Fault mode influence judgment method and semiconductor process equipment
By using the scoring weights and total scores of multiple evaluation subjects in the control system of the semiconductor process equipment to determine the final score of the fault, the problem of poor accuracy of the fault analysis results in the prior art is solved, and the accuracy of the fault mode impact determination is improved.
Patent Information
- Application Number
- CN202311521391.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-11-15
- Publication Date
- 2025-05-16
AI Technical Summary
In the control system of existing semiconductor process equipment, the failure mode impact analysis method mainly relies on expert evaluation, resulting in poor accuracy of the results and is greatly affected by subjective consciousness.
By obtaining multiple evaluation factors scores for failures of semiconductor process equipment by multiple evaluation subjects, the scoring weights and total evaluation factors of each evaluation subject are calculated, and the final scores of each failure are determined, weakening or eliminating the subjective awareness influence of the evaluation subject.
The accuracy of the failure mode affecting the determination result of the semiconductor device control system is improved, and faults with higher order positions can be detected and prioritized to more accurately detect and prioritize fault damage.
Smart Images
Figure CN120013306A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of semiconductor manufacturing, and in particular to a method for determining the impact of a failure mode and semiconductor process equipment. Background Art
[0002] Fault Mode Effect Analysis (FMEA) is a method that conducts a comprehensive risk analysis of faults and sorts faults according to the analysis results to determine the priority of fault handling. Since the smooth operation of the pressure control system of semiconductor process equipment is closely related to the quality of the semiconductor products it produces, using the fault mode effect analysis method to conduct risk analysis on the control system of semiconductor process equipment is an effective means to ensure the safe and stable operation of semiconductor equipment. However, the fault mode effect analysis method currently used in the control system of semiconductor process equipment is mainly based on expert evaluation. The final output result of this method is greatly affected by the subjective consciousness of the experts, resulting in poor accuracy of the fault analysis results. Summary of the invention
[0003] The invention discloses a fault mode impact determination method and semiconductor process equipment, so as to improve the accuracy of the fault mode impact determination result of the control system of the semiconductor equipment.
[0004] In a first aspect, the present invention discloses a method for determining the impact of a fault mode, which is applied to semiconductor process equipment. The method comprises: obtaining scores of a types of evaluation factors for m faults of the semiconductor process equipment from n evaluation subjects, where n, m and a are all integers greater than 1; determining the scoring weights and the total score of the evaluation factors for each fault by each evaluation subject based on the scores of a types of evaluation factors for the m faults by the n evaluation subjects; determining the final score for each fault based on the scoring weights and the total score of the evaluation factors for each fault by the n evaluation subjects; and obtaining a fault mode impact determination result based on the final scores of the m faults.
[0005] In some embodiments, determining the scoring weight of each fault by each evaluation subject based on the a types of evaluation factors scored by the n evaluation subjects for the m faults includes: determining the scoring offset entropy of the a types of evaluation factors scored by each evaluation subject for each fault based on the a types of evaluation factors scored by the n evaluation subjects for the m faults; determining the scoring weight of each fault by each evaluation subject based on the scoring offset entropy of the a types of evaluation factors scored by each evaluation subject for each fault.
[0006] In some embodiments, determining the score offset entropy of the a kind of evaluation factor scores of each evaluation subject for each fault according to the a kind of evaluation factor scores of the m faults by the n evaluation subjects comprises: determining the score offset entropy of the a kind of evaluation factor scores of each evaluation subject for each fault according to the following formula:
[0007]
[0008] Where i = 1, 2, ..., n; j = 1, 2, ..., m; k = 1, 2, ..., a; K represents the entropy constant coefficient; s ijk It represents the score of the kth evaluation factor of the jth fault by the i-th evaluation subject; It represents the score deviation entropy of the kth evaluation factor score of the jth fault by the i-th evaluation subject.
[0009] In some embodiments, determining the scoring weight of each evaluation subject for each fault according to the scoring offset entropy of the a type of evaluation factor scored by each evaluation subject for each fault includes: determining the scoring weight of each evaluation subject for each fault according to the following formula:
[0010]
[0011] Among them, w ij It represents the scoring weight of the i-th evaluation subject on the j-th fault.
[0012] In some embodiments, determining the total evaluation factor score of each fault by each evaluation subject based on the a evaluation factor scores of the m faults by the n evaluation subjects includes: determining the factor coefficient of the a evaluation factor scores of each fault based on the a evaluation factor scores of the m faults by the n evaluation subjects; determining the total evaluation factor score of each fault by each evaluation subject based on the a evaluation factor scores of each fault by each evaluation subject and the factor coefficient of the a evaluation factor scores of each fault.
