China
Africa
Explore chapters and articles related to this topic
Translating Sensor Technology into the Medical Device Environment
Published in Kevin Yallup, Krzysztof Iniewski, Technologies for Smart Sensors and Sensor Fusion, 2017
Devices that are both high risk and nonsubstantially equivalent to a predicate device will be regulated via the PMA pathway. A PMA application requires compilation and presentation of considerable amounts of information. This includes a complete description of the device and components, photos and engineering diagrams, a detailed description of the methods, facilities and controls used to manufacture the device, the proposed labeling and advertising literature, training materials, software documentation, biocompatibility information, and references to applicable standards, also a summary of all clinical, animal, and bench data. An advisory panel is often assembled to provide an external review of the device, and an inspection of manufacturing facilities will be arranged. Finally, after PMA approval, the FDA maintains market surveillance rights as the product is commercialized. The PMA process can take years and cost millions or tens-of-millions of dollars. This is a substantial hurdle for novel, untried technology, such as a new sensor system, to clear, and the process focuses attention on the market potential of the device.
Surface Profilometry
Published in Toru Yoshizawa, Handbook of Optical Metrology, 2015
Toru Yoshizawa, Toshitaka Wakayama
One more result was presented by Zhang, who modified “two plus one phase-shifting algorithm” to alleviate the error due to motion [27]. The schematic diagram of this shape measurement system is shown in Figure 19.32. Their previous work encountered the problem of fast motion such as the facial geometric changes during speaking. In this trial, two fringe images with 90° of phase-shift and a third flat image (not the average of two fringe images with a phase shift of 180°, but a computer-generated uniform flat image) are collected. Their experimental result demonstrated that this system can satisfactorily measure the dynamic geometrical changes. The data acquisition speed attained up to 60 frames/s with an image resolution of 640 × 480 pixels per frame. This system is expected to be applied to online inspection in manufacturing, medical imaging, and computer graphics.
Dimensioning
Published in Ken Morling, Stéphane Danjou, Geometric and Engineering Drawing, 2022
When using the method of parallel dimensioning, it might be possible that a drawing needs more than just one common datum for the horizontal or vertical dimensions. This can have multiple reasons which are usually related to the function of the part or related to inspection or manufacturing. As an example, parts which are machined on a lathe usually require minimum two references for the horizontal dimensions. Figure 7.8 depicts a three-jaw chuck which is used on a lathe to fix the part to be machined. First, all shoulders accessible from one side are turned while the part is fixed in the chuck. After re-chucking, the shoulders on the other side of the part are turned.
Anomaly detection for fabricated artifact by using unstructured 3D point cloud data
Published in IISE Transactions, 2023
Chengyu Tao, Juan Du, Tzyy-Shuh Chang
A variety of image-based methods have been proposed for surface quality inspection in manufacturing processes, such as edge detection based on wavelet filters (Siegel et al., 1998) and smooth sparse decomposition (Yan et al., 2017). Compared with 2D image data, 3D point cloud data has the following three advantages:Capable of capturing tiny surface anomalies, which are indistinguishable in 2D images (Ye et al., 2021) or even to human eyes (Jovančević et al., 2017).Able to provide quantitative 3D geometric information about anomalies, such as maximum depth and orientation (Jovančević et al., 2017), which are valuable in further optimization of process variables (Huang et al., 2022).Robust to lighting conditions (Jovančević et al., 2017).
In-process quality improvement: Concepts, methodologies, and applications
Published in IISE Transactions, 2023
IPQI refers to a set of methodologies of engineering-driven data fusion for process monitoring, root cause diagnosis, and feedback and feed-forward control. The data concerned in IPQI embodies those methodologies throughout the life cycle of a process and product, ranging from product design, process design, in-situ sensors, product quality measurement, and maintenance information, among others. Data fusion is achieved by developing advanced statistical and machine learning methods guided by engineering knowledge. This resulting method or decision-making is further enhanced by optimization methods and control theories (Figure 2). By implementing IPQI, one expects to achieve root cause diagnosis (in addition to change detection), online automatic control (in addition to off-line adjustment), and defect prevention (in addition to defect inspection) in manufacturing systems.
The integrated study for process improvement with economic specification limits, process means setting and quality investment: an extension of the model in Abdul-Kader et al. (2010)
Published in Quality Technology & Quantitative Management, 2020
Inspection, determination of process parameters and quality investment are some of important approaches in quality systems design. Abdul-Kader et al. (2010) developed an integrated model for total inspection with minimization of the expected total cost per item, in which the economic specification limits were first obtained for product inspection, and then, the optimal quality investment level and corresponding improved process mean and standard deviation were determined by applying the quality investment function given by Chen and Tsou (2003). In the present paper, the integrated model given in Abdul-Kader et al. (2010) is modified with specified process capability index value for obtaining the process mean, economic specification limits, and quality investment level based on minimization of the expected total cost per item. Three modified models are introduced and studied, where Model I firstly determines the optimal process mean and economic specification limits and then obtains the optimal quality investment level, Model II firstly determines the optimal process mean and quality investment level and then obtains the economic specification limits, and Model III simultaneously determines the optimal process mean, economic specification limits and the optimal quality investment level. The product is considered with asymmetric quadratic quality loss, tolerance cost, inspection cost, manufacturing cost, part and material cost, rework cost, and scrap cost. For each modified model, the objective function is described and the solution procedure is provided. A numerical example is given for illustration.