14.1.1 Summary

Because the preliminary configuration is based on past experience and statistics, an iterative procedure ensues to fine‐tune the aircraft converging into the correct size of the wing reference area for a family of variant aircraft designs and matched engines. Wallace [3] provides an excellent presentation on the subject and Loftin [4] provides an aircraft sizing exercise in detail.

This chapter includes the following sections:

• Section : Introduction
• Section : Theory
• Section : Coursework Exercises for Civil Aircraft (Bizjet)
• Section : Sizing Analysis and Variant Designs: Civil Aircraft (Bizjet)
• Section : Coursework Exercises for Military Aircraft (AJT)
• Section : Sizing analysis and Variant Designs: Military Aircraft (AJT)
• Section : Aircraft Sizing Studies and Sensitivity Analyses
• Section : Discussions

Coursework content. This is an important chapter. Readers are to compute the parameters that establish the criteria for aircraft sizing and engine matching. The final size is unlikely to be identical to the preliminary configuration obtained in Chapter 8 using statistics of past designs. It is recommended that the readers make use of spreadsheets to carry out the iterations.

14.2.1 Civil Aircraft

Based on circa 2000 fuel prices, the aircraft cost contributes to the direct operating cost (DOC) three to four times the contribution made by the fuel cost. It is not cost effective for aircraft manufacturers to offer custom‐made new designs to each operator with varying payload‐range requirements. Therefore, aircraft manufacturers offer aircraft in a family of variant designs. This approach maintains maximum component commonality within the family to reduce development costs and is reflected in aircraft unit‐cost savings. In turn, it eases the amortisation of nonrecurring development costs, as sales increase. It is therefore important for the aircraft sizing exercise to ensure that the variant designs are least penalised to maintain commonality of components. This is what is meant by producing satisfying robust designs; these are not necessarily the optimum designs. Sophisticated multi‐disciplinary optimisation (MDO) is not easily amenable to ready use. The industries also use parametric search for a satisfying robust design.

To generate a family of variant civil aircraft designs, the tendency is to retain the wing and empennage as almost unchanged while plugging and unplugging the constant fuselage to cope with varying payload capacities. Typically, the baseline aircraft remains as the middle design. The smaller aircraft results in a wing that is larger than necessary, providing better field performances (i.e. takeoff and landing); however, the cruise performance is slightly penalised. Conversely, larger aircraft have smaller wings that improve the cruise performance; the shortfall in takeoff is overcome by providing a higher thrust‐to‐weight ratio (T_SLS/W) and possibly with better high‐lift devices, both of which incur additional costs. The baseline aircraft approach speed, V_app, initially is kept low enough so that the growth of V_app for the larger aircraft is kept within the specifications. Of late, high investment with advanced composite wing‐manufacturing method is in a better position to produce adjusted wing sizes for each variant (large aircraft) in a cost effective manner, offering improved economics in the long run. However, for some time to come, metal wing construction is to continue with minimum changes in wing size to maximise component commonality.

Matched engines are also used in a family to meet the variation of thrust (or power) requirements for the aircraft variants. The sized engines are bought‐out items supplied by engine manufacturers. Aircraft designers stay in constant communication with engine designers in order to arrive at the family of engines required. A thrust variation of up to ±30% from the baseline engine is typically sufficient for larger and smaller aircraft variants from the baseline. Engine designers can produce scalable variants from a proven core gas‐generator module of the engine – these scalable projected engines are known loosely as rubberised engines. The thrust variation of a rubberised engine does not affect the external dimensions of an engine (typically, the bare engine length and diameter change only around ±2%). This book uses an unchanged nacelle external dimension for the family variants, although there is some difference in weight for the different engine thrusts. The generic methodology presented in this chapter is the basis for the sizing and matching practice.

14.2.2 Military Aircraft

It follows the same procedure as in the civil aircraft methodology except that the specifications are different. The turn rate capability is an additional parameter that makes a decisive contribution to the design (see Section 14.3).

