BPT 2423 STATISTICAL PROCESS CONTROL Fundamental Aspects Product

BPT 2423 – STATISTICAL PROCESS CONTROL

§ § Fundamental Aspects Product Life Cycle Curve Measures of Reliability § Failure Rate, Mean Life and Availability § Calculating System Reliability Engineers

§ § § Known the importance of system reliability Identify on how performance during the life of a product, process or system is affected by its design and configuration Able to compute the reliability of systems including systems in series, parallel and hybrid combinations

§ § Reliability (quality over the long term) is the ability of a product to perform its intended function over a period of time and under prescribed environmental conditions Reliability of component dependent on many factors: �quality of research performed at its conception �the original design and any subsequent design changes �the complexity of the design �the manufacturing processes �the handling received during shipping �the environment surrounding its use �the end users and etc.

§ § § Reliability testing enables a company to better comprehend how their products will perform under normal usage as well as extreme or unexpected situations Reliability programs provide information about product performance by systematically studying the product Four factors associated with reliability: 1. Numerical value 2. Intended function 3. Life 4. Environmental conditions

Life cycle of a product is commonly broken down into 3 phases: 1. Early failure (infant mortality) � phase is characterized by failures occurring very quickly after the product has been produced or put into use by the consumer � the curve during this phase is exponential with the number of failures decreasing the longer the product is in use � some early failures are due to inappropriate or inadequate materials, marginal components, incorrect installation or poor manufacturing techniques

2. Chance failure § During the failure, portion of a product’s useful life, failures occur randomly § May be due to inadequate or insufficient design margins § Misapplication or misuse of the product by the consumer can lead to product failure – eg. : overstressing 3. Wear-out § Failures increase in number until few, if any of the product are left § Due to a variety of causes such as related to actual product function or cosmetic (scratched, dented, discolorations, misalignments and interference between components)

§ § Overall system reliability depends on the individual reliabilities associated with the parts, components and subassemblies Reliability values are sought to determine the performance of a product, reveal any recurring patterns of failure and the underlying causes of those failures Reliability test determine what failed, how it failed and the number of hours, cycles, actuations or stresses it was able to bear before failure If result known, decisions can be made concerning product reliability expectations, corrective action steps, maintenance procedures and cost of repair/replacement

Several different types of test exist to judge the reliability of a product : 1. Failure-terminated test �Ended when a predetermined number of failures occur within the sample being tested �Decision concerning whether or not the product is acceptable is based on the number of products that have failed during the test 2. Time-terminated test �Concluded when an established number of hours is reached �Product is accepted on the basis of how many products failed before reaching the time limit 3. Sequential Test - relies on the accumulated results of the tests

Determine Failure Rate, λ Determine Average Life, θ or Determine Availability

Example : Determine the failure rate for a 90 -hour test of 12 items where 2 items fail at 45 and 72 hours, respectively. What is the mean life of the product? λ = 2 45 + 72 + (10) 90 = 2 / 1017 = 0. 00197 θ = 1/λ = 1 / 0. 00197 = 507. 6 hours

Reliability is the probability that a product will not fail during a particular time period. where : s – those units performing satisfactorily n – total number of units being tested Reliability in Series