board will have a higher bad score. However, if the screen printer 130 is cleaned after printing a predetermined number of printed circuit boards according to the set cleaning cycle, it is difficult to directly calculate the bad score of the circuit board printed after printing the predetermined number of circuit boards. Therefore, right censoring of the cleaning screen printer 130 occurs before the defect score reaches a predetermined value, and thus the survival analysis method based on this can be used to predict the defect score when printing more than a predetermined number of circuit boards according to the cleaning cycle, and generate prediction data based on the predicted defect score. For example, when the length of the cleaning cycle is 3, the screen printer 130 can print the first printed circuit board that is printed first after cleaning, the second printed circuit board that is printed second, and the third printed circuit board that is printed third in one cleaning cycle. At this time, the first printed circuit board is printed for the first time after cleaning, so there is no residual solder paste on the template and no quality degradation, but the second printed circuit board may have quality degradation due to residual solder paste generated on the template during the printing process of the first printed circuit board. The third printed circuit board may also have quality degradation due to the accumulation of residual solder paste on the template during the printing process of the first printed circuit board and the second printed circuit board. At this time, the value indicating the degree of quality degradation (e.g., quality degradation occurring in the second printed circuit board and quality degradation occurring in the third printed circuit board) occurring in all circuit boards printed in one cleaning cycle is divided by the length of the cleaning cycle (e.g., 3) to generate prediction data. If the cleaning cycle becomes longer, the residual solder paste accumulated on the template will increase accordingly, so the predicted data will increase with the increase of cleaning cycle.

[0071] Similarly, the processor 111 can determine the average time loss and the average resource loss. A predetermined time is required to clean the screen printer 130, and during the corresponding time period, the screen printer 130 cannot print a circuit board, thereby incurring a loss. For example, during the time required to clean the screen printer 130 once, the circuit board cannot be printed, thereby incurring a loss of overall process delay. In addition, material resources (e.g., solvents, IPA (Isopropyl Alcohol), cleaning paper, etc.) are consumed to clean the screen printer 130, thereby incurring a loss. The average time loss is a value obtained by dividing the loss incurred due to the time required to clean the screen printer 130 by the length of the cleaning cycle, and the average resource loss is a value obtained by dividing the loss incurred according to the resource cost consumed by cleaning the screen printer 130 by the length of the cleaning cycle. The longer the cleaning cycle length, the lower the average time loss and the average resource loss. The time required to clean the screen printer 130 and the cost of the material resources consumed are fixed, so the longer the cleaning cycle, the lower the average time loss and the average resource loss.

[0072] According to an embodiment, the memory 112 may store the time and cost data of cleaning the screen printer 130. The processor 111 may obtain the time and cost data of cleaning from the memory 112, and determine the average time loss and average resource loss based on the length of the cleaning cycle based on the obtained data.

[0073] The processor 111 may determine the optimal cleaning cycle based on the predicted data, the average time loss, and the average resource loss. According to one embodiment, the processor 111 may determine the cleaning cycle length that minimizes the sum of the predicted data, the average time loss, and the average resource loss as the optimal cleaning cycle length. For example, the optimal cleaning cycle may be determined using the following mathematical formula 2.

[0074]

Mathematical formula 2

[0075]

[0076] L(n): Objective function for finding the optimal cleaning cycle

[0077] n: length of the cleaning cycle

[0078] C r : Additional time and resources spent on cleaning as poor quality printed circuit boards are produced

[0079] C d : Time loss in cleaning a screen printing machine

[0080] C m : Resource loss caused by cleaning a screen printing machine

[0081] Q n : Quality degradation occurs in the nth printed circuit board

[0082] t d : Time required to clean a screen printing machine

[0083] t p : The time required to print a printed circuit board

[0084] The processor 111 may update the cleaning cycle after performing a predetermined number of cleanings. According to one embodiment, the processor 111 may receive data indicating the execution of cleaning of the screen printer 130 from the screen printer 130 or the inspection device 120. After the cleaning of the screen printer 130 is performed a predetermined number of times (e.g., 4 times), the processor 111 may update the length of the cleaning cycle using the measurement data accumulated during this period.