14.3.1 Sizing for Takeoff Field Length (TOFL) – Two Engines

To keep analyses simple, only a two‐engine aircraft is sized and engine matched. This is because the sizing and engine matching exercise requires aircraft drag polar and in this book only the two‐engine Bizjet drag polar is worked out. However, three‐ and four‐engine aircraft use exactly the same procedure, equations modified by the number of engines in question.

TOFL is the field length required to clear a 35‐ft (10‐m) obstacle while maintaining a specified minimum climb gradient, γ, with one engine inoperative and flaps and undercarriage extended (Figure 14.1). The Federal Aviation Regulation (FAR) requirements for a two‐engine aircraft minimum second segment climb gradient are given in Section 15.3.3.

For sizing, field length calculations are at the sea level standard day (no wind) and at a zero airfield gradient of paved runway. For further simplification, drag changes are ignored during the transition phase of lift‐off to clear the obstacle (flaring takes about two to four seconds); in words, the equations applied to Vlift‐off are extended to V₂.

Chapter 15 addresses takeoff performance in detail. Eq. (15.8) gives

14.1

where dV/dt = a is the instantaneous acceleration and V is the instantaneous velocity. The aircraft on the ground encounters rolling friction (coefficient μ = 0.025 for a paved, metalled runway). Average acceleration, ā, is taken at 0.7V₂. FAR requires V₂ = 1.2V_stall. This gives V₂² = [2 × 1.44 × (W/S_W)]/(ρC_Lstall). An aircraft stalls at C_Lmax. By substituting the value of V₂, the Eq. 14.1 reduces to:

Its simplified expression for all engines operating is:

14.2

Writing in terms of wing loading Eq. 14.2 can be written as

14.3

where average acceleration, ā = F/m and applied force F = (T − D) − μ(W − L).

Note that, until lift‐off is achieved, W > L and F is the average value at 0.7V₂.

14.4

Substituting in Eq. 14.3 it becomes

14.5

In the foot–pound system (FPS) it can be written as (ρ = 0.00238 slugs and g = 32.2 ft s⁻²)

14.6a

In the SI system it becomes (ρ = 1.225 kg m⁻³ and g = 9.81 m s⁻²)

14.6b

here W/S_W is in N m⁻² to remain in alignment with the units of thrust in newtons.

Checking of the second‐segment climb gradient occurs after aircraft drag estimation, which is explained in Chapter 11. If climb gradient falls short of the requirement, then the T_SLS must be increased. In general, the stringent TOFL requirements also likely to satisfy the second segment climb gradient.

14.3.1.1 Civil Aircraft Design: Takeoff

The contribution of the last three terms (−D/T − μW/T + μL/T) in Eq. 14.4 is minimal and can be omitted at this stage for the sizing calculation. In addition, for the one‐engine‐inoperative condition after the decision speed V₁, the acceleration slows down, making the TOFL longer than the all engines operative case. Therefore, in the sizing computations to produce the specified TOFL, further simplification is possible by applying a semi‐empirical correction factor primarily to compensate for loss of an engine. The correction factors are as follows [2]; all sizing calculations are performed at the MTOW and with T_SLS.

For two engines, use a factor of 0.5 (loss of thrust by a half).

Then, Eqs. (14.6a) and (14.6b) in the FPS system reduces to:

14.7a

For the SI system:

14.7b

For three engines, use a factor of 0.66 (loss of thrust by a third). Then, Eq. (14.6) in the FPS system reduces to:

14.8a

For the SI system:

14.8b

For four engines, use a factor of 0.75 (loss of thrust by a quarter). Then, Eq. (14.6) in the FPS system reduces to:

14.9a

For the SI system:

14.9b

14.3.1.2 Military Aircraft Design: Takeoff

Because military aircraft mostly have a single engine, there is no requirement for one engine being inoperative; ejection is the best solution if the aircraft cannot be landed safely. Therefore, Eqs. (14.6a) and (14.6b) can be directly applied (for a multiengine design, the one‐engine‐inoperative case, generally, uses measures similar to the civil aircraft case). In the FPS system, this can be written as:

14.10a

In the SI system, it becomes:

14.10b

Military aircraft have a thrust, T_SLS/W, that is substantially higher than civil aircraft, which makes (D/T − μW/T + μL/T) even smaller. Therefore, for a single‐engine aircraft, no correction is needed and the simplified equations are as follows:

