Automated material process control method with multivariate analysis for manufacturing industry

Abstract

A method and a system for ensuring product quality in a process for producing a product from a material in a manufacturing plant via a software performed by a computer comprising the following steps of Establishing a data pipeline including a data transformation originated from different source systems, like an ERP, LIMS, or Data Historian, to the computer, wherein the data pipeline provides updated datasets to maintain up-to-date parameter information; Preparing the datasets for a multivariate analysis calculation and performing the multivariate analysis calculation via the software, wherein all relevant parameters, like quality parameter, process parameter, sub-suppliers' quality parameters, are included in the calculation; Visualizing the result data of the multivariate analysis calculation as a 2D plot on a display connected to the computer, wherein the software identifies and marks new production batches and reference batches, time series plots for the relevant parameters and principal components, and selected top parameters which contribute to a principal components' deviation from the new production batch; and Performing quality processes and workflows to improve the production process and product quality by using the visualized result data.

Description

[0001] Automated Material Process Control Method with Multivariate Analysis for Manufacturing Industry

[0002] The disclosed invention relates to a method and a system for ensuring material and finally product quality in a process for producing a product from one or more materials.

[0003] Technical Field

[0004] The invention belongs to the technical fields of process automation, quality control, and supply chain optimization.

[0005] Background

[0006] Modern production processes are very complex matters which are influenced by many variables related to raw material inputs, equipment and tooling, human interactions, and so on. Even the slightest variations can lead to significant quality issues in the finished product, making them inferior or even unsalable. This may result in unacceptable deficiencies for the producer and / or customer, especially in highly regulated industries such as chemical or pharmaceutical.

[0007] A specific problem in this field is furthermore that the unwanted variation in finished goods quality is driven primarily by raw material inputs rather than only process variables. Raw material quality is highly variable, and suboptimal lot selections for product batches can lead to failing customer acceptance limits and a need to scrap or rework the produced batch.

[0008] Another specific point regarding this field is that the feature size of the circuit components has been fast decreased while the number of metal layers has been fast increased resulting in device topographies to exhibit features that inhibited conformal deposition. The need for the global surface planarization of the various thin films layers that constitute the integrated circuit (IC) has tremendously increased.

[0009] Furthermore in particular the high-tech electronics industry, like semiconductor manufacturing, flexible display, high-resolution LED etc, has dramatically advanced over the past 20 years; a development driven by two key factors that led the technology improvement:

[0010] The first major factor is the discovery of new applications of existing materials. One example is hafnium chloride (HfCI-4) for high-k metal gate applications being adopted in 20nm / 10nm logic devices in the semiconductor industry.

[0011] The second major factor is the enhancement of raw material quality which is directly relevant to manufacturing yield as the critical economic and technological KPI (Key Performance Index) in semiconductor mass production. Leading edge semiconductor device makers achieved a 130nm manufacturing process in the 2000’s. At this stage measuring finished good quality parameters at the end of the material production process was sufficient to prevent material-caused defects in the semiconductor manufacturing process. Thus, 99.999% purity products were produced and supplied to these industries.

[0012] As semiconductor fabrication technology advanced to 30nm and below in the 2010s, material suppliers (hereinafter ‘supplier’) have supported semiconductor manufacturing yield by supplying higher purity and adding more parameters in product specifications.

[0013] In the 2020s, innovative semiconductor manufacturing companies have been advancing their technology beyond 5nm, and ultra-high material quality has become an essential factor in achieving the commercialization of technological innovation. Beyond the quality parameters at the end of the product process, parameters related to the manufacturing process (hereinafter ‘process parameter’) and sub-suppliers' material parameters were also considered important. As a result, the number of parameters to monitor in the material manufacturing process was exponentially increased, and the material suppliers faced the challenge of monitoring more than 1 ,000 parameters in every batch production.

[0014] As suppliers support the technological advancement in the semiconductor industry by managing more than 1 ,000 parameters including process parameters, several problems arise. These are for instance:

[0015] 1. It is prohibitively time-consuming to monitor thousands of parameters to identify suspected batches for further investigation, as this requires an in- depth analysis of the individual time-series chart for each parameter.

[0016] 2. Monitoring the adherence of a large number of parameters to univariate statistical bounds inflates the probability of false alarms, where a batch is flagged as out of control due to natural statistical fluctuations rather than because of a genuine underlying issue. Such false alarms are timeconsuming to investigate and can undermine confidence in the reliability of the monitoring solution.