[0085] The processor 111 may transmit the information of the determined cleaning cycle to the screen printer 130, or transmit the instruction indicating cleaning to the screen printer 130. According to one embodiment, the processor 111 may transmit the information of the determined cleaning cycle to the screen printer 130, so that the cleaning cycle is displayed on the display screen of the screen printer 130. The screen printer 130 may calculate the number of printed circuit boards in response to receiving the data of the cleaning cycle, and after printing the circuit boards corresponding to the length of the cleaning cycle, stop the printing action and clean the template. According to one embodiment, the screen printer 130 may provide a cleaning notification to the user through the display screen while cleaning.

[0086] Figure 2 An example of a template with residual solder paste accumulation according to an embodiment of the present disclosure is shown. During the process of printing solder paste on a plurality of printed circuit boards by a screen printer, residual solder paste may accumulate in the opening of the template 200. Figure 2 , the template 200 may include a plurality of openings corresponding to a plurality of pads having various areas and shapes, and residual solder paste may accumulate in the plurality of openings. According to one embodiment, the impact caused by the residual solder paste may vary depending on the area of ​​the opening. For example, in an opening 201 having an area smaller than a predetermined size, a clogging phenomenon may occur due to residual solder paste. Due to the clogging phenomenon, the pads corresponding to the opening 201 through which the solder paste cannot pass smoothly may be under-welded with solder paste. On the contrary, in an opening 202 having an area larger than a predetermined size, residual solder paste may accumulate and harden on the side or bottom of the opening 202. Therefore, during the soldering process, the template 200 cannot be closely attached to the printed circuit board and over-welding of the solder paste may occur.

[0087] Therefore, the screen printing machine can perform cleaning according to a predetermined cleaning cycle. Cleaning of the screen printing machine means cleaning of the screen printing machine and / or the template 200. The screen printing machine can obtain data for the cleaning cycle and perform cleaning based on the obtained data. Figures 3 to 8 The method for determining the optimal cleaning cycle of a screen printing machine is described in detail.

[0088] Figure 3 A chart showing measurement data and prediction ranges of indicators indicating solder paste status representing an embodiment of the present disclosure is shown. The processor may obtain measurement data of indicators indicating solder paste status from an inspection device. The indicators indicating solder paste status may include: the degree of over-soldering or under-soldering of the printed solder paste for each of the multiple pads included in the printed circuit board, the standard deviation of the amount of printed solder paste, the degree of deviation from the alignment position of the circuit board and the template, and at least one of the process capability index. The processor may determine the bad scores of the various indicators indicating solder paste status based on the obtained measurement data. The processor may determine the bad scores for the various indicators based on the difference between the predicted value 302 and the actual value 301 of the indicator indicating solder paste status. For example, referring to Figure 3 , the processor can determine the defective scores of the over-soldering degree and the under-soldering degree of the solder paste based on the amount of solder paste printed on the printed circuit board in the measurement data. Figure 3 In the figure, the x-axis is the number of printed circuit boards printed by the screen printer, and the y-axis is the amount of solder paste printed on the printed circuit board. Figure 3 The amount of solder paste predicted based on the number of printed circuit boards (predicted value) 302 , the error ranges 303 , 304 , and 305 of the solder paste, and the actual measured value 301 based on the measurement data are shown in a graph form 300 .

[0089] The processor may set credible intervals 303, 304, 405 with a predetermined error range from the predicted value 302. For example, the processor may set a first credible interval 303 with a first error range from the predicted value, a second credible interval 304 with a second error range, and a third credible interval 305 with a third error range. The processor may use the data stored in the original memory to set the predicted value 302 and the credible intervals 303, 304, 305 with a predetermined error range. The processor may determine the defective score of the printed circuit board based on the difference between the measured value 301 and the predicted value 302. For example, in Figure 3 In the chart of FIG. 3 , the processor can determine the defect score of the solder paste over-soldering degree when the measured value 301 is greater than the predicted value 302, and can determine the defect score of the solder paste under-soldering degree when the measured value 301 is less than the predicted value 302. The greater the error between the measured value 301 and the predicted value 302, the higher the defect score of the solder paste over-soldering and under-soldering degree can be determined.

[0090] The processor may determine a higher bad score as the difference between the predicted value 302 and the measured value 301 is greater. For example, the second bad score of the printed circuit board when the measured value 301 is within the second credible interval 304 may be higher than the first bad score of the printed circuit board when the measured value 301 is within the first credible interval 303, and the third bad score of the printed circuit board when the measured value 301 is within the third credible interval 305 may be higher than the second bad score.

