[OFFICIAL]

INTERNAL .

FUNDAMENTALS OF
RELIABILITY ENGINEERING
ASSIGNMENT
Date 15/10/2022
eg 30/09/2016

Full name Evans Mahopolo

Name and surname

Company name
 [OFFICIAL]

Context:
You have recently been appointed as the reliability engineer for mobile equipment at the Americano Mine.
During the last Engineering meeting, the Engineering Manager raised concerns about the effectiveness of the
tactical maintenance on the diesel engines of a specific haul truck. Her specific concerns were as follows:
• There seems to be a duplication in the maintenance plans on the engine with both OEM
recommendations and in-house condition monitoring plans
• The duration of some of the services seem to be excessive
• Despite this elaborate maintenance plans, some engine failures occurred during the recent past.

At the meeting you were asked to investigate the failures and the current maintenance plans, and recommend
a way forward. You are expected to make a presentation at next month’s Engineering Meeting.

Additional Information
Your first action was to meet with the Fleet Manager, the Section Engineer and the Maintenance Planner to get
more information about the usage of the trucks, the maintenance plans and the engine failures. You made the
following notes during these meetings:
• There are no log books to record the running hours or downtime of the engine, since the trucks operate
7 days a week across three 8-hour shifts (00h00 to 08h00; 08h00 to 16h00; 16h00 to 24h00).
• The maintenance history (service dates and duration, breakdown dates and duration) are captured in
the EAMS
• The OEM has recommended the following maintenance plans for the diesel engines, which are still in
place despite the fact that the trucks are outside of their warranty periods:
o A minor service (8 hours’ duration) should be done every 500 running hours
o A major service (12 hours’ duration) should be done every 1,000 running hours. This service
includes all the tasks of a minor service plus more.
o Engines should be replaced after 10,000 running hours to prevent unexpected breakdowns
• There is a pre-shift operational safety inspection of the truck at the start of each shift, lasting one hour,
in other words three inspections per day.
• On maintenance days some of these inspections are skipped:
o If the maintenance is shorter than 15 hours, 1 inspection is skipped (only 2 inspections done)
o If the maintenance is between 15 and 23 hours, 2 inspections are skipped (only 1 done)
o If the maintenance takes longer than 23 hours, all 3 inspections are skipped (none done)
• The mine also conducts condition-based maintenance (CBM) on these engines, using online sensors
providing continuous readings with appropriate alarm limits and off-line inspections during services. No
additional downtime, over and above the normal service durations, are required for CBM activities.
o The P-F interval for the monitoring techniques in use, is about 1,000 hours
o Preparation for an engine replacement can be made within 250 hours of a decision being made
• The following costs were also provided to you:
o The cost of a new engine is high at US$100,000 per engine
o The cost of planned downtime is fairly low at US$ 2,000 per hour
o The cost of unplanned downtime is extreme at US$20,000 per hour

Note: For simplicity, assume that all engine replacements (both planned and unplanned) and service activities
start at midnight (00:00 am.)

Assignment
2 of 6
 [OFFICIAL]

Assignment: [Total: 40 marks. Estimated duration: 4 hours]

Question 1: [5]
Explain how you will approach this task by describing your analytical process and listing the activities that you
will perform to provide the Engineering Manager with the information that she requested.

Answer:
Problem definition: I would seek to understand the problems associated with failures of the truck, problems in
maintenance management that lead to us having duplicate plans running concurrently. I will do this through
stakeholder engagement e.g conversations with maintenance and production personnel, machine studies and
so on.
Data Collection and processing: I will collect equipment failure data and clean it up for analysis. In this step, I
will make sure that the data collected will be reliable enough to depict the true performance of the equipment in
the current state (initial capability) by removing any outliers. I will also check the master data on the CMMS for
any non-value adding plans.
Analyse data: The next step will be to analyse data collected on the above step. In this stage I will use tools
such as histograms, Jack knife etc to check data distribution with the view of arriving at a conclusion of the
failure distribution.
Decide on alternative solution: Based on the results of data analysis, I will prescribe alternative solutions to
our plant problems such as new maintenance tactics and work management strategies.
Plan and implement alternative solution: In the next engineering meeting, I will propose an alternative
solution and how the solution should be implemented.
Set key performance indicators to measure the gain: I will also set up KPIs to measure the improvement of
my alternative solution and compare with the initial performance state of our truck.

Question 2: [5]
Refer to the spread sheet called “FRE Assignment Data - Learner.xls”. It contains the dates and maintenance
activities for the diesel engines on truck MAC-001 as extracted from the EAMS for the period from 1 January
2014 to 31 December 2020. (For simplicity, assume that all functional failures and planned engine
replacements relate to the same dominant failure mode.) Calculate the running hours for each of the six
engines that were replaced during the 7 years, using the data that you collected from the meetings. Explain
your logic in your answer and attach your spread sheet to show your calculations.
Answer:
The table bellow shows a summary of running hours calculated. Full working is shown on tab Q2-running hrs
on the spread sheet. The logic:
• Start dates are counted after the installation has been completed, eg engine 2 started 1 day after it
replacement engine one (assuming that replacement includes time taken to remove an old one and
install a new one)
• End date: is calculated at a day before the replacement of the preceding one.
• Replacement times are excluded on down time of each engine because they affect overall truck
availability but not individual engine availability. Replacement times are just indicated on the table but
not included on the calculations.
• Down time due to inspections is calculated as default inspection hours minus skipped inspection