[0017] 3. Interaction effects between two or more parameters, which may significantly affect material quality and impact device yield cannot be discovered through univariate analyses. Interaction effects represent the combined impact of multiple relevant factors on the dependent measure, such as material related defect rates. When an interaction effect is present, the impact of one factor depends on the level of the other factor. As the number of parameters increases, the number of possible interaction effects increases super-linearly, which makes this issue especially challenging in modern quality control scenarios. A higher level of attention and domain knowledge of all parameter trends, deviations, and potential interaction effects is required to ensure quality but the available resources cannot support this requirement.

[0018] 4. Manual checking all QC process is extremely time-consuming and its automation is necessary.

[0019] To handle these issues and requirements there are in general many ways known in the state of the art to improve the production process and ensure the quality of the fabricated products. Nowadays, most of these methods are data-driven, where the relevant parameters are monitored and controlled based on data collected from the production process.

[0020] One of those approaches is known from the International Patent Application WO 2022 / 194995 A1 which discloses a method for developing or improving a process for producing a product from a material comprising steps of acquiring raw material data from at least two different sources for the production process and its relevant parameters by using a Data Collecting computer; using the acquired raw material data related to the production process to perform a Process Mapping step by using a Process Mapping computer; assigning the acquired raw material data related to the relevant parameters of the production process to its corresponding process parts by performing a Data Mapping step by using a Data Mapping computer; analyzing the therefore mapped process description with a specific software performed on an Analyzing computer thereby identifying and validating one or more existing characteristics related to the quality or performance of the production process; and using the identified and validated characteristics to develop the production process or improve its performance.

[0021] Another know prior art is the International Patent Application WO 2023 / 2 12132 A1 which discloses a method to improve bio- / chemical and semiconductor production processes performed on a production site the following steps of Gathering / uploading and preparing data about the production process in form of process variables by the at least one computer; Optimizing and scaling, depending on the process variables, a suitable Machine Learning Model by the at least one computer; Creating statistical univariate and / or multivariate profiles of a process operation of the production site via the optimized Machine Learning Model by the at least one computer; Monitoring at least one online process variable of the production process via a digital dashboard provided by the at least one computer by applying the created statistical univariate and / or multivariate profiles to it to detect anomalies, disruptions and / or malfunctions of the production process; and Adapting the production process by either correcting the detected anomalies, disruptions and / or malfunctions and / or indicating the user to interrupt the production process if required quality standards are breached or comprised.

[0022] These approaches are unfortunately either too general and not adapted to the mentioned specialized quality issues in the semiconductor industry or restricted to their specific Use Cases and therefore not applicable here.

[0023] As mentioned there are many ways known in the state of the art to supervise the whole production process. As most of them are data-driven the relevant parameters and variables of the production process are monitored and checked regularly if they do not meet their target values.

[0024] To comply with and enhance such state of the art it would therefore be desirable to find a new approach to operate automatic production management systems which further enhances the production process regarding quality and reliability, and which can specifically handle the multiple and complex process parameters setting efficiently.

[0025] Summary of the invention This task can be solved by a method for ensuring product quality in a process for producing a product from a material in a manufacturing plant via a software performed by a computer comprising the following steps of Establishing a data pipeline including a data transformation originated from different source systems, like an ERP, LIMS, or Data Historian, to the computer, wherein the data pipeline provides updated datasets to maintain up-to-date parameter information; Preparing the datasets for a multivariate analysis calculation and performing the multivariate analysis calculation via the software, wherein all relevant parameters, like quality parameter, process parameter, sub-suppliers’ quality parameters, are included in the calculation; Visualizing the result data of the multivariate analysis calculation as a 2D plot on a display connected to the computer, wherein the software identifies and marks new production batches and reference batches, time series plots for the relevant parameters and principal components, and selected top parameters which contribute to a principal components’ deviation from the new production batch; and Performing quality processes and workflows to improve the production process and product quality by using the visualized result data. From a user standpoint, the user can easily figure out whether a new batch deviates from previous reference batches from looking at the visualized the 2D plot. If the new batch deviates by location outside a sigma boundary, the user can check the time series plots to identify any abnormal trends. The user can select top parameters which contribute to the major deviation among numerous parameters (i.e. , 1 ,000), and the user can investigate these selected parameters further. The core of the invention is in general an automated workflow including all types of data, like raw material, process, and finished and intermediate good quality, from each batch, transformation, combination, execution of multivariate analysis, visualization of the analysis, and integration into existing quality processes and workflows to ensure and control the quality of each new batch produced in a timeefficient manner. This invention allows users to focus on a selected number of top contributing parameters rather than spending significant time and effort analyzing all parameters and investigating every abnormal trend. Moreover, the invention enables users to analyze and investigate the relevant process parameters in a very efficient manner.