In the FPS system, this can be written as:

14.11a

In the SI system, this becomes:

14.11b

14.3.2 Sizing for the Initial Rate of Climb (All Engines Operating)

The initial unaccelerated rate of climb is a user specification and not a FAR requirement; this is when the aircraft is heaviest in climb. In general, the FAR requirement for the one‐engine‐inoperative gradient provides sufficient margin to give a satisfactory all‐engine initial climb rate. However, from the operational perspective, higher rates of climb are in demand when it is sized accordingly. Military aircraft (some with a single engine) requirements stipulate faster climb rates and sizing for the initial climb rate is important. The methodology for aircraft sized to the initial climb rate is described in this section. Sizing exercise for climb is at unaccelerated rate of climb. Figure 14.2 shows a typical climb trajectory.

For a steady‐state climb, the Eq. (15.5) gives the expression for rate of climb, RC = V × sin γ.

Steady‐state force equilibrium gives T = D + W × sin γ or sin γ = (T − D)/W.

This gives

14.12

Equation 14.12 is written as

14.13

Equation 14.13 is based on climb thrust rating, which is lower than T_SLS. Equation 14.13 needs to be written in terms of T_SLS. The ratio, T_SLS/T = k_cl varies depending on the engine bypass ratio (BPR).

14.14

14.15

The drag polar is now required to compute the relationships given in Equation 14.14.

14.3.3 Sizing to Meet Initial Cruise

There are no FAR or Milspec regulations to meet the initial cruise speed; initial cruise capability is a user requirement. Therefore, both civil and military aircraft sizing for initial cruise use the same equations. At a steady‐state level flight, thrust required (aeroplane drag, D) = thrust available (T_a); that is:

14.16

Dividing both sides of the equation by the initial cruise weight, W_{in_cr} = k × MTOW due to fuel burned to climb to the initial cruise altitude. The factor k lies between 0.95 and 0.98 (climb performance details are not available during the conceptual design phase), depending on the operating altitude for the class of aircraft, and it can be fine‐tuned through iterations: in the coursework exercise, one round of iterations is sufficient. The factor cancels out in the following equation but is required later. Henceforth, in this part of cruise sizing, W represents the MTOW, in line with the takeoff sizing:

14.17

The drag polar gives the C_D value to correspond to the initial cruise C_L (because they are non‐dimensional, both the FPS and SI systems provide the same values). Initial cruise:

14.18

The thrust‐to‐weight ratio sizing for initial cruise capability is expressed in terms of T_SLS. Eq. 14.18 is based on the maximum cruise thrust rating, which is lower than the T_SLS. Eq. 14.18 must be written in terms of T_SLS. The ratio, T_SLS/T_a = k_cr; varies depending on the engine BPR. The factor k_cr is computed from the engine data supplied. Then, Eq. 14.18 can be rewritten as:

14.19

Variation in wing size affects aircraft weight and drag. The question now is: How does the C_D change with changes in W and S_W? (T_a changes do not affect the drag because it is assumed that the physical size of an engine is not affected by small changes in thrust.) The solution method is to work with the wing only – first by scaling the wing for each case and then by estimating the changes in weight and drag and iterating – which is an involved process.

This book simplifies the method by using the same drag polar for all wing‐loadings (W/S_W) with little loss of accuracy. As the wing size is scaled up or down (the AR invariant), it changes the parasite drag. The induced drag changes as the aircraft weight increases or decreases. However, the C_D increase with wing growth is divided by a larger wing, which keeps the C_D change minimal.

14.3.4 Sizing for Landing Distance

The most critical case is when an aircraft must land at its maximum landing weight of 0.98 MTOW. In an emergency, an aircraft lands at the same airport for an aborted takeoff procedure, assuming a 2% weight loss due to fuel burn in order to make the return circuit. Pilots prefer to approach as slow as possible for ease of handling at landing.

For the Bizjet class of aircraft, the approach velocity, V_app (FAR requirement at 1.3 V_stall) is less than 125 kts to ensure that it is not constrained by the minimum control speed, V_c. Wing C_Lstall is at the landing flap and slat setting. For sizing purposes, an engine is set to the idle rating to produce zero thrust.

