China
Africa
Explore chapters and articles related to this topic
Customer-Focused Quality
Published in John Nicholas, Lean Production for Competitive Advantage, 2018
But the mean and standard deviation can be determined only if the process is stable. A stable process is one where fundamental features of the process are repetitive and unchanging. In fact, the term “stable process” implies that the mean and standard deviation of a process are constant; that is, they will remain the same until something happens to change the process. In the production of circular rods, for example, the average diameter of the rods produced will be the same today and tomorrow as it was yesterday—given that nothing happens to alter the average up or down, and the same applies to the standard deviation of the diameter. Even though the diameters of individual rods vary slightly, in a stable process the mean and standard deviation of the rods remain the same.
Quality Engineering and Methods
Published in Jong S. Lim, Quality Management in Engineering, 2019
The control chart uses the average and the standard deviation of sampling data from an established stable process. The most popular chart type is x-bar and R-Chart as discussed in the previous section. For example, the previously discussed PPAP process requires a minimum of 25 subgroups containing at least 100 readings from consecutive parts of the significant production run.25
Process Stability
Published in Gisi Philip, Sustaining a Culture of Process Control and Continuous Improvement, 2018
A stable process is one that is predictable. A process can be considered stable if the output parameters of interest are relatively consistent over time. For example, if a process produces steel rods whose average length and standard deviation are consistent and predictable with time, the process from which they are produced is considered stable. Characteristics of a stable process include:
Robust profile alignment based on penalised-spline smoothing
Published in International Journal of Production Research, 2019
In the simulation, we consider the reference profile as and the following unaligned profile: where b=0.5, u=0.2, v=4, and ω is a parameter that defines the lengths of the profiles to be aligned. In this context, ω is set as when the profiles to be aligned are longer than the reference, whereas when the profiles to be aligned are shorter than the reference. The first part in Model (14) displays the main shapes of the profiles, and the second part shows the periodic vibrations in applications. In these settings, the unaligned profiles are different from the reference surface both in amplitude and phase. The observed profiles are generated with the random errors ϵ generated i.i.d. from the normal distribution with and for a stable process and severe noises, respectively. For these sampling situations and the analysis in Section (2.4), the penalty parameter λ used in Model (7) was.
Process tracking and monitoring based on discrete jumping model
Published in Journal of Quality Technology, 2018
Statistical process control (SPC) techniques have been widely used in industry for quality control and process improvement (see Montgomery, 2007). The basic idea of most SPC methods is to first establish a probability distribution, called “null” distribution, that describes the randomness of the statistic being monitored when the process works normally. The newly collected samples will be statistically tested against the null distribution, then decisions will be made based on the testing result. The process is “in-control” if the new sample is likely from the null distribution; otherwise, the process is “out-of-control.” One notable limitation of these SPC methods is that prior knowledge regarding the process change is rarely considered in the hypothesis testing procedure for decision-making. For example, when we consider a stable process with very small out-of-control probability, the criteria of out-of-control should be different from that of a process with high variability. However, in most existing SPC techniques, this information is not considered and the decision is solely based on statistical testing results.
Plantwide control systems design and evaluation applied to biodiesel production
Published in Biofuels, 2021
Bruno Firmino da Silva, Jones Erni Schmitz, Ivan Carlos Franco, Flávio Vasconcelos da Silva
The process global overview needs to be set according to previous goal definitions. In this case, the main purpose aims at a satisfactory stable process, with commercial product quality and higher probability to obtain profits. The two PWC strategies were tested with disturbances in the oil flow rate with a variation between 5% and −5%. The control systems kept the process controlled over the disturbances applied and the biodiesel purity with low variations was always larger than 98.4%wt, which is above the commercially required purity of 96.5%wt.