[0026] Advantageous and therefore preferred further developments of this invention emerge from the associated sub claims and from the description and the associated drawings.

[0027] One of those preferred further developments of the disclosed method comprise that a Principal Component Analysis (PCA) is used as multivariate analysis calculation. PCA is a method of multivariate analysis.

[0028] It is a linear dimensionality reduction technique that works by data transformation into new coordinates such that the largest variation in the data can be easily identified. Therefore, it enables users to identify interaction effects not captured by univariate analysis.

[0029] Another one of those preferred further developments of the disclosed method comprise that for calculating the PCA boundaries of the reference batches the software uses statistically sufficient historical data for each relevant parameter with at least one year of data to capture potential seasonality effects. This allows to identify long-term trends, ensuring comprehensive quality benchmarks.

[0030] Another one of those preferred further developments of the disclosed method comprise that for performing the quality processes and workflows to improve the production process and product quality statistical process control (SPC) programs are used wherein a multivariate SPC is integrated into the used SPC programs. Most SPC programs review univariate control charts to identify trends and outlier batches against control limits and specifications. Including Multivariate SPC will enhance SPC programs by identifying multivariate trends and batches which are out of control to the set sigma level in multivariate space. These abnormal conditions can be missed through univariate SPC programs.

[0031] Another one of those preferred further developments of the disclosed method comprise that performing the quality processes and workflows to improve the production process and product quality a discrepant review board or material review board process is performed with integrating the multivariate analysis calculation wherein an auto notification of new batches with multivariate analysis violations triggers the discrepant review board process where a batch is placed on hold until it can be investigated and dispositions made. The included plots and trends enable the user to investigate root causes of the multivariate out of control condition.

[0032] Another one of those preferred further developments of the disclosed method comprise that performing the quality processes and workflows to improve the production process and product quality a risk assessment for changes is performed and the quality of the batch after the change is compared to the process of recorded batches to ensure no change or deviations in quality have occurred. Interaction effects could be missed if only univariate control charts are reviewed. Including the tool’s multivariate analysis of changed batch(es) vs process of record batches will strengthen the risk assessment.

[0033] Another one of those preferred further developments of the disclosed method comprise that performing the quality processes and workflows to improve the production process and product quality a supporting failure analysis and complaint investigations is done for suspected material quality variation. If a customer reports a performance issue on a material batch, the multivariate analysis with the suspected batch as the focal batch can help the investigation team confirm if the batch was abnormal compared to historic batches which had no performance issue through the tool’s multivariate analysis. Parameters contributing to a batch’s outlier condition can be quickly identified for further investigation of root cause. It is furthermore understood that all those four mentioned specific embodiments for the quality processes and workflows can be used interchangeably, together or only some of them - whatever is most suitable for the user.

[0034] Another one of those preferred further developments of the disclosed method comprise that a used manufacturing plant production plan, manufacturing process control systems, and laboratory database are maintained within a closed network structure to ensure data security and integrity. That increases data security and thereby helps to protect sensitive production and quality information.

[0035] Another one of those preferred further developments of the disclosed method comprise that the multivariate analysis algorithm is automatically initiated once data entry for a new batch is completed and every new data point is added to the previously aggregated dataset, facilitating timely analysis and quality assurance of each batch. It ensures immediate analysis of new batches and thereby facilitating proactive quality assurance.

[0036] Another one of those preferred further developments of the disclosed method comprise that the results of the multivariate analysis calculation are added to databases dedicated for multivariate analysis, thereby enhancing the data's utility for ongoing and future quality assessments. It further provides an up-to-date foundation for analysis and decision-making, improving ongoing and future quality assessments.