Assignment
3 of 6
 [OFFICIAL]

Summary of running hours

Engine Running hours
Engine 1 9868.5
Engine 2 2212
Engine 3 9465.5
Engine 4 10334
Engine 5 9874
Engine 6 9965

Question 3: [5]
Use this information to calculate the MTBF for the engine based on the principle of “normal life expectancy” as
discussed during the training. Explain your reasoning and show all calculations on a separate tab in your
spread sheet.
Answer:
Two engine replacements were excluded in the calculations. Engine 1 replacement is just based on OEM
recommendation with no detailed information on failure, hence this period will also be excluded. Engine 2 failed
prematurely. Based on the principle of “normal life expectancy", its data point will be regarded as an outlier.
Number of valid engine
replacements 4.00
Operating hours (Excluding outlier) 39,638.50
MTBR (Hrs) 9,909.63

Question 4: [5]
If you were to implement a Usage-Based Maintenance (UBM) tactic for the engine, at what interval would you
replace the engine, based on your calculations thus far? (Refer to the training notes for guidance.) How does
this answer compare with the OEM’s recommended replacement interval?

Answer.
Replacement Interval = Average life - 3σ.

Engine 3 Operating Hrs 9,465.50

Engine 4 Operating Hrs 10,334.00
Engine 5 Operating Hrs 9,874.00
Engine 6 Operating Hrs 9,965.00
MTBF 9909.625
Standard deviation 356.686925
Replacement interval calculated 8839.564225

Recommended replacement 8800 hours

interval:

MTBF suggests that the OEM’s recommendation of 10 000 is approximately 3 standard deviations away from
the mean calculated on real plant data. There could various reasons to the difference e.g. operational context.

Assignment
4 of 6
 [OFFICIAL]

Question 5: [5]
Plot a histogram of the valid engine run times (using intervals of 500 hours). Comment on the shape of the
histogram and explain its impact on the maintenance tactics to be considered on the engine.
Answer:
Calculation on Q5 on the spread sheet

2.5

2
Number of failures

1.5

0.5

Operating Hrs

The shape depicts a wear out curve. There seem to be a normal distribution of data around the mean of about
9500 to 10000 hours. Based on this data, we may adopt Usage based tactic.

Question 6: [10]
Comment on the feasibility and cost-effectiveness of using each of the four principal maintenance tactics on the
engine, based on your analysis thus far and the cost information provided to you. What maintenance tactic/plan
would you recommend for the engine? Motivate your answer.
Answer:
The four principal tactics are run until failure, Re-design, Condition based maintenance and usage-based
maintenance.
a. Run to failure: Run to failure will mean that we let all the engines fail and then replace them. This will
be a costly tactic as we would lose $20 000 per hour in down time every time, we let this happen
compared to a planned down time of $2 0000. Therefore, run to failure is uneconomic.

b. Redesign: Redesign will be more applicable if engine failures were out of control and cannot be
addressed through other maintenance tactics. The calculation of the life of these engines are close to
design specifications and hence there is no economic justification to redesign.

c. Usage based maintenance: This strategy means that, based on calculations, we can replace the
engine at specified intervals before failure. Based on expected life calculations and the distribution of
failures, we can estimate the replacement life and implement UBM. However, the downside of this is
that we have a risk of replacing prematurely before actual failure happens and thus losing out on the
benefit of the remaining life. We may also set our replacement interval slightly later than the actual
failure and face the consequences of an unplanned down time. In short, the UBM interval prediction is
as accurate as the process used to calculate it.

d. Condition Based Maintenance: Here we monitor the performance of the engine and set acceptable
limits to raise alarms to replace it at the earliest sign of functional failure. This tactic has a lot of

Assignment
5 of 6
 [OFFICIAL]

advantages as we can monitor the condition online and plan replacements/follow on tasks without
incurring inspection down time. Moreover, this tactic is predictive in nature as we can implement
techniques like oil sampling and vibration analysis that can warn us before things get out of hand. We
can also realise full benefits on asset life (ie. avoid premature replacement) while also guarding against
late replacement intervals that will lead to unplanned down time. The downside is that, there is need to
invest more money into condition monitoring devices, training staff on technologies as well as on data
capturing and analysis.

Question 7: [5]
Analyse the historical data about the actual duration of the 1,000-hourly Full Service by calculating the mean
and mode of these durations and comparing it with the OEM’s estimated duration. What would your
recommendation be about the estimated duration to be used by the Maintenance Planner in future? Motivate
your answer and estimate the savings to the mine.

Answer:
I excluded outliers as part of my data clean up. Calculation on Q7 tab on spread sheet.

Hours
OEM full-service recommendation 12
Mean 10.75
Mode 10.5
Mean difference with OEM 1.25
Mode difference with OEM 1.5

I recommend that the planner uses the mode 10.5hrs to plan for service durations (frequently appearing time).
Savings = 1.5 hrs X $2000/hr = $3000.

Assignment
6 of 6