Uncertainty Measurement and Analysis for Quality and Reliability in Manufacturing Process

Recent technological advancements in computing, sensing and communication have led to the development of cyber-physical manufacturing processes, where a computing subsystem monitors the manufacturing process performance in real-time by analyzing sensor data and implements the necessary control to improve the product quality. This paper develops a predictive control framework where control actions are implemented after predicting the state of the manufacturing process or product quality at a future time using process models. In a cyber-physical manufacturing process, the product quality predictions may be affected by uncertainty sources from the computing subsystem (resource and communication uncertainty), manufacturing process (input uncertainty, process variability and modeling errors), and sensors (measurement uncertainty). In addition, due to the continuous interactions between the computing subsystem and the manufacturing process, these uncertainty sources may aggregate and compound over time. In some cases, some process parameters needed for model predictions may not be precisely known and may need to be derived from real time sensor data. This paper develops a dynamic Bayesian network approach, which enables the aggregation of multiple uncertainty sources, parameter estimation and robust prediction for online control. As the number of process parameters increase, their estimation using sensor data in real-time can be computationally expensive. To facilitate real-time analysis, variance-based global sensitivity analysis is used for dimension reduction. The proposed methodology of online monitoring and control under uncertainty, and dimension reduction, are illustrated for a cyber-physical turning process.

<p xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" class="first" dir="auto" id="d13128e120">Laser powder bed fusion (LPBF), a typical metal additive manufacturing (MAM) process, generates large thermal gradients and rapid cooling rates, making it challenging to fully understand the relationship between nano-scaled solidification microstructure and process parameters. This study aims to uncover the nano-scaled solidification microstructure characteristics of 316L stainless steel (SS) produced through LPBF. A deeper understanding of microstructural development is crucial for fabricating high-performance materials by effectively controlling the AM process. <p xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" dir="auto" id="d13128e122"> <b>1. Introduction</b> <p xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" dir="auto" id="d13128e127">Laser powder bed fusion (LPBF) is currently the most versatile metal additive manufacturing (MAM) process, enabling the production of complex parts without the design limitations of traditional manufacturing methods. The rapid and repeated thermal cycles create intense thermal gradients, leading to metastable chemical, structural, and mechanical states, resulting in metallurgical defects that compromise material properties. Extensive research of LPBF-processed 316L SS has been conducted to improve the mechanical properties and highlight the need to correlate different microstructural features with the manufacturing process. Most studies have focused on characterizing microstructures from macro to sub-micrometer scales, with few investigating features at the nano or atomic levels. Therefore, this study aims to explore these nano/atomic scale microstructural features to better understand the relationship between solidification microstructure and process parameters. <p xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" dir="auto" id="d13128e129"> <b>2. Experiment</b> <p xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" dir="auto" id="d13128e134">The nominal composition of the 316L SS powder was 18Cr-14Ni-2.5Mo-0.03C (wt.%), and the powder size was under 53µm. LPBF fabrication was conducted using a 3D printer (EOSM290, EOS GmbH, Germany) equipped with a Yb-fiber laser by the “X-scan strategy,” i.e., the laser beam was scanned bidirectionally along the X-axis without rotation.

The International Organization for Standardization (ISO) Guide to the Expression of Uncertainty in Measurement (GUM) describes a generic procedure for determining an estimate for the value of the measurand and its associated combined standard uncertainty from the estimates and their associated standard uncertainties for various input quantities. A user of the ISO-GUM who is interested in understanding, managing or improving the measurement procedure needs the details, usually expressed as an uncertainty budget, on how the estimate for the value of the measurand and its associated combined standard uncertainty were calculated. In particular, a user may be interested in quantifying the degrees of contribution to the combined standard uncertainty from its components. When the measurement equation is a linear function of uncorrelated input variables, the contribution from a component is usually quantified by the product of the component of uncertainty and its sensitivity coefficient. This paper introduces a coefficient of contribution that is suitable for both uncorrelated and correlated input variables. The proposed coefficient of contribution is useful for a variety of measurement equations. Correlations between input variables can significantly alter the relative importance of the contributions to the combined standard uncertainty from its components.

