The primary objective of an airline powerplant engineer
responsible from turbofan aircraft engines is
to technically manage the engines to be available for
revenue flight, while achieving the desired goals of
cost of use, reliability and safety, with adequate engine
performance level for operational flexibility.
For an airline, flight safety is a must. Thus, aviation
authorities throughout the world strictly bound the
range in which an airline could move to meet the
above goals. Since, approximately 40-45% of total
aircraft maintenance cost is directly arising from
engine maintenance costs, meeting these objectives
becomes a very important issue as far as the competitiveness
of the airline business is concerned.
Under these circumstances, airlines always seek to
find and explore methods in order keep the powerplants
running with an optimized cost versus on
wing life.
The engine on wing maintenance concept is known
as “on-condition maintenance”. In this concept, engines
are continuously monitored during their on
wing operation in order to prevent failures and to
meet goals on reliability and safety. In other words,
engines are kept on wing as long as reliability,
safety and performance levels are acceptable. All of
the above conditions end up with raising the following
important question: What is the time on wing of
an aircraft engine that will fulfill all of these goals
at the required levels? This question forms the basis
of analyzes performed in this study.
It is known that there is an optimum time at which
engine should have essential restoration done, in
order to meet an optimized cost for removal. In
other words, at times prior to the optimum one; the
opportunity costs which are arising due to repairing
an engine before all its useful life is consumed result
in an increase in maintenance cost. At times greater
than the optimum; the number of parts which need
repair, the difficulty and cost of repair and the number
of parts which must be scrapped increase. As a
result, the maintenance cost increases.
In order to search for this optimum, available data
is gathered from one airline and analyzed both analytically
and statistically. The problem is mathematically
modeled by making assumptions, without effecting
the accuracy of the real problem. In order to achieve a better understanding, the objectives
(maintenance cost, performance, reliability and
safety) are studied as sub problems. The data relating
these objectives to the engine on wing time are
searched and analyzed. For each of these objectives,
mathematical expressions are derived. As far as the
integrity of the goals of maintenance cost, performance,
reliability and safety is concerned, a multi objective
optimization problem is proposed. The model
is converted into a fitness function form by an airline
engineering perspective and multi objective problem
of on wing life optimization is solved by using a Genetic
Algorithm based solver. Genetic Algorithm
method has been selected as being a robust and reliable
technique for multi objective engineering optimization
problems. Since performance deterioration
characteristics end up with three different types of
deterioration curves, three sub problems have been
solved. This ensured the diversity in terms of actual
engine performance deterioration characteristics,
which is a result of operational, configurational and
manufacturing diversities encountered in the real
operation. Performance related diversity is believed
to cover the above mentioned facts of the actual
problem and enables a better forecast on the optimum
on wing life of commercial turbofan aircraft
engines.
All the results are gathered considering airline
engineering priorities and the optimum on wing
life of a CFM56-3C1 engine is calculated. A removal
planned within this optimum removal interval
will achieve all the priorities in terms of minimum
maintenance cost, adequate performance,
while not sacrificing from reliability and safety.
The result of the optimization is compared with
the actual removals and found to be in line with
the actual removal time frame. This proves the
proficiency of proposed method and the mathematical
model.
Additionally, the operational factors affecting the
aircraft engine on wing life is discussed in order to
derive a generalized mathematical model for adaptation
to other airline engine fleets and to other engine
types. Thus, a generaized model is formulated
in order to find the optimum time on wing for any
type of commercial turbofan engine operated in any
fleet.