[0037] Another one of those preferred further developments of the disclosed method comprise that the database for the visualization of the multivariate analysis results is periodically renewed, ensuring that the visualization database reflects the most recent data and analysis outcomes. That guarantees that visualizations represent the latest data, aiding in accurate quality evaluation.

[0038] Another one of those preferred further developments of the disclosed method comprise that the software provides a user interface for the software shown on the display, in particular in form of an application website, that allows users to specify a batch number and material number, with each combination being unique and automatically assigned, to retrieve and display relevant multivariate analysis results. This allows and enhances user engagement and simplifies the process of accessing relevant quality information.

[0039] Another one of those preferred further developments of the disclosed method comprise that the software allows users to input a different batch number to retrieve and visualize the multivariate analysis results for that specific batch, enabling comparative quality assessments across multiple batches. This enables an effective comparative analysis, which can inform quality improvements across different production runs.

[0040] Another one of those preferred further developments of the disclosed method comprise that for the visualization the software is configured to show on the display plots based on the multivariate analysis result data, such as 2D scatter plots and time series charts for each principal component, providing visual insights into the quality parameters of the production batches. It thereby provides intuitive visual insights into the data, making it easier to identify and understand quality parameters and deviations.

[0041] Another one of those preferred further developments of the disclosed method comprise that users can request and view time series plots for the top parameters contributing to principal components deviation by using a designated button on the display, thereby facilitating detailed analysis of parameter trends over time. This enhances the detail and depth of analysis available to users, aiding in pinpointing specific areas for quality improvement.

[0042] A further component of the claimed invention is a system for ensuring product quality in a manufacturing plant comprising a software, performed on a computer, a closed network structure integrating manufacturing plant production plans, manufacturing process control systems, at least one display and databases to secure data integrity and confidentiality, configured to perform the following method steps of Establishing a data pipeline including a data transformation originated from different source systems, like an ERP, LIMS, or Data Historian, to the computer, wherein the data pipeline provides updated datasets to maintain up-to-date parameter information; Preparing the datasets for a multivariate analysis calculation and performing the multivariate analysis calculation via the software, wherein all relevant parameters, like quality parameter, process parameter, sub-suppliers’ quality parameters, are included in the calculation; Visualizing the result data of the multivariate analysis calculation as a 2D plot on a display connected to the computer, wherein the software identifies and marks new production batches and reference batches, time series plots for the relevant parameters and principal components, and selected top parameters which contribute to a principal components’ deviation from the new production batch; and Performing quality processes and workflows to improve the production process and product quality by using the visualized result data.

[0043] Only requirement for this computer program to perform the whole method as described is, that the used program and its respective hardware components are able to perform the method completely and automatically. Such a program can then be stored on a Computer-readable storage medium and / or data carrier signal which cause the involved computers to carry out the method steps of Establishing a data pipeline including a data transformation originated from different source systems, like an ERP, LIMS, or Data Historian, to a computer, wherein the data pipeline provides updated datasets to maintain up-to-date parameter information; Preparing the datasets for a multivariate analysis calculation and performing the multivariate analysis calculation via the computer program, wherein all relevant parameters, like quality parameter, process parameter, subsuppliers’ quality parameters, are included in the calculation; Visualizing the result data of the multivariate analysis calculation as a 2D plot on a display connected to the computer, wherein the computer program identifies and marks new production batches and reference batches, time series plots for the relevant parameters and principal components, and selected top parameters which contribute to a principal components’ deviation from the new production batch; and Performing quality processes and workflows to improve the production process and product quality by using the visualized result data.

[0044] The storage medium can be any suitable digital memory like a usb drive, a harddisk, a flashdrive and so on. From that memory it can also be provided via remote communication means using respective data carrier signals, like ethernet, wired or wireless, or any other suitable network transmission means, for transmitting the software to its target hardware.

[0045] Detailed description

[0046] The system, method and software product according to the invention and functionally advantageous developments of those are described in more detail below with reference to the associated drawings using a preferred exemplary embodiment. In the drawings, elements that correspond to one another are provided with the same reference numerals.

[0047] The drawings show: Figure 1 : a general overview about the workflow and its method steps

[0048] Figure 2: an conceptual user interface of the application

[0049] The invention will be explained in more detail by presenting one preferred exemplary embodiment which disclose respective ways for a proactive quality control system for materials based on data like historical performance, customer factors, attributes, and material processing, raw material, and intermediate factors and attributes.