Uncertainty assessment is a quantitative method about accuracy grade. This essay mainly focused on the principle of uncertainty analysis, including establish the uncertainty analysis model, assess the uncertainty components of type A and type B, and calculate the sensitivity coefficient and combined standard uncertainty. Because the result is easy to go wrong when using partial differential to solve the sensitivity coefficient, two kinds of mathematical models without using partial differential are put forward. Based on the uncertainty assessment of boiler emission flue gas heat loss, three models are used to calculate the combined standard uncertainty of boiler flue gas heat loss respectively, the results are contrasted and analyzed which certified the correctness of the two new models. In this way, a new calculation method is bought up, as well as simplifying the work of uncertainty assessment. Keywords-uncertainty; mathematical model; sensitivity coefficient; emission flue gas heat loss

: Manufacturing process models are developed for a data-driven, reconfiguarable manufacturing facility. A human assembly process model is developed to characterize assembly processes. Machine and tool resources are characterized for manufacturing processes to provide a broad characterization of manufacturing process coverage in a hypothetical final assembly factory for infantry fighting vehicles. A sample population of manufacturing process parameters and resources is collected and incorporated into a manufacturing capability and process model library. Requirements for a virtual manufacturing environment are developed, and an initial, prototype version is implemented to facilitate process model validation, configuration, and experimentation.

This paper presents an experimental characterization of grounded coplanar waveguides (GCPWs) with uncertainty analysis. GCPW lines to be analyzed were fabricated using different subtractive printed circuit board (PCB) manufacturing processes: panel plating method and pattern plating method. Using a modified multiline method, we have extracted the propagation constants with uncertainty propagation analysis. This considered instrumentation noise, length uncertainty, and impedance mismatch caused by variations in cross-section due to the fabrication capabilities of each manufacturing process. By reformulating the propagation constant, we obtained the attenuation per unit length and the effective relative permittivity as a function of frequencies up to 43.5GHz with 95% confidence interval. These results showed good agreement with full-wave simulations and highlighted the differences between the PCB manufacturing processes.

Control chart (CC) is used to monitor the special causes that arise during the process monitoring. These special causes produce continual shifts in the process parameters that last until it is identified and removed. There is a need for such techniques, which present the true representation of the entire process. Rational subgrouping is an essential concept in Statistical Process Control (SPC) which is seldom overlooked by the practitioner. Hence, most of the manufacturing, engineering, and production processes give output products in the form of batches over smaller intervals of time. The aim of this study is to provide a median based design for Tukey and Tukey-EWMA control charts under subgrouping. It will use the idea of boxplot to monitor the process behavior. This study also provides a brief discussion regarding selecting and forming subgroups from the process data. The performance of the median based Tukey and Tukey-EWMA charts are judged using Average, Median and Standard-Deviation run-length as performance measures. We have considered subgroup sizes of m=1,5 &10 at pre-specified ARL0 equal to 370. To real-life applications of the median based tukey designs are also presented to show their implementation in food manufacturing and hard-bake processes.

Non-contact thermal measurement techniques such as on-line thermography can be valuable tools for process monitoring and quality control. Many manufacturing processes such as welding or casting are thermally driven, or exhibit strong correlation between thermal conditions and product characteristics. Infrared inspection of self-emitted radiation can provide valuable insight into process parameters not routinely observed yet which dominate product quality. Recent advances in IR system technology coupled with significant reductions in cost are making thermography a viable tool for such on-line monitoring. This paper describes the implementation of a novel rugged thermal imaging system based on a dual-wavelength technique for a large intelligent process monitoring project. The object of the portion described herein is to deploy a non- contact means of monitoring tooling surface thermal conditions. The technical and practical challenges of developing such a non-contact thermal measurement system for continuous inspection in an industrial environment are discussed, and methods of resolving them are presented. These challenges include implementation of a wavelength filter system for quantitative determination of the surface temperature. Also, unlike visible-spectrum machine vision applications, surface emissivity of the test object as well as reflections from other IR emitters must be taken into account when measuring infrared radiation for a part or process. However, the primary issues that must be addressed prior to deployment are compensation for ambient temperature conditions and optimization of the calibration process. Other issues center on remote camera control, image acquisition, data synchronization, and data interpretation. An example application of this system, along with preliminary data, is described.