[0013] In some embodiments, the a evaluation factor scores include a first evaluation factor score, a second evaluation factor score, and a third evaluation factor score, and determining the factor coefficient of the a evaluation factor score of each fault according to the a evaluation factor scores of the m faults by the n evaluation subjects includes: determining the factor coefficient of the first evaluation factor score of each fault according to the following formula:
[0014]
[0015] The factor coefficient of the second evaluation factor score of each fault is determined according to the following formula:
[0016]
[0017] The factor coefficient of the third evaluation factor score of each fault is determined according to the following formula:
[0018]
[0019] Among them, s ij1 represents the first evaluation factor score of the i-th evaluation subject on the j-th fault, s ij2 represents the second evaluation factor score of the i-th evaluation subject on the j-th fault, s ij3 represents the third evaluation factor score of the i-th evaluation subject on the j-th fault, max(s j1 ) represents the maximum value of the first evaluation factor score of the jth fault, min(s j1 ) represents the minimum value of the first evaluation factor score of the jth fault; max(s j2 ) represents the maximum value of the second evaluation factor score of the jth fault, min(s j2 ) represents the minimum value of the second evaluation factor score of the jth fault; max(s j3 ) represents the maximum value of the third evaluation factor score of the jth fault, min(s j3 ) represents the minimum value of the third evaluation factor score of the j-th fault.
[0020] In some embodiments, the a evaluation factor scores include a first evaluation factor score, a second evaluation factor score, and a third evaluation factor score, and determining the factor coefficient of the a evaluation factor score of each fault according to the a evaluation factor scores of the m faults by the n evaluation subjects includes: determining the factor coefficient of the first evaluation factor score of each fault according to the following formula:
[0021]
[0022] The factor coefficient of the second evaluation factor score of each fault is determined according to the following formula:
[0023]
[0024] The factor coefficient of the third evaluation factor score of each fault is determined according to the following formula:
[0025]
[0026] Among them, s ij1 represents the first evaluation factor score of the i-th evaluation subject on the j-th fault, s ij2 represents the second evaluation factor score of the i-th evaluation subject on the j-th fault, s ij3represents the third evaluation factor score of the i-th evaluation subject on the j-th fault, max(s j1 ) represents the maximum value of the first evaluation factor score of the jth fault, min(s j1 ) represents the minimum value of the first evaluation factor score of the jth fault; max(s j2 ) represents the maximum value of the second evaluation factor score of the jth fault, min(s j2 ) represents the minimum value of the second evaluation factor score of the jth fault; max(s j3 ) represents the maximum value of the third evaluation factor score of the jth fault, min(s j3 ) represents the minimum value of the third evaluation factor score of the j-th fault.
[0027] In some embodiments, determining the total evaluation factor score of each evaluation subject for each fault according to the a evaluation factor scores of each evaluation subject for each fault and the factor coefficients of the a evaluation factor scores of each fault comprises: determining the total evaluation factor score of each evaluation subject for each fault according to the following formula:
[0028] q ij =s ij1 x j +s ij2 y j +s ij3 z j or q′ ij =s ij1 x′ j +s ij2 y′ j +s ij3 z′ j ;
[0029] Among them, x j The first factor coefficient of the first evaluation factor score of the jth fault, y j The first factor coefficient of the second evaluation factor score of the jth fault, z j The first factor coefficient of the third evaluation factor score of the jth fault, q ij represents the total score of the first evaluation factor of the jth fault by the i-th evaluation subject; x′ j The second factor coefficient representing the first evaluation factor score of the jth fault, y′ j The second factor coefficient representing the second evaluation factor score of the jth fault, z′ j The second factor coefficient of the third evaluation factor score of the jth fault, q′ ij It represents the total score of the second evaluation factor of the jth fault by the i-th evaluation subject.
[0030] In some embodiments, determining the final score of each fault according to the score weights of the n evaluation subjects for each fault and the total score of the evaluation factors includes: determining the final score of each fault according to the following formula:
[0031]
[0032] Among them, p j represents the first final score of the jth fault, p′ j represents the second final score of the j-th fault.
[0033] In a second aspect, the present invention discloses a semiconductor process equipment, including a process chamber and a control system, wherein the control system includes at least one memory and at least one processor, wherein the memory stores a computer program, and the processor executes the computer program to implement a fault mode impact determination method as described in any one of the above items.