[0050] The used hardware infrastructure for the embodiment is exemplary for this preferred embodiment. It can change for an other embodiments. Especially the kind of involved computers can differ greatly, depending on how much of the steps is performed by human users with the help of computers and application software or done automatically by specific computers using for instance Al based software.

[0051] The general manufacturing process using an Enterprise Resource Planning (ERP) platform is now divided into four steps and in this preferred embodiment of the invention, the previously described main four invention method steps as shown in Figure 1 are added in two main process steps after the completion of step 4.

[0052] The general manufacturing process (step 1 to step 4) are the following:

[0053] 1 . A production order is placed in an ERP system.

[0054] 2. The batch is produced from raw materials. During production, process parameters are recorded in a data management system.

[0055] 3. Samples are taken for measuring quality values on the finished batch.

[0056] 4. Quality values are measured and uploaded to database systems. Three different systems in 1 ), 2), and 4) are synced with the data pipeline. As a result, process parameters, quality parameters, and raw materials’ quality parameters with respect to every batch are updated. The new steps now start as follows:

[0057] 5. The multivariate analysis is completed, and the result is shown as type of a) 2D scatter plot, b) Time-Series chart per each principal component, c) top contributing parameter. Figure 2 shows an example of a conceptual user interface displayed on of the application

[0058] 6. Quality engineers, quality control, quality assurance, engineers, operations, or process owners (users) can utilize the result for quality assurance. First, they can proactively check whether a new production batch is located within a certain boundary or not from the PCA 2D plot or they can be automatically notified for any sigma boundary violations of new batches. They can also refer to time-series charts to check if there is any increasing or decreasing trend. Second, they can check top parameters contributing to principal components. With these top parameters, they can manage their resources efficiently. Several scenarios are possible: a. If a batch is out of the sigma boundaries, the discrepant review board (material review board) process is initiated for further investigation and disposition. If the result turns out some of process parameters are ranked top parameters, process engineers will investigate if any abnormal process steps happened. b. If some raw materials’ quality (sub-suppliers' product quality) values hold the top rank, the incoming raw material quality engineer or supplier quality will investigate further. c. If some of the container parameters are included in the top parameters, container engineers and operations will work together to find any causes. d. Results of the investigations are shared with DRB / MRB to decide on batch disposition.

Claims

Claims1. A method for ensuring product quality in a process for producing a product from a material in a manufacturing plant via a software performed by a computer comprising the following steps:• Establishing a data pipeline including a data transformation originated from different source systems, like an ERP, LIMS, or Data Historian, to the computer, wherein the data pipeline provides updated datasets to maintain up-to-date parameter information;• Preparing the datasets for a multivariate analysis calculation and performing the multivariate analysis calculation via the software, wherein all relevant parameters, like quality parameter, process parameter, sub-suppliers’ quality parameters, are included in the calculation;• Visualizing the result data of the multivariate analysis calculation as a 2D plot on a display connected to the computer, wherein the software identifies and marks new production batches and reference batches, time series plots for the relevant parameters and principal components, and selected top parameters which contribute to a principal components’ deviation from the new production batch;• Performing quality processes and workflows to improve the production process and product quality by using the visualized result data.

2. The method of claim 1 , wherein a Principal Component Analysis (PCA) is used as multivariate analysis calculation.

3. The method of claim 2, wherein for calculating the PCA boundaries of the reference batches the software uses statistically sufficient historical data for each relevant parameter with at least one year of data to capture potential seasonality effects.

4. The method of claim 1 , wherein for performing the quality processes and workflows to improve the production process and product quality statistical process control (SPC) programs are used wherein a multivariate SPC is integrated into the used SPC programs.

5. The method of claim 1 , wherein for performing the quality processes and workflows to improve the production process and product quality a discrepant review board or material review board process is performed with integrating the multivariate analysis calculation wherein an auto notification of new batches with multivariate analysis violations triggers the discrepant review board process where a batch is placed on hold until it can be investigated and dispositions made.

6. The method of claim 1 , wherein for performing the quality processes and workflows to improve the production process and product quality a risk assessment for changes is performed and the quality of the batch after the change is compared to the process of recorded batches to ensure no change or deviations in quality have occurred.

7. The method of claim 1 , wherein for performing the quality processes and workflows to improve the production process and product quality a supporting failure analysis and complaint investigations is done for suspected material quality variation.