Mesenchymal stem/stromal cells (MSCs) have been identified as a promising therapeutic option for osteoarthritis, graft vs. host disease and cardiovascular diseases, among others. For widespread application of these therapies, robust and scaled manufacturing processes are required that reliably yield high amounts of high quality MSCs. One of the primary challenges in MSC manufacturing is achieving robustness, due to the high donor-to-donor and batch-to-batch variability seen in MSC manufacturing. To achieve more consistent manufacturing, standardization of the manufacturing process and analytical methods to determine cell quality and control process parameters will be needed. Traditionally, MSCs are cultivated in two dimensional (2D) systems, such as flasks or plates. However, these systems are limited in their scalability. To enhance volumetric productivity, upscaling may be achieved using agitated bioreactors where the MSCs are grown on microcarriers or other types of scaffolds. In this article, we have reviewed existing publications on the manufacturing of MSCs in agitated bioreactor systems regarding the process conditions used and the quality parameters measured to define more clearly the most relevant cell quality and process parameters. Key cell quality parameters measured are cell number and viability, immunophenotype and differentiation potential, while key process parameters include the cultivation system (cell source, bioreactor type, media composition), physiochemical properties of the media such as pH and dissolved oxygen (DO), as well as nutrient supply. Defining these parameters more clearly will support the development of robust MSC manufacturing processes at scale using improved process control and facilitate the widespread clinical application of MSC-based cell therapies.

The analysis and improvement of product quality was not accurate for the incomplete and uncertain of data in manufacturing process. To get the optimizable improve rules of product quality, rough theory has been adopted to analyze the manufacturing process. Firstly, the quality decision table has been presented. Then the incomplete information have been disposed based on the Similar Relation and proposed the method for getting the optimizable quality rules. Finally, a knuckle pivot manufacturing process is provided to verify the effectiveness of the proposed method.

In sheet metal forming processes, the success of the forming process depends on several primary parameters such as the geometry of the tool, Sheet Material Properties, and the Friction in the contact area (tool/workpiece interface). In addition, the process parameters like punch crosshead, deformation temperature, surface roughness of both punch head and sheet blank, lubricant state, and the value of the applied load. Friction in the actual contact area plays a significant effect in obtaining a high-quality part and reducing the production cost. The study provides a new idea to specify the suitable Coefficient of Friction (COF) used when forming Aluminium AA1060, taken according to Coulomb’s Friction Law and Stribeck Diagram for different lubricant states. The methodology of this research is based on the finite element analysis using ABAQUS/CAE 2020 of the Nakajima Formability Test for different Coefficients of Friction (COF) that are taken from Stribeck’s Diagram according to sheet material type, where the suitable Coefficient of Friction (COF) that achieves ISO 12004-2 condition to consider the sheet material successful in the test and valid for use in the manufacturing process. Results show that the Coefficient of Friction (COF) significantly affects the Formability behaviour and specifies the Location of Tearing (crack), where at higher (COF), the Formability of Aluminium AA1060 is decreased. The Tearing (crack) occurs near the circumference of the sample. Consequently, the sample was considered failed in the test and invalid to use in manufacturing processes. At the same time, when (COF) is low, the Formability of the sheet material increases, and tearing occur near the centre of the sample apex, so the sample is considered successful in the test and valid to use in the manufacturing process according to ISO 12004-2