[0034] The fault mode impact determination method and semiconductor process equipment disclosed in the present invention obtain a types of scores of m faults of semiconductor process equipment from n evaluation subjects, where n, m and a are all integers greater than 1, and determine the scoring weight and total score of each evaluation subject for each fault according to the a types of scores of the m faults by the n evaluation subjects, and determine the final score of each fault according to the scoring weight and total score of each evaluation subject for each fault, and determine the mode impact of the m faults according to the final score of the m faults, so that the subjective consciousness influence of the evaluation subject can be weakened or eliminated to a certain extent by introducing the scoring weight of each evaluation subject for each fault, thereby improving the accuracy of the fault mode impact determination result of the control system of the semiconductor equipment.
[0035] Moreover, in the present invention, the final score of each fault can be determined according to the scoring weight and the total score of each evaluation subject for each fault, and then the m faults can be clearly sorted according to the final scores of the m faults, so that the faults with higher rankings can be discovered more accurately and handled with priority, so as to minimize the harm of the faults. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the background technology, the drawings required for use in the embodiments of the present invention or the background technology will be described below.
[0037] Figure 1 The figure is a flow chart of a current failure mode effect determination method.
[0038] Figure 2 The present invention discloses a flow chart of a method for determining the impact of a fault mode.
[0039] Figure 3 A schematic diagram of the structure of a semiconductor process equipment disclosed in an embodiment of the present invention. DETAILED DESCRIPTION
[0040] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0041] The pressure control system of semiconductor equipment is closely related to product quality. As a highly coupled Many Input Many Output (MIMO) system, it usually faces the following problems when a fault occurs: Failures usually do not occur alone, and which fault should be handled first when multiple faults occur. If some high-priority faults are not handled in a timely manner, it will cause equipment downtime and safety accidents, and bring corresponding economic losses. According to Pareto's law, the degree of harm caused by the top 20% of faults in the fault sequence accounts for more than 80% of all faults. Therefore, the possible faults in the system should be sorted to accurately detect and prioritize the high-ranking faults.
[0042] like Figure 1 As shown, a current failure mode effect analysis method includes:
[0043] Step 1: Combine historical data and design defect analysis to determine the fault set {PF j |j=1,2,……,m}.
[0044] Step 2: n evaluation subjects, such as experts, conduct a comprehensive risk analysis of the fault and score the three evaluation factors, namely fault severity (S1), fault frequency (S2) and fault detectability (S3). The score range of each evaluation factor is 0 to 1, and the scoring results are shown in Table 1 below, where max i () means taking the maximum value of the row of data.
[0045] Table 1
[0046]
[0047]
[0048] Step 3: The evaluation subject divides the evaluation factors into intervals. Generally, the S1 evaluation factor intervals are 0-B1, B1-B2, B2-1, the S2 evaluation factor intervals are 0-C1, C1-C2, C2-C3, C3-1, and the S3 evaluation factor intervals are 0-D1, D1-D2, D2-1.
[0049] Step 4: Prepare a fault ranking table based on the divided intervals. The fault ranking table is shown in Table 2 below:
[0050] Table 2
[0051]
[0052]
[0053] Among them, H (High) in the fault level column indicates that the fault handling priority is high, M (middle) indicates that the fault handling priority is medium, and L (Low) indicates that the fault handling priority is low. When multiple faults occur, the handling can be postponed.
[0054] Step 5: Obtain the fault level corresponding to each fault based on the final scoring result. Here is an example: Assume that the final scoring result of fault PF1 is: S1 PF1 ∈(B1, B2), S2 PF1 ∈(C2, C3), S3 PF1 ∈(D1, D2), by looking up the table, we can get that the level of fault PF1 is M.
[0055] However, whether it is the scoring process in step 2 or the fault level classification in the fault ranking table in step 4, the subjective consciousness of the evaluation subject, such as the expert, has a certain guiding role, and it may be difficult to achieve a truly objective evaluation. For example, mechanical experts pay more attention to mechanical failures, electrical experts pay more attention to electrical failures, and software experts pay more attention to software failures. In addition, the final result of the fault ranking is not specifically quantified, but only divided into three boundaries of H, M, and L. There are still many faults within the same boundary, and it is impossible to determine the processing priority of faults within the same boundary.
