Taibah University
College of Engineering
Industrial Engineering Department
IE-442-Engineering Reliability and
Maintainability
Chapter 9: Maintainability
Maintenance
• Reactive Maintenance: Performed in response to unplanned or unscheduled
downtime of the unit, usually as a result of a failure whether it be internal or
external.
• Proactive Maintenance
– Preventive Maintenance: scheduled downtime, usually periodical, in which a
well-defined set of tasks (inspection, repair, replacement, cleaning, etc.) are
performed.
– Predictive Maintenance: Aimed at detecting the state of a part (whether it
should be replaced, repaired, or whether it is near failure) through diagnostic tools.
– Should be performed when it is cost-effective. Proactive Maintenance should
reduce the number of unscheduled failures
Total Downtime
• Maintainability: Probability that a failed system or
component will be resolved or repaired to a specified
condition within a period of time when maintenance is
performed in accordance with prescribed procedures.
• Repair time (Different from Total Downtime!): An inherent
design feature of the component, which reflects its
maintainability
Total Downtime
Factors of Maintainability
• Since Maintainability is a probability, it can be characterized by
specifying a repair-time probability distribution.
Measures of Maintainability
• Mean Time to Repair: MTTR
• Median Time to Repair
• Mode (Most Likely Repair Time)
• Time in which a certain percentage of failures must be repaired
• Mean Repair Time plus Mean Preventive Maintenance Downtime
• Number of Maintenance Hours per Operating Hour
Repair Time Distribution
• • Probability that a repair will be accomplished within time t
• T: Random variable: Time to repair a failed unit
• h(t’): Probability density function of time to repair a failed unit
• Mean and Variance of Time to Repair distribution
Repair Time Distribution
Exponential repair times
System Repair Time
• System MTTR as a weighted average of its subsystems MTTRs. Weights
are functions of relative number of
failures. If
- qi: Number of identical subsystems of type i
- fi: Expected number of failure of the ith unique subsystem over the
system design life
- toi: Total number of operating hours of the ith component over the
system design life
Then
System Repair Time
System Repair Time/ system having redundant
components
• Case 1: A component is repaired as soon as it fails
• Case 2: A component is repaired only when all k component have failed
– All the failed components are repaired at the same time
– Only one component can be repaired at a time
Reliability Under Preventive Maintenance
• For complex systems, increased reliability can often be achieved through
a preventive maintenance program.
– Can reduce the effect of aging and wearout
• Assumption: the system is restored to its original condition
• R(t): Reliability of system without maintenance
• T: Interval between preventive maintenance
• Rm(t): Reliability of the system with preventive maintenance
Reliability Under Preventive Maintenance
• R(t)n: Probability of surviving n maintenance intervals
• R(t-nT): Probability of surviving t-nT time units past the last preventive maintenance
Reliability Under Preventive Maintenance
Reliability Under Preventive Maintenance
Reliability Under Preventive Maintenance
Under preventive maintenance
State-Dependent Systems with Repair
• For 2 active redundant components
• Assumptions:
– Components have the same constant failure rate λ
– Components have the same constant repair rate r
• State 1: Both component are operating
• State 2: One component operating, one component in repair
• State 3: Both component have failed
State-Dependent Systems with Repair
• For 2 components one on-line and one on standby. Now on-line can be repaired when it
fails
• Assumptions:
– On-line component has constant repair rate r
State-Dependent Systems with Repair
State-Dependent Systems with Repair
State-Dependent Systems with Repair
Questions?