Researchers at the University of Queensland, Australia, have developed and refined an optical feedback technique for measuring surface roughness. We spoke to one of the authors, Jeremy Herbert, to find out more. Optical feedback interferometry is a non-contact sensing technique that uses a laser as not only a coherent light source, but also as a receiver of information. The beam from the laser is reflected off a target where it is modulated by some property of the material. This light is then reinjected back into the laser cavity where it mixes with the standing wave and disturbs the normal operation of the laser. By monitoring this disturbance through either the output power or the terminal voltage of the laser, information about the target can be determined. Schematic of the roughness measurement experiment showing the direction of translation as parallel to the lasing axis The primary advantage of optical feedback is that a sensor can detect changes in optical path length on the order of nanometres, and/or minute changes in target reflectivity using compact and low cost components; in some configurations, a laser diode is the only optical component. This allows extremely sensitive measurement of quantities such as displacement, velocity and refractive index of transparent media. Another more recent application is the imaging of targets using exotic wavelengths (such as terahertz radiation) for which simple detectors are not widely available. The sensitivity of instrumentation based on optical feedback is a double-edged sword. While this type of sensor can measure minute changes in the properties of the target, it is also very sensitive to changes in environmental parameters such as temperature and vibration. A common approach to minimise the effects of these unwanted influences on the output signal is to perform repeated measurements and average them. Unfortunately in some circumstances, such as the one outlined in the article, it is not possible to perform repeated measurements, and so an alternative strategy is required. Roughness target locations where a focused translating beam would experience higher (green) and lower reflectivity (red). The black arrow shows the direction of scanning. In the article, a method of determining information about the roughness of a surface using optical feedback was described and demonstrated. The method does not rely on repeated measurement, but instead uses the windowed standard deviation of the optical feedback signal to track the effective reflectivity of the surface as it moves. A method to compute the windowed standard deviation in a real-time manner was presented in an earlier article also published in Electronics Letters (Herbert, J, Wilson, S, Rakic, A and Taimre, T, ‘FPGA implementation of a high-speed, real-time, windowed standard deviation filter’, Electronics Letters, 2016, 52, (1), pp. 22–23). This work is significant because it demonstrates a viable method for the use of optical feedback signals to measure surface roughness without necessitating repeated measurements of the target. It provides evidence for the idea that an online surface roughness sensor could be developed, with the small size and low cost advantages of laser feedback, for use in the manufacturing process of objects with specific roughness tolerances. More broadly, this approach could be applicable to other scenarios in which repeating a measurement is impossible, such as in determining information about a particle randomly passing through the beam. Possible applications of this work in the short term could be in the area of tribology, the measurement of mechanical wear, and in manufacturing processes that require tight control of surface roughness. Longer-term applications are also possible in areas where the measurement quantity of interest can be modelled as the problem of surface roughness, such as in the measurement of particle size. Since this work was completed, further efforts have been directed toward applying the methods in the article to other scenarios in which repeated measurements are not possible. Examples of these situations include the characterisation of uniquely shaped particles passing through the beam, and in the online control of manufacturing processes such as polishing. Optical feedback sensors are unusual in that they are able to measure nanometre-sized changes in optical path length, and yet can be constructed using a small number of compact and relatively inexpensive components. As manufacturers compete to have the most effective quality control and manufacturing processes, sensing techniques such as the one outlined in this article become more and more relevant. Over the coming years, further work in this area could result in low cost sensors that could improve the yield of manufacturing processes through the increased availability of information about process parameters.

Biopharma Is Going Digital … Bit by Bit

Nowadays new applications based on the 3D printing technique demand increasingly strict product quality requirements. The in-situ monitoring of variables associated with the manufacturing process through the application of different techniques could help to evaluate the process and ultimately to ensure product quality. In this regard, the acquisition and evaluation of variables and indexes derived from thermographic analysis during the process are key for an early defect detection and can contribute to quality estimation. In this work, a new methodology is proposed for the monitoring and analysis of the additive manufacturing process based on the processing of thermographic images from an LWIR (Long Wave Infrared) camera. The methodology and the suitability of the variables and indexes extracted during the monitoring of the manufacturing process are discussed for the case of a 3D fused filament fabrication of polymers.

Additive manufacturing is a promising manufacturing process with diverse applications, but ensuring the quality and reliability of the manufactured products is a key challenge. The digital twin has emerged as a technology solution to address this challenge, allowing real-time monitoring and control of the manufacturing process. This paper proposes a digital twin system framework for additive manufacturing that integrates machine learning models, employing Unity, OctoPrint, and Raspberry Pi for real-time control and monitoring. Particularly, the system utilizes machine learning models for defect detection, achieving an Average Precision (AP) score of 92%, with specific performance metrics of 91% for defected objects and 94% for non-defected objects, demonstrating high efficiency. The Unity client user interface is also developed for control and visualization, facilitating easy additive manufacturing process monitoring. This research article presents a detailed description of the proposed digital twin framework and its workflow for implementation, the machine learning models, and the Unity client user interface. It also demonstrates the effectiveness of the integrated system through case studies and experimental results. The main findings show that the proposed digital twin system met its functional requirements and effectively detects defects and provides real-time control and monitoring of the additive manufacturing process. This paper contributes to the growing field of digital twin technology and additive manufacturing, providing a promising solution for enhancing the quality and reliability in the field of additive manufacturing.