14.20

At landing V_app = 1.3 V_stall.

Therefore,

14.21

14.22

14.4.1 Takeoff

Requirements. TOFL 4400 ft (1341 m) to clear a 35‐ft height at ISA + sea level. The maximum lift coefficient at takeoff (i.e. flaps down to 20° and no slat) is C_Lstall(TO) = 1.9 (obtained from testing and computational fluid dynamics (CFD) analysis). Installation losses are taken to 7% that reduces takeoff thrust to 0.93 T_SLS. Using Eq. (14.7a), the expression reduces to:

Using Eq. (14.7a), the expression reduces to W/S_W = 4400 × 1.9 × (0.93 × T_SLS/W)/37.5 = 207.3 × (T/W).

Using Eq. (14.7b), it becomes W/S_W = 4.173 × 1341 × 1.9 × (0.93 × T_SLS/W) = 9889.2 × (T_SLS/W)

The result is computed in Table 14.1.

The industry must also examine other takeoff requirements (e.g. an unprepared runway) and hot and high ambient conditions.

14.4.2 Initial Climb

From the market requirements, an initial climb starts at an 1400‐ft altitude (ρ = 0.00232 slugs/ft³) at a speed V_EAS = 250 kt (Mach 0.38) = 250 × 1.68781 = 422 ft s⁻¹ (128.6 m s⁻¹) and the required rate of climb, RC = 2600 ft min⁻¹ (792.5 m min⁻¹) = 43.33 ft s⁻¹ (13.2 m s⁻¹). For the class of TFE731 engine data, Figure 13.3 gives the uninstalled k_cl = T/T_SLS = 0.705. Taking 5% installation loss during climb, the installed k_cl = T/T_SLS = 0.95 × 0.705 = 0.67 that gives T_SLS/T = 1.5. The undercarriage and high‐lift devices are in a retracted position. The result is computed in Table 14.2.

Lift coefficient,

Using Eq. 14.14,

14.4.3 Cruise

Requirements. Initial cruise speed must meet the high‐speed cruise (HSC) at Mach 0.75 and at 41 000 ft (flying higher than bigger jets in less congested traffic corridors). Using Eq. 14.19, the result is computed in Table 14.3 (for the initial cruise aircraft weight use k = 0.972 in.).

In FPS at 41 000 ft:

In SI at 12 519 m altitude:

Equation 14.18 gives the initial cruise:

Equation 14.19 gives:

For the class of TFE731 engine data, Figure 13.4 gives the uninstalled k_cr = T/T_SLS = 0.23. Taking 4% installation loss during cruise, the installed k_cl = T/T_SLS = 0.94 × 0.23 = 0.222 that gives T_SLS/T = 4.5.

In FPS:

In SI:

Using Figure 11.2 for the drag polar, Table 14.3 gives the computed values for Bizjet cruise sizing.

14.4.4 Landing

From the market requirements, V_app = 120 knots = 120 × 1.68781 = 202.5 ft s⁻¹ (61.72 m s⁻¹). Landing C_Lmax = 2.1 at a 40° flap setting (from testing and CFD analysis). For sizing purposes, the engine is set to the idle rating, producing zero thrust. Using Eq. 14.22 the following is obtained.

In the FPS system, W/S_W = 0.311 × 0.002378 × 2.1 × (202.5)² = 63.8 lb ft⁻². In the SI system, W/S_W = 0.311 × 1.225 × 2.1 × (61.72)² = 3052 N m⁻². Because the thrust is zero (i.e. idle rating) at landing, the W/S_W remains constant.

14.5.1 Variants in the Family of Aircraft Designs

The family concept of aircraft design is discussed in previous chapters and highlighted again at the beginning of this chapter. Maintaining large component commonality (genes) in a family is a definite way to reduce design and manufacturing costs – in other words, ‘design one and get two or more almost free’. This encompasses a much larger market area and, hence, increased sales to generate resources for the manufacturer and nation. The amortisation is distributed over larger numbers, thereby reducing aircraft costs.