[0091] The processor can determine a cleaning cycle of the screen printing machine, and the screen printing machine can perform cleaning according to the determined cleaning cycle. Figure 3 , the processor can set the length of the cleaning cycle to 10. The screen printer can clean the template after printing 10 printed circuit boards. The cleaning results are Figure 3 It can be confirmed from the graph 300 of FIG. 1 that the amount of solder paste printed on the printed circuit board decreases sharply after the number of printed circuit boards reaches 10. The processor can determine the defective score in the same way for other indicators indicating the state of the solder paste. The various methods for the processor to determine the defective score of the printed circuit board will be described in detail below. Figure 4 and Figure 5 Described in detail in.

[0092] Figure 4 An embodiment of the present disclosure is shown in which a device determines a bad score by pad group, and determines a bad score of a circuit board based on the bad score of each pad group. The processor may divide a plurality of pads included in a printed circuit board into at least one pad group (e.g., group 1, group 2, ..., group n) according to their shapes. Each pad group may include at least one pad having the same shape. The processor may determine a bad score by pad group. The processor may determine a bad score for each pad group based on measurement data, and determine a bad score for the printed circuit board based on the bad score of each pad group. According to one embodiment, the processor may arrange a plurality of pad groups according to their area sizes from the first pad group to the nth pad group. For example, the first pad group may include the pad with the largest area among all pad groups, and the nth pad group may include the pad with the smallest area among all pad groups. However, the manner of arranging a plurality of pad groups is not limited thereto.

[0093] Figure 4Each matrix may be a matrix for the processor to determine the bad score of the printed circuit board. Each upper matrix 400, 410, 420, 430, 440 corresponds to an index indicating the state of the solder paste, and the lower matrix 450 is a matrix for statistically analyzing the data of all upper matrices 400, 410, 420, 430, 440. For example, the first matrix 400 may correspond to the degree of over-soldering of the solder paste, the second matrix 410 may correspond to the degree of under-soldering of the solder paste, the third matrix 420 may correspond to the process capability index, the fourth matrix 430 may correspond to the degree of deviation from the alignment position of the printed circuit board and the template, the fifth matrix 440 may correspond to the standard deviation of the solder paste amount, and all cells of the first matrix 400 to the fifth matrix 440 may correspond to the bad score of the index corresponding to each matrix. For example, each cell of the matrix may store the bad score corresponding to the index of the matrix. The lower matrix 450 may store the result of statistically analyzing the data of the bad scores stored in the first matrix 400 to the fifth matrix 440. For example, each cell in a column of the lower matrix 450 may represent a defective score of a pad group in a printed circuit board.

[0094] Each column of the matrix may correspond to a printed circuit board. For example, the first column 401 of the first matrix 400 may store a bad score for a recently printed circuit board. Each row of the matrix may correspond to one of a plurality of pad groups included in the printed circuit board. For example, the first row of the first matrix 400 to the fifth matrix 440 may store a bad score for a first pad group among a plurality of pad groups.

[0095] The processor may determine the bad scores of each indicator for at least one pad belonging to the pad group. For example, the processor may determine the bad scores of multiple indicators for the i-th pad group. The i-row 1 column 402 of the first matrix may store the bad scores indicating the over-welding degree of the solder paste printed on the pad belonging to the i-th pad group of the most recently printed circuit board, and the i-row 1 column 412 of the second matrix may store the bad scores indicating the under-welding degree of the solder paste printed on the pad. The processor may store the bad scores of each indicator of the pad belonging to the i-th pad group in the i-row 1 column 402, 412, 422, 432, 442 of the first matrix to the fifth matrix.

[0096] The processor may count the bad scores of each indicator to determine the bad scores of each pad group. For example, the processor may count the bad scores 402 of the first indicator to the bad scores 442 of the fifth indicator for the i-th pad group to determine the bad score 452 of the i-th pad group. On the matrix, the processor may count the bad scores stored in the i-row 1-column cells 402, 412, 422, 432, 442 of the first matrix 400 to the fifth matrix 440 to determine the data to be stored in the i-row 1-column 452 of the lower end matrix 450. According to one embodiment, the processor may set the weighted values ​​of each indicator and use the weighted values ​​to determine the bad scores of each pad group. The processor may add the product of the weighted values ​​of each indicator of the pad group and the bad score to determine the bad score of the corresponding pad group.