[0056] Based on this, the present invention discloses a fault mode impact determination scheme, which reduces or eliminates the subjective influence of the evaluation subject to a certain extent by introducing the scoring weight of each evaluation subject on each fault, and improves the accuracy of the fault mode impact determination result of the control system of the semiconductor device. In addition, according to the scoring weight and total score of each evaluation subject on each fault, the final score of each fault can be determined in the present invention, and then the m faults can be clearly sorted according to the final scores of the m faults, so that the faults with higher rankings can be more accurately discovered and processed preferentially.
[0057] As an optional implementation of the content disclosed in the present invention, the embodiment of the present invention discloses a method for determining the impact of a fault mode, such as Figure 2 As shown, the failure mode impact determination method includes:
[0058] S101: Obtain scores of a types of evaluation factors of m faults of semiconductor process equipment from n evaluation subjects, wherein n, m and a are all integers greater than 1.
[0059] First, the historical data of semiconductor process equipment and the analysis results of the design defects of semiconductor process equipment are combined in advance to determine the m faults of semiconductor process equipment. The fault set is recorded as {PF j |j=1, 2, ..., m}. The m faults may be faults of a control system of a semiconductor process equipment, and the control system may be at least one of a pressure control system, a temperature control system, and a motion control system, wherein the pressure control system is used to control the air intake and exhaust volume of a process chamber of the semiconductor process equipment, the temperature control system is used to control the temperature in the process chamber, and the motion control system is used to control the movement rate of a robot arm or a carrier in the process chamber.
[0060] Then, n evaluation subjects conduct a comprehensive risk analysis on the m faults, and score a kinds of evaluation factors in turn, and obtain various evaluation factor scores of each evaluation subject for each fault. In some embodiments, a kinds of evaluation factors may include a first evaluation factor, a second evaluation factor, and a third evaluation factor. The first evaluation factor may be the severity of the fault (S1), the second evaluation factor may be the frequency of the fault (S2), and the third evaluation factor may be the detectability of the fault (S3). The scoring range of these three evaluation factors is 0 to 1 point. Of course, the present invention is not limited to this. In other embodiments, a kinds of evaluation factors may be two evaluation factors, or four, five or even more evaluation factors, which will not be repeated here.
[0061] Based on this, the executing subject of the fault mode impact determination method disclosed in the present invention, such as the processor of the control system of the semiconductor process equipment, can obtain the a types of evaluation factor scores of n evaluation subjects for m faults of the semiconductor process equipment, and perform subsequent steps based on the a types of evaluation factor scores of the n evaluation subjects for the m faults of the semiconductor process equipment.
[0062] S102: According to the scores of a types of evaluation factors of m faults by n evaluation subjects, the scoring weight of each evaluation subject for each fault and the total score of the evaluation factors are determined.
[0063] Obtain the scores of a evaluation factors of m faults by n evaluation subjects, and determine the score weight of each evaluation subject for each fault and the total score of the evaluation factors. Among them, the evaluation subject, such as the expert set, is recorded as {evaluation subject i|i=1,2,……,n}; the evaluation factor score of each fault is recorded as {s ijk |k=1,2,…,a;i=1,2,…,n;j=1,2,…,m},s ijk It represents the score of the kth evaluation factor of the jth fault given by the i-th evaluation subject.
[0064] In order to eliminate the influence of the subjective consciousness of the evaluation subject as much as possible, it is necessary to dynamically assign weights to the evaluation subjects, that is, it is necessary to determine the scoring weights of each evaluation subject. In some embodiments, the scoring offset entropy of each evaluation subject's scoring of the a evaluation factors of m faults can be determined based on the scores of n evaluation subjects on the a evaluation factors of m faults, and then the scoring weight of each evaluation subject on each fault can be determined based on the scoring offset entropy of each evaluation subject's scoring of the a evaluation factors of each fault.
[0065] Among them, entropy is a quantitative concept. The score offset entropy can reflect the difference between the score of any evaluation subject on any evaluation factor of any fault and the scores of other evaluation subjects on the same evaluation factor of the fault. The higher the difference between the scores of the evaluation subject and other evaluation subjects on the same evaluation factor of the same fault, the greater the score offset entropy of the evaluation subject for the evaluation factor of the fault, and the smaller the score weight of the evaluation subject for the fault.
[0066] It can be understood that the higher the difference in the scores between the evaluation subject and other evaluation subjects on the same evaluation factor of the same fault, the more subjective the evaluation subject's score of the evaluation factor for the fault is. The smaller the scoring weight of the fault by the evaluation subject is, the less subjective influence of the evaluation subject's score of the evaluation factor for the fault on the final score of the fault can be reduced.
