2.1 Averages

2.1 Averages

In aviation maintenance engineering, data-driven decisions are made daily. Technicians and engineers analyze parameters ranging from engine temperatures and tire pressures to component failure rates and structural dimensions. To make sense of this data, we use measures of central tendency, commonly referred to as averages. In the EASA Part-66 examination, you must demonstrate a thorough understanding of the mathematical principles behind the arithmetic mean, the median, the mode, and the weighted average, as well as their direct applications in aircraft maintenance.

The Arithmetic Mean

The arithmetic mean (often simply called the "mean" or "average") is the sum of a series of values divided by the total number of values in that series. Mathematically, it is expressed as:

$\bar{x} = \frac{1}{n}\sum_{i=1}^{n} x_i = \frac{x_1 + x_2 + \dots + x_n}{n}$

Where:

• $\bar{x}$ represents the mean.
• $x_i$ represents each individual data point.
• $n$ is the total number of data points.

Aviation Application: Cylinder Head Temperatures (CHT)

Consider a six-cylinder reciprocating engine where a technician records the CHT of each cylinder during a test run:

• Cylinder 1: $380^{\circ}\text{F}$
• Cylinder 2: $395^{\circ}\text{F}$
• Cylinder 3: $410^{\circ}\text{F}$
• Cylinder 4: $390^{\circ}\text{F}$
• Cylinder 5: $400^{\circ}\text{F}$
• Cylinder 6: $385^{\circ}\text{F}$

To find the mean CHT:

$\bar{x} = \frac{380 + 395 + 410 + 390 + 400 + 385}{6} = \frac{2360}{6} \approx 393.33^{\circ}\text{F}$

While the arithmetic mean provides a useful baseline of the engine's overall thermal state, it has a significant limitation: it can mask individual anomalies. For instance, if Cylinder 3 was running at a dangerously high $450^{\circ}\text{F}$ due to a partially clogged fuel injector, the mean would only rise to $400^{\circ}\text{F}$, potentially concealing the localized thermal stress.

---

The Median

The median is the middle value in a dataset when the values are arranged in ascending or descending order. If the dataset has an odd number of observations, the median is the single middle value. If the dataset has an even number of observations, the median is the arithmetic mean of the two middle values.

• For an odd number of elements ($n$), the median position is: $\frac{n+1}{2}$
• For an even number of elements ($n$), the median is: $\frac{x_{n/2} + x_{(n/2)+1}}{2}$

Aviation Application: Component Reliability Analysis

Analyzing the reliability and lifespan of components often involves skewed datasets. Suppose a maintenance department tracks the service life (in operating hours) before failure of five hydraulic pumps: $\text{Dataset: } 1200\text{ hrs, } 1450\text{ hrs, } 1500\text{ hrs, } 1600\text{ hrs, } 4800\text{ hrs}$

Let's compare the mean and the median:

1. Arithmetic Mean: $\bar{x} = \frac{1200 + 1450 + 1500 + 1600 + 4800}{5} = \frac{10550}{5} = 2110\text{ hours}$
2. Median: Since $n = 5$ (an odd number), the median is the third value in the sorted list, which is $1500\text{ hours}$.

In this scenario, the arithmetic mean ($2110\text{ hours}$) is heavily skewed upward by a single outlier—a pump that lasted an exceptionally long time ($4800\text{ hours}$). The median ($1500\text{ hours}$) is a much more representative average of the typical lifespan a technician can expect from these pumps before they fail.

---

The Mode

The mode is the value that occurs most frequently in a dataset. A dataset may have one mode (unimodal), more than one mode (bimodal or multimodal), or no mode at all if every value occurs only once.

Aviation Application: Tool and Failure Code Tracking

In hangar operations, the mode is highly useful for identifying systemic issues or high-use items:

• Failure Codes: If an avionics shop logs the following error codes during troubleshooting of a VHF navigation system: E12, E34, E34, E88, E34, E12, the mode is E34. This tells the engineer that error E34 is the most common failure mode, directing their troubleshooting focus.
• Tooling: Analyzing the sizes of fasteners removed during a wing inspection panel service helps a lead technician stock the most common sockets. If the sizes removed are $3/8\text{ in}$, $5/16\text{ in}$, $3/8\text{ in}$, $7/16\text{ in}$, and $3/8\text{ in}$, the mode is $3/8\text{ in}$.

---

Weighted Averages

A weighted average (or weighted mean) is used when some data points contribute more to the final average than others. Each data point ($x_i$) is assigned a weight ($w_i$), which reflects its significance, volume, or mass. The formula is:

$\bar{x}_w = \frac{\sum_{i=1}^{k} w_i x_i}{\sum_{i=1}^{k} w_i} = \frac{w_1 x_1 + w_2 x_2 + \dots + w_k x_k}{w_1 + w_2 + \dots + w_k}$

---

Aviation Application: Weight and Balance (Center of Gravity)

The most critical application of weighted averages in aviation is the calculation of an aircraft's Center of Gravity (CG). An aircraft is balanced around a pivot point, and its longitudinal CG must fall within strict limits defined by the manufacturer for safe flight.

The CG is calculated along a horizontal axis relative to a reference point called the datum.

• Weight ($W$): The mass of the aircraft or individual components (acts as the weight $w_i$).
• Arm ($A$): The distance from the datum to the center of gravity of the component (acts as the value $x_i$).
• Moment ($M$): The product of weight and arm ($M = W \times A$).

