project & quality management quality management reliability

Project & Quality Management

Quality Management

Reliability

Reliability Management

Why is it needed?

• Reliable operation of critical equipment

• Planning of maintenance activities

• Improved ‘quality’ of an item

Reliability Management

Reliability management is concerned with performance and conformance over the expected life of the product

“the probability that a product or a piece of equipment performs its intended function for a stated period of time under specified operation conditions’”

Definition of Reliability

The definition has four important elements:

• Probability

• Time

• Performance

• Operating conditions

Definition of Reliability

Probability• A value between 0 and 1• Precise meaning

e.g. probability of 0.97 means that 97 of 100

items will still be working at stated time

under stated conditions

Definition of Reliability

Performance• Some criterion to define when product has

failed

e.g. bearing clearances in an engine or amount

of emissions from a car

Definition of Reliability

Operating conditions• These describe the operating conditions that

correspond to the stated product life. e.g. for a car engine this might mean

→ Speed→ Loading→ Effects of an expected amount of

misuse such as over-revving and stalling.

Reliability Measurement

This is based on the Failure Rate

i.e.TimeOperatingTotal

FailedItemsrateFailure

Some products are scrapped when they fail e.g. hairdryer

Others are repaired e.g. washing machine.

Failure rate over the life of a product

The failure rate is expected to vary over the life of a product – ‘Bathtub Curve’

Time

Fa

ilu

re R

ate

A

C

D

B

Bathtub Curve

A-B Early Failure• ‘Teething’ problems. Caused by design/material

flaws

B-C Constant Failure• Lower than initial failure rate and more or less

constant until end of life

C-D End of life failure• Failure rate rises again due to components

reaching end of life

Simplifying Assumption

• Exponential distribution of failure rate is assumed. This means that the failure rate remains constant over life of product

• Bathtub curve becomes a straight line

Calculating Failure Rate

Time

Fa

ilu

re R

ate

Calculating Failure Rate

Failure rate TimeOperatingTotal

FailedItems

usually expressed by the Greek letter lambda ()

The probability of a product surviving until time (t) is given by the following function:

Reliability at time (t) =

e is the exponential function

te

Procedure

To establish reliability of an item:

• Conduct a series of tests until a number of

them fail.

• Calculate failure rate (Lambda).

• Calculate reliability for a given time using

Reliability at time (t) = e-t

Example

Trial data shows that 105 items failed during a test with a total operating time of 1 million hours. (For all items i.e. both failed and passed).

The failure rate 41005.1

1000000

105 x

Example

Find the reliability of the product after 1000 hours i.e. (t) =1000

Reliability at 1000 hours:

R(1000) = 0.9

te

)10001005.1( 4 xxe

Therefore the item has a 90% chance of surviving for 1000 hours