[0097] The processor may perform statistics on the bad scores of each pad group included in the printed circuit board to determine the bad score of the printed circuit board. According to one embodiment, the processor may set a weighted value for each pad group, and determine the bad score of the printed circuit board considering the weighted value. For example, the processor may determine the weighted value of each pad group based on the number of pads included in each pad group. For example, the more pads included, the larger the weighted value of each pad group may be set by the processor. According to one embodiment, the processor may multiply the weighted value of each pad group by the bad score and then add them all up to determine the bad score of the printed circuit board.

[0098] Figure 5 The present invention shows an embodiment in which a device according to an embodiment of the present invention determines a defect score of a printed circuit board based on measurement data of an indicator indicating a solder paste state. Figure 4 The description content that is repeated will not be repeated. Figure 5 Show the use Figure 4 An embodiment of determining a defective score of a printed circuit board using a matrix as described in the accompanying drawings. Figure 5 A graph 500 shows first to fifth matrices 510 to 550 corresponding to a plurality of indicators indicating solder paste states and defect scores of printed circuit boards determined by counting the first to fifth matrices 510 to 550.

[0099] The first matrix 510 may represent the bad scores of each pad group of the over-soldering degree of the solder paste, the second matrix 520 may represent the bad scores of each pad group of the under-soldering degree of the solder paste, the third matrix 530 may represent the bad scores of each pad group of the process capability index of the printing process, the fourth matrix 540 may represent the bad scores of each pad group of the degree of deviation from the alignment position of the printed circuit board and the template, and the fifth matrix 550 may represent the bad scores of each pad group of the standard deviation of the amount of printed solder paste. In the matrix, the higher the bad score of each cell, the darker the color. The processor can count the bad scores of each indicator of each matrix to determine the bad score of the printed circuit board.

[0100] In the graph 500, the x-axis indicates the number of printed circuit boards, and the y-axis indicates the defective score. In the graph 500, it can be confirmed that the defective score 503 of the 71st and 72nd printed circuit boards is higher than the reference value. Corresponding to the determination that the defective score 503 of the 71st and 72nd printed circuit boards is higher than the reference value, the processor can determine the cleaning cycle of the screen printer.

[0101] Figure 6 The apparatus of an embodiment of the present disclosure uses survival analysis and determines the optimal cleaning cycle based on the measurement data and the amount of resources consumed by the screen printing machine cleaning. The processor can generate prediction data based on the bad score and calculate the average time loss 602 and the average resource loss 601 to determine the optimal cleaning cycle. Figure 6 A graph 600 is shown describing a method by which a processor determines an optimal wash cycle based on forecast data 603 , average time loss 602 , and average resource loss 601 .

[0102] The processor may calculate the defect score of the Nth printed circuit board when the screen printer is not cleaned and prints N circuit boards using survival analysis, and may generate prediction data 603 as a value indicating the degree of quality degradation of the Nth printed circuit board. Figure 6 If the screen printer continues to print the printed circuit board without cleaning, the amount of solder paste accumulated on the template will increase, so the prediction data 603 can be expressed as a graph tilted to the upper right.

[0103] The processor may determine the average time loss 602 and the average resource loss 601 based on the time and resources consumed for cleaning the screen printing machine once. According to one embodiment, the processor may obtain the time and resource related data required for cleaning the screen printing machine once from the memory. Since the required time and resources have been determined, the longer the cleaning cycle is, the smaller the average time loss 602 obtained by dividing the loss incurred by the required time by the cleaning cycle and the average resource loss 601 obtained by dividing the loss incurred by the required resources by the cleaning cycle will be. Therefore, in Figure 6 In the graph 600 , the average time loss 602 and the average resource loss 601 may be represented as graphs sloping downward to the right.

[0104] The processor may determine the cleaning cycle in which the sum of the prediction data 603, the average time loss 602, and the average resource loss 601 reaches the minimum as the optimal cleaning cycle. Figure 6 , when the cleaning cycle is 2, the total loss 610 is the smallest, so the processor can determine the cleaning cycle to be 2.

