Reliability Maintenance Engineering Day 1 session 4 Estimating Reliability
Three day live course focused on reliability engineering for maintenance programs. Introductory material and discussion ranging from basic tools and techniques for data analysis to considerations when building or improving a program.
2. USING FIELD AND TEST DATA TO
ESTIMATE RELIABILITY
Day 1 Session 4
3. Objectives
• Discussing reliability data types
• Considering complete and censored data
• Analyzing trend plotting, 80-20 plots and
statistical plots
• Drawing a best-fit regression line with Weibull
• Recording data related to reliability
performance
• Providing continuity in data collection systems
6. Reliability Data Types
• Pass/Fail data
• Attribute data
• Time or cycles to failure
• Variables data
• Start and stop
definitions
7. Reliability Data Types
• Complete data
– All units have failed
• Censored data
– Right censored – not all
have failed yet
– Left censored – not sure
when some started
– Interval censored – not
sure when failure occurred
between inspections.
12. Trend Plotting
Advantages
• Easy
• Visual
• Actual data
Disadvantages
• Difficult to see changes
• Lagging indicator
• May require adjustments
for scheduled downtime
13.
14. 80 – 20 or Pareto plotting
Advantages
• Easy
• Visual
• Prioritization build in
Disadvantages
• Difficult to see changes over
time
• Doesn’t include costs
impact
16. Life Data CDF
Advantages
• Accounts for all data
• Visual
• Easy to read
Disadvantages
• Difficult to see changes
• Requires probability paper
or software
• Requires time to failure data
CDF – Cumulative Distribution Function -
17.
18. MCF plotting
Advantages
• Very Easy to create
• Easy to see patterns
Disadvantages
• Difficult to summarize
• Requires more than two
failures/repairs per
equipment
• Doesn’t account for
scheduled downtime or
extended repairs
MCF – Mean Cumulative Function
34. Comparisons over time
• Break up data by time
– Before/after change
– Before/after experiment
– Before/after overhaul
• Labels
• Show magnitude of
impact clearly, if any.
35. Data collection over time
• Vendor data
• Installation
• Maintenance
• Decommissioning
37. Summary
• Discussing reliability data
types
• Considering complete and
censored data
• Analyzing trend plotting, 80-
20 plots and statistical plots
• Drawing a best-fit regression
line with Weibull
• Recording data related to
reliability performance
• Providing continuity in data
collection systems
Using field and test data
to estimate reliability