Therefore, the CG location (the average arm) is the weighted average of the arms, using weights as the weighting factors:

$CG\ (Arm) = \frac{\sum Moments}{\sum Weights} = \frac{W_1 A_1 + W_2 A_2 + \dots + W_n A_n}{W_1 + W_2 + \dots + W_n}$

Worked Example: Helicopter Empty Weight CG

A helicopter is placed on three scales for an empty weight determination. The datum is located at the nose of the helicopter. The measurements are:

• Nose Gear: Weight ($W_N$) = $345\text{ kg}$, Arm ($A_N$) = $1100\text{ mm}$ aft of datum.
• Left Main Gear: Weight ($W_{LM}$) = $1020\text{ kg}$, Arm ($A_{LM}$) = $3200\text{ mm}$ aft of datum.
• Right Main Gear: Weight ($W_{RM}$) = $1015\text{ kg}$, Arm ($A_{RM}$) = $3200\text{ mm}$ aft of datum.

To find the Empty Weight Center of Gravity (EWCG):

1. Calculate Individual Moments: $\text{Nose Gear Moment: } M_N = 345\text{ kg} \times 1100\text{ mm} = 379,500\text{ kg}\cdot\text{mm}$ $\text{Left Main Gear Moment: } M_{LM} = 1020\text{ kg} \times 3200\text{ mm} = 3,264,000\text{ kg}\cdot\text{mm}$ $\text{Right Main Gear Moment: } M_{RM} = 1015\text{ kg} \times 3200\text{ mm} = 3,248,000\text{ kg}\cdot\text{mm}$

2. Calculate Total Weight: $\sum W = 345 + 1020 + 1015 = 2380\text{ kg}$

3. Calculate Total Moment: $\sum M = 379,500 + 3,264,000 + 3,248,000 = 6,891,500\text{ kg}\cdot\text{mm}$

4. Calculate Center of Gravity: $CG = \frac{6,891,500\text{ kg}\cdot\text{mm}}{2380\text{ kg}} \approx 2895.59\text{ mm}$

The weighted average CG is $2895.59\text{ mm}$ aft of the datum. This is a critical value that must be logged in the aircraft's weight and balance records.

---

Aviation Application: EASA Fleet Weight Averages

Under EASA Part-CAT regulations (specifically CAT.POL.A.305), air operators of large fleets of identical aircraft of the same type may establish a fleet average weight and a fleet average CG rather than weighing every single aircraft continuously.

However, strict mathematical rules apply to ensure safety:

• A representative sample of the fleet must be weighed at regular intervals.
• The dry operating weight (DOW) of any individual aircraft in the fleet cannot deviate from the fleet average by more than $0.5%$ of the maximum structural landing weight (or $1%$ of the maximum zero fuel weight).
• The CG position of any individual aircraft cannot deviate from the fleet average CG by more than $0.5%$ of the Mean Aerodynamic Chord (MAC).

Worked Example: Fleet Deviation Compliance

An airline operates a fleet of 4 twin-engine utility aircraft. The maximum structural landing weight of this model is $6,000\text{ kg}$. The weighed empty weights of the four aircraft are: $\text{Aircraft A: } 3,200\text{ kg, Aircraft B: } 3,220\text{ kg, Aircraft C: } 3,190\text{ kg, Aircraft D: } 3,210\text{ kg}$

1. Calculate the Fleet Average Weight: $\text{Average Weight} = \frac{3200 + 3220 + 3190 + 3210}{4} = \frac{12820}{4} = 3,205\text{ kg}$
2. Determine the Allowable Deviation Limit: $\text{Limit} = 0.5\% \text{ of } 6,000\text{ kg} = 0.005 \times 6000 = 30\text{ kg}$
3. Check Individual Deviations:
   • Aircraft A: $|3200 - 3205| = 5\text{ kg}$ (Compliant)
   • Aircraft B: $|3220 - 3205| = 15\text{ kg}$ (Compliant)
   • Aircraft C: $|3190 - 3205| = 15\text{ kg}$ (Compliant)
   • Aircraft D: $|3210 - 3205| = 5\text{ kg}$ (Compliant)

Since no individual aircraft deviates from the fleet average weight of $3,205\text{ kg}$ by more than $30\text{ kg}$, the airline is legally permitted to use the fleet average weight for performance calculations.

---

Aviation Application: Fuel Density Averages

Fuel density changes significantly with temperature, which affects fuel load calculations. Jet A-1 density typically ranges from $0.78\text{ kg/L}$ (in hot environments) to $0.84\text{ kg/L}$ (in cold environments). If you mix fuel from different sources with different temperatures, a weighted average density must be calculated to find the total fuel mass.

Worked Example: Mixed Density Refueling

An aircraft has $4,000\text{ liters}$ of fuel remaining in its tanks at a density of $0.80\text{ kg/L}$. During a quick turnaround, $10,000\text{ liters}$ of fuel is uploaded from a fuel bowser at a colder temperature, having a density of $0.82\text{ kg/L}$. What is the average density of the fuel mixture in the aircraft tanks?

Using the weighted average formula, where the volumes act as weights ($w_i$) and the densities act as values ($x_i$):

$\text{Average Density} = \frac{(4000\text{ L} \times 0.80\text{ kg/L}) + (10000\text{ L} \times 0.82\text{ kg/L})}{4000\text{ L} + 10000\text{ L}}$ $\text{Average Density} = \frac{3200\text{ kg} + 8200\text{ kg}}{14000\text{ L}} = \frac{11400\text{ kg}}{14000\text{ L}} \approx 0.814\text{ kg/L}$

Understanding these relationships is key to preventing fuel mass calculation errors, which can lead to critical range or structural issues.