[0067] In some embodiments of the present invention, the score deviation entropy of the a-type evaluation factor scores of each evaluation subject for each fault can be determined according to the following formula:
[0068]
[0069] Where i = 1, 2, ..., n; j = 1, 2, ..., m; k = 1, 2, ..., a; K represents the entropy constant coefficient; s ijk It represents the score of the kth evaluation factor of the jth fault by the i-th evaluation subject; It represents the score deviation entropy of the kth evaluation factor score of the jth fault by the i-th evaluation subject.
[0070] On this basis, in some embodiments of the present invention, the scoring weight of each evaluation subject for each fault can be determined according to the following formula:
[0071]
[0072] Among them, w ij It represents the scoring weight of the i-th evaluation subject on the j-th fault.
[0073] Taking a = 3 as an example, the score deviation entropy of the first evaluation subject's score on the first evaluation factor of the first fault is The score deviation entropy of the first evaluation subject's score of the second evaluation factor for the first fault and the score deviation entropy of the third evaluation factor score of the first fault by the first evaluation subject As shown below:
[0074]
[0075]
[0076]
[0077] The scoring weight w of the first evaluation subject for the first fault 11 As shown below:
[0078]
[0079] Among them, s 111 It represents the score of the first evaluation factor of the first fault given by the first evaluation subject; s 112 It represents the score of the second evaluation factor of the first fault given by the first evaluation subject; s 113 It indicates the score of the third evaluation factor of the first fault given by the first evaluation subject.
[0080] In order to meet different requirements on the failure mode impact determination results, in some embodiments of the present invention, different factor coefficients are set according to different requirements, so as to determine different total evaluation factor scores according to different factor coefficients.
[0081] In some embodiments of the present invention, the factor coefficients of the a kinds of evaluation factor scores for each fault can be determined based on the a kinds of evaluation factor scores for m faults by n evaluation subjects, and then the total evaluation factor score for each fault by each evaluation subject can be determined based on the a kinds of evaluation factor scores for each fault and the factor coefficients of the a kinds of evaluation factor scores.
[0082] In some embodiments, the a evaluation factor score of each evaluation subject for each fault can be multiplied by the factor coefficient of the a evaluation factor score to obtain the total evaluation factor score of each evaluation subject for each fault. Based on this, the proportion of the corresponding evaluation factor score in the total score can be increased or decreased by setting a suitable factor coefficient, thereby meeting the requirements of the corresponding fault mode affecting the determination result.
[0083] In some embodiments, different factor coefficients can be set according to different requirements of different types of customers for the failure mode impact determination results. Among them, different types of customers can be divided into benefit-oriented customers and cost-oriented customers. The factory system of benefit-oriented customers is segmented management. Unless the degree of failure is extremely serious, they do not want to stop the equipment to affect the process of the equipment; cost-oriented customers want to stop the equipment for minor failures to avoid affecting the overall factory system and causing greater losses.
[0084] Taking the example that the a-type evaluation factor score includes the first evaluation factor score, the second evaluation factor score and the third evaluation factor score, the first evaluation factor may be the fault severity (S1), the second evaluation factor may be the fault occurrence frequency (S2), and the third evaluation factor may be the fault detectability (S3), in the benefit-oriented customer demand, the importance of each evaluation factor is S3, S2, and S1 in turn. Therefore, among the calculated factor coefficients, the increase in the factor coefficient of the S3 evaluation factor> the increase in the factor coefficient of the S2 evaluation factor> the increase in the factor coefficient of the S1 evaluation factor, so that the proportion of the S3 evaluation factor> the proportion of the S2 evaluation factor> the proportion of the S1 evaluation factor; in the cost-oriented customer demand, the importance of each evaluation factor is S1, S2, and S3 in turn. Therefore, among the calculated factor coefficients, the increase in the factor coefficient of the S1 evaluation factor> the increase in the factor coefficient of the S2 evaluation factor> the increase in the factor coefficient of the S3 evaluation factor, so that the proportion of the S1 evaluation factor> the proportion of the S2 evaluation factor> the proportion of the S3 evaluation factor. It should be noted that the factor coefficients cannot act as weights because the sum of the three is not 1.