[0105] Figure 7The present invention shows an embodiment of a device according to an embodiment of the present invention using measurement data and survival analysis to determine an optimal cleaning cycle. The processor can obtain measurement data from the inspection device. According to an embodiment, the processor can start T in the sequence used to determine the optimal cleaning cycle. s After that, in the first data collection period P of the screen printing machine cleaning performed a predetermined number of times (for example, 3 times), datacollect1 During the first data collection interval P datacollect1 In the first data collection interval P, the processor can obtain measurement data from the inspection device but does not calculate the bad score. According to one embodiment, the processor can be based on the first data collection interval P datacollect1 The obtained measurement data sets the prediction range of each indicator used to determine the bad score.

[0106] The processor may be configured to perform a datacollect2 Increase the cleaning cycle and collect data. According to one embodiment, the second data collection interval P datacollect2 It may be a period of time during which a predetermined number of screen printing machine cleanings are performed. According to one embodiment, the processor may datacollect2 Confirm whether the predicted data generated by survival analysis are consistent with the actual measured values.

[0107] The processor may be configured to collect data at first and second data collection intervals P datacollect1 , P datacollect2 After completion, determine the optimal cleaning cycle for the screen printing machine as described above. Figure 7 , the processor may determine the length of the cleaning cycle to be 4. According to one embodiment, the processor may update the cleaning cycle after the screen printing machine has been cleaned a predetermined number of times. When updating the cleaning cycle, the processor may update the cleaning cycle using all accumulated data. For example, referring to Figure 7 , the processor can update the cleaning cycle after performing 4 cleanings. At the first update time point of the cleaning cycle, the processor can update the cleaning cycle based on the data collected in the second data collection interval P datacollect2 The period T from the end to the first update time point co1 At a second update time point in the cleaning cycle, the processor may update the cleaning cycle based on the data accumulated during the second data collection interval P. datacollect2 The period T from the end to the second update time point co2 The accumulated data is used to update the cleaning cycle. At the third update time point and the fourth update time point, all the accumulated data can be used to update the cleaning cycle. Thus, the processor can determine the cleaning cycle that reflects the current state of the screen printing machine and can always maintain the optimal cleaning cycle.

[0108] Various embodiments of the present disclosure may be implemented in software in a machine-readable storage medium. The software may be software for implementing various embodiments of the present disclosure. The software may be inferred from various embodiments of the present disclosure by a programmer skilled in the art of the present disclosure. For example, the software may be a program including machine-readable instructions (e.g., code or code fragments). A machine is a device that can operate according to instructions extracted from a storage medium, such as a computer. In one embodiment, the machine may be the device 110 of an embodiment of the present disclosure. In one embodiment, the processor of the machine may run the extracted instructions, and the constituent elements of the machine may perform functions corresponding to the corresponding instructions. In one embodiment, the processor may be the processor 111 of an embodiment of the present disclosure. The storage medium may refer to all kinds of recording media (recording medium) that store data readable by a machine. The storage medium may include, for example, ROM (read-only memory), RAM (random access memory), CD-ROM (read-only optical disk drive), magnetic tape, floppy disk, optical data storage device, etc. In one embodiment, the storage medium may be a memory 112. In one embodiment, the storage medium may also be implemented in a distributed form in a computer system connected via a network. The software may be distributed, stored, and run in a computer system, etc. The storage medium may be a non-transitory storage medium. A non-transitory storage medium refers to a tangible medium that exists independently of semi-permanent or temporary storage of data, and does not include a signal that is transmitted transiently.

[0109] Figure 8 A sequence diagram of a method for determining an optimal cleaning cycle for an apparatus according to an embodiment of the present disclosure. According to an embodiment, in step 800, the apparatus may determine a defect score indicating the quality of a circuit board based on data of at least one indicator indicating a process state. The apparatus may obtain measurement data of multiple indicators indicating a solder paste state from an inspection device. The apparatus may set a predicted value and a predicted range for each indicator data based on the stored data, and determine the defect score based on the difference between the measured value and the predicted value.

[0110] According to one embodiment, the device may classify a plurality of pads included in the printed circuit board into a plurality of pad groups according to their shapes, and determine a bad score for each pad group. The device may determine the bad score of each indicator, and perform statistics on the bad scores of each indicator to determine the bad score of the pad group. According to one embodiment, the device may determine the weighted value of each indicator, and determine the bad score of the pad group using the weighted value of the indicator. For example, the device may determine the value of the sum of the products of the weighted values ​​of each indicator and the bad score as the bad score of each pad group. The device may perform statistics on the bad scores of each pad group to determine the bad score of the printed circuit board. According to one embodiment, the device may determine the weighted value of each pad group, and determine the bad score of the printed circuit board based on the determined weighted value. For example, the device may determine the value of the sum of the products of the bad scores of each pad group and the weighted value as the bad score of the printed circuit board.

