Reliability has been widely used to measure equipment performance in military and commercial industry since the early 1940s. Since then, the importance of the reliability has grown at a phenomenal rate. Now, reliability is a key equipment characteristic that has significant influence over equipment production efficiency and cost of owning and operating a piece of equipment. In addition, better reliability leads to a competitive advantage. In the semiconductor manufacturing equipment industry, reliability plays even greater role enabling semiconductor manufacturers to compete globally.
Currently, some semiconductor manufacturers also track high-level metrics, such as overall equipment efficiency (OEE) or cost of ownership (CoO), for equipments in their factories. Reliability is a key element of such metrics.
Currently, some semiconductor manufacturers also track high-level metrics, such as overall equipment efficiency (OEE) or cost of ownership (CoO), for equipments in their factories. Reliability is a key element of such metrics.
Reliability is the probability of performing intended functions continuously without a failure (stoppage or interruption) for a specified time under the stated operational conditions. Mathematically, it is written as:
R(t) = Pr[T > t]
where t, specific time of interest; T, random variable; R(t), reliability at time t; Pr[], probability of.
Factors such as operating environment, operating stress level, operating speed, operator skill level, and maintenance procedures and policies can affect the reliability of any equipment. Therefore, a given reliability level applies to a specified set of the operational conditions. If the value of any factor varies from specified operational conditions, the reliability level may differ.
Example: the reliability of a blower in a card cage, operating in its ambient environment at 60% of its rated power, will be 0.85 at 2 years after installation.
Example: the reliability of a blower in a card cage, operating in its ambient environment at 60% of its rated power, will be 0.85 at 2 years after installation.
Quantifying Reliability
The reliability metrics are various terms used to quantify the numerical value of reliability levels. In semiconductor manufacturing industry, we use neither reliability R(t) nor failure rate to measure level of equipment reliability. Instead, we use metrics based on mean life. These metrics consist of at least four words, as shown in figure. Two of them, MEAN and BETWEEN, are mandatory. Other words relate to the measures of life and events.
Using the algorithm given in figure, we can define metrics appropriate for any situation. Take the word MEAN, select a word for measure of life; take the word BETWEEN, and select the desired event. Examples
1. Mean time between failures
2. Mean cycles between unscheduled maintenance
3. Mean wafers between failures.
These metrics are widely used to track reliability of semiconductor manufacturing equipment and are recommended by the Semiconductor Equipment Manufacturing International (SEMI), an association of semiconductor manufacturing equipment suppliers, in SEMI Specification E10-0304E for definition and measurement of equipment reliability, availability, and maintainability (RAM).
1. Mean time between failures
2. Mean cycles between unscheduled maintenance
3. Mean wafers between failures.
These metrics are widely used to track reliability of semiconductor manufacturing equipment and are recommended by the Semiconductor Equipment Manufacturing International (SEMI), an association of semiconductor manufacturing equipment suppliers, in SEMI Specification E10-0304E for definition and measurement of equipment reliability, availability, and maintainability (RAM).
Build-In Reliability
Building-in reliability is a process that assures all parts, subsystems, modules that are made and assembled according to engineering drawings and specifications without degrading the designed reliability or introducing new failure modes. Important steps of this process are:
Assembly instructions. Prepare detailed instructions for each assembly step. These instructions should include proper parts, materials, step-by-step assembly procedures, tools, limitations, inspection procedures, etc.
Training. To minimize assembly errors, it is essential that every assembly operator be trained in basic assembly methods and in all the assembly operations assigned to him or her.
Burned-in. All parts and the system itself should be properly and adequately burned-in, debugged, or stress-screened before shipment.
Product reliability acceptance test (PRAT). Conduct a PRAT on randomly selected units before shipping to assure the reliability level of the product line as it is shipped.
Packaging and shipping. Equipment must be packed properly for the intended shipping mode. Select shipping mode that does not impart any undue stress on the equipment.