[0085] In some embodiments of the present invention, the a-type evaluation factor score includes a first evaluation factor score, a second evaluation factor score, and a third evaluation factor score. In order to meet the needs of benefit-oriented customers, the first factor coefficient of the first evaluation factor score of each fault can be determined according to the following formula:
[0086]
[0087] The first factor coefficient of the second evaluation factor score of each fault is determined according to the following formula:
[0088]
[0089] The first factor coefficient of the third evaluation factor score of each fault is determined according to the following formula:
[0090]
[0091] Among them, s ij1 represents the first evaluation factor score of the i-th evaluation subject on the j-th fault, s ij2 represents the second evaluation factor score of the i-th evaluation subject on the j-th fault, s ij3 represents the third evaluation factor score of the i-th evaluation subject on the j-th fault, max(s j1 ) represents the maximum value of the first evaluation factor score of the jth fault, min(s j1 ) represents the minimum value of the first evaluation factor score of the jth fault; max(s j2 ) represents the maximum value of the second evaluation factor score of the jth fault, min(s j2 ) represents the minimum value of the second evaluation factor score of the jth fault; max(s j3 ) represents the maximum value of the third evaluation factor score of the jth fault, min(s j3 ) represents the minimum value of the third evaluation factor score of the jth fault. Among them, the first factor coefficient set of the first evaluation factor score is {x j |j=1,2,……,m}, the first factor coefficient set of the second evaluation factor score is {y j |j=1,2,……,m},the first factor coefficient set of the second evaluation factor score is {z j |j=1,2,……,m}.
[0092] On this basis, in some embodiments of the present invention, the total score of the evaluation factors of each fault by each evaluation subject is determined according to the following formula: ij =s ij1 x j +s ij2 y j +s ij3 z j Among them, x j The first factor coefficient of the first evaluation factor score of the jth fault, y j The first factor coefficient of the second evaluation factor score of the jth fault, z j The first factor coefficient of the third evaluation factor score of the jth fault, q ij It represents the total score of the first evaluation factor of the jth fault by the i-th evaluation subject.
[0093] In other embodiments, in order to meet the cost-oriented customer requirements, the second factor coefficient of the first evaluation factor score of each fault may be determined according to the following formula:
[0094]
[0095] The second factor coefficient of the second evaluation factor score of each fault is determined according to the following formula:
[0096]
[0097] The second factor coefficient of the third evaluation factor score of each fault is determined according to the following formula:
[0098]
[0099] Among them, the second factor coefficient set of the first evaluation factor score is {x′ j |j=1,2,……,m},the second factor coefficient set of the second evaluation factor score is {y′ j |j=1,2,……,m},the second factor coefficient set of the third evaluation factor score is {z′ j |j=1,2,……,m}.
[0100] On this basis, in some embodiments of the present invention, the total score of the evaluation factors of each fault by each evaluation subject is determined according to the following formula: ij =s ij1 x′ j +s ij2 y′ j +s ij3 z′ j Among them, x′ j The second factor coefficient representing the first evaluation factor score of the jth fault, y′ j The second factor coefficient representing the second evaluation factor score of the jth fault, z′ j The second factor coefficient of the third evaluation factor score of the jth fault, q′ ij It represents the total score of the second evaluation factor of the jth fault by the i-th evaluation subject.
[0101] It can be understood that the first factor coefficient can be a factor coefficient that meets the needs of efficiency-oriented customers, and the second factor coefficient can be a factor coefficient that meets the needs of cost-oriented customers. However, the present invention is not limited to this. In other embodiments, different types of factor coefficients can be set according to different types of customer needs, which will not be repeated here.
[0102] S103: Determine the final score of each fault according to the score weights of the n evaluation subjects for each fault and the total score of the evaluation factors.
[0103] In some embodiments of the present invention, in order to meet the needs of benefit-oriented customers, the final score of each fault is determined according to the following formula: Among them, p j represents the first final score of the jth fault, and the final score set of m faults is {p j |j=1,2,……,m}. In other embodiments, in order to meet the cost-oriented customer requirements, the final score of each fault is determined according to the following formula: Among them, p′ j represents the second final score of the jth fault, and the final score set of m faults is {p′ j |j=1,2,……,m}.
[0104] S104: Obtain a fault mode impact determination result according to the final scores of the m faults.
[0105] The final score set of m faults is {p j |j=1,2,……,m} or {p′ j |j=1,2,……,m}, the final scores of the m faults can be sorted and output as the fault mode impact determination result. The higher the final score, the higher the fault sequence and the higher the priority of fault handling.