[0111] According to an embodiment, in step 802, the apparatus may generate prediction data indicating the degree of quality degradation of the Nth circuit board when the screen printer is not cleaned and the N circuit boards are printed in sequence. The apparatus may use survival analysis to predict the defective scores of printed circuit boards for which actual measurement data cannot be obtained. The apparatus may generate prediction data indicating the degree of quality degradation of the printed circuit board based on the predicted defective scores.

[0112] According to an embodiment, in step 804, the device may determine the length of the cleaning cycle based on the predicted data and the amount of resources consumed when performing cleaning of the screen printing machine. The device may calculate the average time loss and average resource loss that vary according to the length of the cleaning cycle, and determine the cleaning cycle when the sum of the average time loss, the average resource loss and the predicted data is the smallest as the optimal cleaning cycle.

[0113] According to one embodiment, the device may update the cleaning cycle after performing a predetermined number of screen printer cleanings. Since the optimal cleaning cycle may be changed according to the state of the screen printer or the state of the printed circuit board, the device may update the cleaning cycle after performing a predetermined number of cleanings. The device may update the cleaning cycle based on all measurement data accumulated up to the update time point.

[0114] exist Figure 8 In the sequence diagram shown, each step of the method or algorithm of the present disclosure is described in sequence, but each step can be performed in any combination of the present disclosure in addition to being performed in sequence. The description based on this flowchart excludes the situation where no changes or revisions are applied to the method or algorithm, and does not mean that any step is necessary or recommended. In one embodiment, at least a portion of the steps can be performed in parallel, repeatedly or exploratory. In one embodiment, at least a portion of the steps can be omitted, or other steps can be added.

[0115] The above describes the technical concept of the present disclosure based on various embodiments, but the technical concept of the present disclosure includes various substitutions, deformations and changes that can be realized within the scope understandable to those skilled in the art of the present disclosure. In addition, such substitutions, deformations and changes should be understood to be included in the attached claims.

Claims

1. A device for determining an optimal cleaning cycle of a screen printing machine, comprising: More than one processor; and One or more memories storing instructions, which, when executed by the one or more processors, cause the one or more processors to perform operations; wherein the one or more processors determine a plurality of defect scores indicating the printing quality of each of the plurality of circuit boards based on at least one indicator indicating the state of solder paste printed on the plurality of circuit boards, generating prediction data based on the plurality of defect scores, the prediction data being used to indicate the degree of degradation in printing quality of the Nth circuit board when the screen printer is not cleaned before printing the N circuit boards in sequence, and determining a cleaning cycle based on the prediction data and the amount of resources consumed when performing cleaning of the screen printing machine, The cleaning cycle corresponds to the number of circuit boards printed by the screen printer after cleaning the screen printer and before the next cleaning.

2. The device according to claim 1, wherein: the one or more processors, Determine a weighted value corresponding to each of the at least one indicator, determining a forecast range for the at least one indicator, and The plurality of defect scores for each of the plurality of circuit boards are determined based on the degree to which each of the at least one indicator exceeds the prediction range and the weighted value.

3. The device according to claim 2, wherein: the one or more processors, determining a defect score of each pad group according to a plurality of pad groups distinguished according to pad shapes, and The plurality of bad scores for each of the plurality of circuit boards are determined based on the bad scores of the respective pad groups.

4. The device according to claim 3, wherein: the one or more processors, determining a weighted value of each pad group based on the number of pads included in each of the plurality of pad groups, and The plurality of bad scores for each of the plurality of circuit boards are determined based on the bad scores of the respective pad groups and weighted values ​​of the respective pad groups.

5. The device according to claim 4, wherein: The one or more processors determine the weight value of a pad group including a predetermined number of pads or less as 0 among the plurality of pad groups.

6. The device according to claim 1, wherein: The at least one indicator comprises: At least one selected from the group consisting of an over-soldering degree of the printed solder paste, an under-soldering degree of the printed solder paste, a standard deviation of the amount of the printed solder paste, and a degree of deviation from an alignment position between the circuit board and the template.