8. The method of claim 1 , wherein a used manufacturing plant production plan, manufacturing process control systems, and laboratory databaseare maintained within a closed network structure to ensure data security and integrity.

9. The method of claim 1 , wherein the multivariate analysis algorithm is automatically initiated once data entry for a new batch is completed and every new data point is added to the previously aggregated dataset, facilitating timely analysis and quality assurance of each batch.

10. The method of claim 1 , wherein the results of the multivariate analysis calculation are added to databases dedicated for multivariate analysis, thereby enhancing the data's utility for ongoing and future quality assessments.11 . The method of claim 1 , wherein the database for the visualization of the multivariate analysis results is periodically renewed, ensuring that the visualization database reflects the most recent data and analysis outcomes.

12. The method of claim 1 , wherein the software provides a user interface for the software shown on the display, in particular in form of an application website, that allows users to specify a batch number and material number, with each combination being unique and automatically assigned, to retrieve and display relevant multivariate analysis results.

13. The method of claim 12, wherein the software allows users to input a different batch number to retrieve and visualize the multivariate analysis results for that specific batch, enabling comparative quality assessments across multiple batches.

14. The method of claim 1 , wherein for the visualization the software is configured to show on the display plots based on the multivariateanalysis result data, such as 2D scatter plots and time series charts for each principal component, providing visual insights into the quality parameters of the production batches.

15. The method of claim 14, wherein users can request and view time series plots for the top parameters contributing to principal components deviation by using a designated button on the display, thereby facilitating detailed analysis of parameter trends over time.

16. A System for ensuring product quality in a manufacturing plant comprising a software, performed on a computer, a closed network structure integrating manufacturing plant production plans, manufacturing process control systems, at least one display and databases to secure data integrity and confidentiality, configured to perform the following method steps:• Establishing a data pipeline including a data transformation originated from different source systems, like an ERP, LIMS, or Data Historian, to the computer, wherein the data pipeline provides updated datasets to maintain up-to-date parameter information;• Preparing the datasets for a multivariate analysis calculation and performing the multivariate analysis calculation via the software, wherein all relevant parameters, like quality parameter, process parameter, sub-suppliers’ quality parameters, are included in the calculation;• Visualizing the result data of the multivariate analysis calculation as a 2D plot on a display connected to the computer, wherein the software identifies and marks new production batches and reference batches, time series plots for the relevant parameters and principal components, and selected top parameters which contribute to a principal components’ deviation from the new production batch;Performing quality processes and workflows to improve the production process and product quality by using the visualized result data.

17. A Computer program comprising instructions which cause the involved computer to carry out the following method steps:• Establishing a data pipeline including a data transformation originated from different source systems, like an ERP, LIMS, or Data Historian, to a computer, wherein the data pipeline provides updated datasets to maintain up-to-date parameter information;• Preparing the datasets for a multivariate analysis calculation and performing the multivariate analysis calculation via the computer program, wherein all relevant parameters, like quality parameter, process parameter, sub-suppliers’ quality parameters, are included in the calculation;• Visualizing the result data of the multivariate analysis calculation as a 2D plot on a display connected to the computer, wherein the computer program identifies and marks new production batches and reference batches, time series plots for the relevant parameters and principal components, and selected top parameters which contribute to a principal components’ deviation from the new production batch;• Performing quality processes and workflows to improve the production process and product quality by using the visualized result data.

18. A Computer-readable storage medium and / or data carrier signal having stored thereon the computer program of claim 17 which cause the involved computer to carry out the following method steps:• Establishing a data pipeline including a data transformation originated from different source systems, like an ERP, LIMS, or DataHistorian, to the computer, wherein the data pipeline provides updated datasets to maintain up-to-date parameter information;• Preparing the datasets for a multivariate analysis calculation and performing the multivariate analysis calculation via the computer program, wherein all relevant parameters, like quality parameter, process parameter, sub-suppliers’ quality parameters, are included in the calculation;• Visualizing the result data of the multivariate analysis calculation as a 2D plot on a display connected to the computer, wherein the computer program identifies and marks new production batches and reference batches, time series plots for the relevant parameters and principal components, and selected top parameters which contribute to a principal components’ deviation from the new production batch;• Performing quality processes and workflows to improve the production process and product quality by using the visualized result data.

Images