[0106] Take m=4, n=4, and a=3 as an example. The four faults are: PF1-process exhaust gas leakage, PF2-pipeline corrosion, PF3-lack of dry cleaning function, and PF4-hardening of heating belt. The three evaluation factors are: fault severity (S1), fault frequency (S2), and fault detectability (S3). The scoring weights of the four evaluation subjects for the four faults are shown in Table 3 below:
[0107] Table 3
[0108]
[0109] The final scores of the four faults that meet the needs of benefit-oriented customers are shown in Table 4 below:
[0110] Table 4
[0111]
[0112]
[0113] The final scores of the four faults that meet the cost-oriented customer needs are shown in Table 5 below:
[0114] Table 5
[0115] Fault Final score Troubleshooting priority <![CDATA[PF1]]> 0.67 2 <![CDATA[PF2]]> 0.44 3 <![CDATA[PF3]]> 0.41 4 <![CDATA[PF4]]> 0.82 1
[0116] It can be seen that in the embodiments of the present invention, not only can the subjective influence of the evaluation subject be weakened or eliminated to a certain extent by introducing the scoring weights of each evaluation subject for each fault, thereby improving the accuracy of the fault mode impact determination results of the control system of the semiconductor device, but also a final score for each fault can be determined based on the scoring weights and total scores of each evaluation subject for each fault, rather than separately determining the final scores of each evaluation factor of each fault, thereby more clearly sorting multiple faults, and more accurately discovering and giving priority to faults with higher rankings. In addition, appropriate factor coefficients can be set according to different needs so that the fault mode impact determination results meet different needs.
[0117] As another optional implementation of the contents disclosed in the present invention, an embodiment of the present invention further discloses a semiconductor process equipment, which includes but is not limited to a vertical furnace equipment and the like.
[0118] In some optional embodiments, such as Figure 3 As shown, the vertical furnace equipment includes a process chamber 1 and a control system (not shown in the figure). Among them, a wafer boat 10 can be arranged in the process chamber 1, and a wafer 11 can be placed in the wafer boat 10. The wafer 11 undergoes oxidation, annealing, low-pressure chemical deposition and other processes in the process chamber 1. Among them, the control system may include a pressure control system, a temperature control system and a motion control system. The control system may include a processor and a memory, and a computer program is stored in the memory. When the computer program is executed by the processor, the fault mode impact determination method disclosed in any of the above embodiments is implemented.
[0119] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0120] The above embodiments only express several implementation methods of this specification, and the descriptions are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of this specification, which all belong to the protection scope of this specification. Therefore, the protection scope of the patent of this specification shall be based on the attached claims.
Claims
1. A method for determining the impact of a failure mode, characterized in that: Applied to semiconductor process equipment, the method comprises: Obtaining a types of evaluation factor scores of m faults of the semiconductor process equipment from n evaluation subjects, where n, m and a are all integers greater than 1; According to the scores of the a evaluation factors of the m faults by the n evaluation subjects, determining the score weight of each evaluation subject for each fault and the total score of the evaluation factors; Determine the final score of each fault based on the score weights of n evaluation subjects for each fault and the total score of the evaluation factors; A failure mode impact determination result is obtained according to the final scores of the m faults.
2. The method according to claim 1, characterized in that The step of determining the scoring weight of each evaluation subject for each fault according to the scores of the a evaluation factors of the m faults by the n evaluation subjects comprises: According to the scores of the a evaluation factors of the m faults by the n evaluation subjects, determining the score offset entropy of the a evaluation factor scores of each evaluation subject for each fault; According to the score deviation entropy of a kind of evaluation factor score of each evaluation subject on each fault, the score weight of each evaluation subject on each fault is determined.
3. The method according to claim 2, characterized in that The step of determining the score offset entropy of the a kind of evaluation factor scores of each evaluation subject for each fault according to the a kind of evaluation factor scores of the m faults by the n evaluation subjects comprises: The score deviation entropy of each evaluation subject's score of a type of evaluation factor for each fault is determined according to the following formula: Where i = 1, 2, ..., n; j = 1, 2, ..., m; k = 1, 2, ..., a; K represents the entropy constant coefficient; s ijk It represents the score of the kth evaluation factor of the jth fault by the i-th evaluation subject; It represents the score deviation entropy of the kth evaluation factor score of the jth fault by the i-th evaluation subject.
4. The method according to claim 3, characterized in that Determining the scoring weight of each evaluation subject for each fault according to the scoring offset entropy of a type of evaluation factor scored by each evaluation subject for each fault includes: The scoring weight of each evaluation subject for each fault is determined according to the following formula: Among them, w ij It represents the scoring weight of the i-th evaluation subject on the j-th fault.
