Reliability is a problem that exists in every link of product design, manufacture and use. Simply put, the so-called reliability is the degree to which the product is not prone to failure. As we all know, products are usually qualified when they leave the factory, but as time goes by, the function and performance of the product will gradually change, which will eventually lead to failure. Although this trend cannot be changed, designing and manufacturing products that do not fail within a specified time is a topic that both enterprises and consumers care about. As far as the frequent communication failures of U.S. fighter jets in the early days of World War II, and as recently as the quality problems of a certain car brand’s engine that were exposed during March 15 last year, in essence, they are all caused by the product’s poor reliability. Reasonable application of reliability analysis can help technicians in R&D, engineering, quality and other departments improve the stability of product quality, reduce product life cycle costs and after-sales service costs, and improve customer satisfaction and loyalty.
What is puzzling is that many companies have realized the importance of reliability analysis, but they still deliberately avoid reliability analysis. Why? There are many reasons, one of the main reasons is that the general enterprises feel that the routine statistical quality control is already complicated enough, and the research on the reliability needs to use a lot of more advanced statistical knowledge, for those who have not been trained in formal statistical methods It would be daunting, which objectively greatly limits the promotion of reliability methods in enterprises.
JMP has a powerful and professional reliability and survival platform, which can support users to conduct user analysis and predict the reliability of single components, repairable products and even the entire system in an interactive and visual way. Let's take a typical example to see how to use JMP to carry out reliability analysis.
example: In order to analyze the reliability of an electrical product, a company collected a batch of service life data of the product (as shown in Figure 1, when "censored" =0 When means "time" is the exact failure time, when "censored" =1 When means that the exact failure time is unknown, but it must be greater than the value displayed in "Time"). Under this set of representative data, let's study the failure characteristics of the product? When the probability of failure is 90% What is the reliable lifetime of this product?
Figure 1 Reliability test raw data table (part)
According to the theory of reliability method, to solve these two problems, we need to solve a basic problem first: what distribution does this group of life data obey? This is not an easy problem to solve. You have to try, compare, and verify one by one, such as Weibull distribution, lognormal distribution, exponential distribution, etc., to say the least, there are more than a dozen. However, general quality engineers feel dizzy when they hear these professional statistical terms, and because the life data contains the "censored" feature, the judgment is more complicated, and usually requires a series of lengthy statistical analysis reports and statistical indicators to judge .
When using JMP software for analysis, we found that there is a command in JMP called "fit all distributions", which can fit all conventional reliability distributions one by one in a few seconds, and then automatically screen out the best distribution fit. For example, in the figure below, the "loglogistic" distribution is the best distribution JMP found after a quick comparison of all life distributions. If you don’t know much about statistical principles and just want to see what the distribution looks like intuitively, just look at the blue curve on the picture and the light blue confidence interval around it; if you understand statistical principles better, think For an in-depth understanding of the statistical basis for discrimination, you can also see the relevant indicators in the "Model Comparison" table below. In short, you can take what you need and complete the most basic task of distribution model identification.
Figure 2 Visual display of reliability distribution model comparison
In addition, while obtaining the best fitting distribution, various reliability characteristics related to the quality of the product (such as reliable life, failure probability, failure probability density, failure rate, etc.) can also be expressed graphically. For example, in the figure below, both the "Distribution Profiler" and "Quantile Profiler" show the relationship between failure probability and product life (the main difference between the two is that the variables represented by the X-axis and the Y-axis are just opposite), The "hazard rate profiler" shows the law that the hazard rate (that is, the so-called Hazard Rate) changes with the product life (this can be used to realize the most famous "bathtub curve" in reliability theory), " The Density Profiler shows how the failure probability density varies with product life.
Figure 3 Profiler of main characteristic variables in reliability analysis
Using these graphs, the first question "How is the failure characteristic of this product?" can be explained visually. Now let's answer the second question "When the probability of failure is 90%, what is the reliable life of this product?" Enter "0.9" on the X-axis of the "Quantile Profiler", and you can get it on the Y-axis The red value "4017.671" and the black value "[2798.93,5767.09]" indicate that the reliable life at this time should be 4017.671, and its confidence interval should be [2798.93,5767.09].
There are many other contents about reliability analysis, such as accelerated life analysis, regression analysis of life data, maintainability analysis and so on. These have a corresponding graphical intelligent analysis platform in the JMP software. A very direct benefit is that it allows engineers with non-statistical backgrounds to quickly overcome the barriers of statistics and directly conduct analysis and research on technical issues, thereby improving product reliability. Degree, reduce customer complaint rate and after-sales service costs. We will continue to bring you introductions and case applications of these platforms in the future, so stay tuned.