7. The device according to claim 1, wherein: Also includes a communication circuit connected to the screen printing machine in a communicative manner, The communication circuit transmits the cleaning cycle to the screen printer, The screen printing machine performs a predetermined action in response to the reception of the cleaning cycle so as to perform cleaning according to the cleaning cycle.

8. The device according to claim 1, wherein: The one or more processors utilize survival analysis to generate the prediction data.

9. The device according to claim 1, wherein: The resources consumed when performing the cleaning include the time required for cleaning the screen printing machine and the cost of purchasing items.

10. The device according to claim 9, wherein: the one or more processors, The cleaning cycle that can minimize the amount of resources consumed and the reduction in printing quality when performing cleaning of the screen printing machine is determined based on the prediction data, the time required for cleaning the screen printing machine, and the cost of purchasing items.

11. The device according to claim 1, wherein: the one or more processors, continuously obtaining the at least one indicator and storing it in the memory, and After a predetermined number of cleanings are performed, the cleaning cycle is updated based on the data stored in the memory.

12. A method for determining an optimal cleaning cycle of a screen printing machine, comprising: The step of determining a plurality of defect scores indicating a printing quality of each of the plurality of circuit boards based on at least one indicator indicating a state of solder paste printed on the plurality of circuit boards; A step of generating prediction data based on the plurality of defect scores, the prediction data being used to indicate the degree of quality degradation of the Nth circuit board when the screen printer is not cleaned and the N circuit boards are printed in sequence; and the step of determining a cleaning cycle based on the predicted data and an amount of resources consumed when performing cleaning of the screen printing machine; The cleaning cycle corresponds to the number of circuit boards printed by the screen printer after cleaning the screen printer and before the next cleaning.

13. The method according to claim 12, wherein: The steps of determining the bad score include: The step of determining a weighted value corresponding to each of the at least one indicator; The step of determining a prediction range for said at least one indicator; and The step of determining the plurality of defect scores for each of the plurality of circuit boards based on the degree to which each of the at least one indicator exceeds the prediction range and the weighted value.

14. The method according to claim 13, wherein: The steps of determining the bad score include: The step of determining a defect score of each pad group according to a plurality of pad groups classified according to pad shapes; and The step of determining the plurality of bad scores for each of the plurality of circuit boards based on the bad scores of the respective pad groups.

15. The method according to claim 14, wherein: The steps of determining the bad score include: A step of determining a weighted value of each pad group based on the number of pads included in each of the plurality of pad groups; and The step of determining the plurality of bad scores for each of the plurality of circuit boards based on the bad scores of the respective pad groups and the weighted values ​​of the respective pad groups.

16. The method according to claim 15, wherein: The step of determining weighted values ​​according to the plurality of pad groups further comprises: The step of determining the weight value of a pad group including a pad number less than a predetermined number among the plurality of pad groups to be 0.

17. The method according to claim 12, wherein: The at least one indicator comprises: At least one selected from the group consisting of an over-soldering degree of the printed solder paste, an under-soldering degree of the printed solder paste, a standard deviation of the amount of the printed solder paste, and a degree of deviation from an alignment position between the circuit board and the template.

18. The method according to claim 12, wherein: The step of generating the prediction data includes the step of generating the prediction data using survival analysis.

19. The method according to claim 12, wherein: The steps of determining the cleaning cycle include: The step of continuously obtaining the at least one indicator and storing it in a memory; and The step of updating the cleaning cycle based on the data stored in the memory after performing a predetermined number of cleanings.

20. A computer-readable recording medium, wherein the computer-readable recording medium is a non-transitory computer-readable recording medium having instructions recorded thereon, wherein when the instructions are executed by one or more processors, the one or more processors are caused to perform operations, wherein: The instructions include: The one or more processors determine a plurality of defect scores indicating a printing quality of each of the plurality of circuit boards based on at least one indicator indicating a state of solder paste printed on the plurality of circuit boards; A step of generating prediction data based on the plurality of defect scores, the prediction data being used to indicate the degree of quality degradation of the Nth circuit board when the screen printer is not cleaned and the N circuit boards are printed in sequence; and the step of determining a cleaning cycle based on the predicted data and an amount of resources consumed when performing cleaning of the screen printing machine; The cleaning cycle corresponds to the number of circuit boards printed by the screen printer after cleaning the screen printer and before the next cleaning.

Images