5. The method according to claim 1, characterized in that Determining the total score of the evaluation factors of each fault by each evaluation subject according to the scores of the a evaluation factors of the m faults by the n evaluation subjects comprises: Determine the factor coefficient of the a evaluation factor score of each fault according to the a evaluation factor scores of the m faults by the n evaluation subjects; According to the a kinds of evaluation factor scores of each evaluation subject for each fault and the factor coefficients of the a kinds of evaluation factor scores of each fault, the total evaluation factor scores of each evaluation subject for each fault are determined.
6. The method according to claim 5, characterized in that The a evaluation factor scores include a first evaluation factor score, a second evaluation factor score and a third evaluation factor score. The factor coefficients of the a evaluation factor scores of each fault are determined according to the a evaluation factor scores of the m faults by the n evaluation subjects, including: The factor coefficient of the first evaluation factor score of each fault is determined according to the following formula: The factor coefficient of the second evaluation factor score of each fault is determined according to the following formula: The factor coefficient of the third evaluation factor score of each fault is determined according to the following formula: Among them, s ij1 represents the first evaluation factor score of the i-th evaluation subject on the j-th fault, s ij2 represents the second evaluation factor score of the i-th evaluation subject on the j-th fault, s ij3 represents the third evaluation factor score of the i-th evaluation subject on the j-th fault, max(s j1 ) represents the maximum value of the first evaluation factor score of the jth fault, min(s j1 ) represents the minimum value of the first evaluation factor score of the jth fault; max(s j2 ) represents the maximum value of the second evaluation factor score of the jth fault, min(s j2 ) represents the minimum value of the second evaluation factor score of the jth fault; max(s j3 ) represents the maximum value of the third evaluation factor score of the jth fault, min(s j3 ) represents the minimum value of the third evaluation factor score of the j-th fault.
7. The method according to claim 5, characterized in that The a evaluation factor scores include a first evaluation factor score, a second evaluation factor score and a third evaluation factor score. The factor coefficients of the a evaluation factor scores of each fault are determined according to the a evaluation factor scores of the m faults by the n evaluation subjects, including: The factor coefficient of the first evaluation factor score of each fault is determined according to the following formula: The factor coefficient of the second evaluation factor score of each fault is determined according to the following formula: The factor coefficient of the third evaluation factor score of each fault is determined according to the following formula: Among them, j1 represents the first evaluation factor score of the i-th evaluation subject on the j-th fault, si j2 represents the second evaluation factor score of the i-th evaluation subject on the j-th fault, si j3 represents the third evaluation factor score of the i-th evaluation subject on the j-th fault, max(s j1 ) represents the maximum value of the first evaluation factor score of the jth fault, min(s j1 ) represents the minimum value of the first evaluation factor score of the jth fault; max(s j2 ) represents the maximum value of the second evaluation factor score of the jth fault, min(s j2 ) represents the minimum value of the second evaluation factor score of the jth fault; max(s j3 ) represents the maximum value of the third evaluation factor score of the jth fault, min(s j3 ) represents the minimum value of the third evaluation factor score of the j-th fault.
8. The method according to claim 6 or 7, characterized in that: Determining the total evaluation factor score of each fault by each evaluation subject according to the a evaluation factor scores of each fault by each evaluation subject and the factor coefficients of the a evaluation factor scores of each fault comprises: The total score of the evaluation factors of each evaluation subject for each fault is determined according to the following formula: q ij = s ij1 x j + s ij2 y j + s ij3 z j or q' ij = s ij1 x' j + s ij2 y' j + s ij3 z' j ; Among them, x j The first factor coefficient of the first evaluation factor score of the jth fault, y j The first factor coefficient of the second evaluation factor score of the jth fault, z j The first factor coefficient of the third evaluation factor score of the jth fault, q ij represents the total score of the first evaluation factor of the jth fault by the i-th evaluation subject; x′ j The second factor coefficient representing the first evaluation factor score of the jth fault, y′ j The second factor coefficient representing the second evaluation factor score of the jth fault, z j ′ represents the second factor coefficient of the third evaluation factor score of the jth fault, q′ ij It represents the total score of the second evaluation factor of the jth fault by the i-th evaluation subject.
9. The method according to claim 8, characterized in that Determining the final score of each fault based on the score weights of each fault and the total score of the evaluation factors by the n evaluation subjects includes: The final score for each fault is determined according to the following formula: or Among them, p j represents the first final score of the jth fault, p j ′ represents the second final score of the jth fault.
10. A semiconductor process equipment, comprising a process chamber and a control system, characterized in that: The control system includes at least one memory and at least one processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the fault mode impact determination method according to any one of claims 1 to 9.