Relex Weibull
	The Relex Weibull reliability model enables you to improve future product reliability by analyzing historical failure data trends. Once the failure data points and analysis parameters are entered, it can determine the failure distribution that best fits the data, and therefore best serves to forecast future failures. This data can also be used for maintenance planning, cost-effective replacement strategies, spare parts forecasting, and warranty analysis. The Relex Weibull reliability model is an integral part of the Relex Reliability Studio and can be used in combination with the Relex RBD and Relex Reliability Prediction modules. Its intuitive interface and failure data graphing capabilities make it quite useful for estimating parameters and plotting failure data and confidence intervals.

Weibull analysis can be used to study a variety of fields, practices, and disciplines. It can employ several different failure distributions, depending upon the specific situation. For example, the Weibull distribution is one of the most widely used distributions for failure data analysis. It is useful for mechanical, chemical, electrical, electronic, materials, and human failure analysis. The Weibull distribution can analyze the data from burn-in (infant mortality), useful life, and wear-out periods - meaning that it is effective in increasing, constant, and decreasing failure rate situations.

Some of the questions that Weibull analysis can answer include:

• What type of failure mechanism is the root cause?
• How many failures are expected?
• How reliable is the existing part compared to a possible new design?
• When should I replace an existing part with a new one to minimize maintenance costs?

Weibull analyses study the relationship between product reliability and product lifespan. They provide insight into the decrease in reliability as the usage of a product or system increases. The primary advantage of Weibull analysis is that it can provide reasonably accurate failure analyses and failure forecasts with extremely small data samples. This facilitates cost-effective and efficient component testing.