Today, all manufacturers produce a family of derivative variants. The Airbus320 series has four variants and more than 3000 have been sold. The Boeing737 family has six variants, offered for nearly four decades, and nearly 6000 have been sold. It is obvious that in three decades, aircraft manufacturers have continuously updated later designs with newer technologies. The latest version of the Boeing737–900 has vastly improved technology compared to the late 1960s 737–100 model. The latest design has a different wing; the resources generated by large sales volumes encourage investing in upgrades – in this case, a significant investment was made in a new wing, advanced cockpit/systems, and better avionics, which has resulted in continuing strong sales in a fiercely competitive market.

The variant concept is market and role driven, keeping pace with technology advancements. Of course, derivatives in the family are not the optimum size (more so in civil aircraft design), but they are a satisfactory size that meets the demands. The unit‐cost reduction, as a result of component commonalities, must compromise with the non‐optimum situation of a slight increase in fuel burn. Readers are referred to Figure 16.6, which highlights the aircraft unit‐cost contribution to DOC as more than three to four times the cost of fuel, depending on payload‐range capability. The worked‐out examples in the next section offer an idea of three variants in the family of aircraft.

14.5.2 Bizjet Family

Figure 14.4 shows the final configuration of the family of ; the baseline aircraft is in the middle. It proposes one smaller (i.e. 4–6 passengers) and one larger (i.e. 14–16 passengers) variant from the baseline design that carries 10–12 passengers by subtracting and adding fuselage plugs from the front and aft of the wing box. The baseline and variant details are worked out in Section 8.9. Table 14.4 gives the summary of the Bizjet variants.

The short variant engine is derated by about 17% resulting in slightly lower engine weight and the long variant engine is up‐rated by about 15% resulting in slightly higher engine weight.

14.6.1 Takeoff – Military Aircraft

Requirements. TOFL = 800 m (≈2600 ft) to clear 35 ft (10.7 m) at ISA + sea at NTC. The maximum lift coefficient at TO (20° flaps down and no slat) is taken as C_{Lmax_TO} = C_{Lstall_TO} = 1.85. Installation is taken at 3.5%.

Using Eq. (14.11a), the expression becomes (in FPS system)

Using Eq. (14.11b), it becomes (in SI system)

The result is computed in Table 14.5 and listed in tabular form.

14.6.2 Initial Climb – Military Aircraft

From market requirement, initial climb speed V = 350 kt = 350 × 1.68781 = 590.7 ft s⁻¹ and the required rate of climb, RC = 10 000 ft min⁻¹ (50 m s⁻¹) = 164 ft s⁻¹ (50 m s⁻¹). From the Adour 861 class engine data T_SLS/T ratio, k_cl = 1.06. The result is computed in Table 14.6.

Lift coefficient,

Using Eq. 14.15,

14.6.3 Cruise – Military Aircraft

Market Specification. Initial cruise speed and altitude is 0.75 Mach and 36 000 ft (most of training is takes place below the tropopause), for the initial cruise aircraft weight take, k = 0.975 in Eq. 14.14. The result is computed in Table 14.7.

In FPS

In SI,

Equation 14.18 gives initial cruise

Equation 14.19 gives.

In FPS

In SI

Once again, make a table (see Table 14.7) and draw a plot.

14.6.4 Landing – Military Aircraft

From the market requirements,

Landing C_Lstall = 2.5 at a 40° double slotted flap setting.

Using Eq. 14.22,

In FPS system,

In the SI system,

Because at landing the thrust is taken to be zero, the W/S_W remains constant.

14.6.5 Sizing for the Turn Requirement of 4g at Sea Level

In a way, turn sizing is relatively simple and here the parabolic drag polar AJT is used to make use of the equation derived in Chapter 15. The approximated AJT parabolic drag polar is given in Section 11.20.1 as follows.

The turn requirement is n = 4 at sea level (STD) day, but is stipulation for speed at which it can be achieved. It gives the opportunity to initiate sizing analyses in many ways different from these methods. The turn sizing is carried out by establishing the relationship between, wing loading, W/S_W, and thrust loading T_SLS/W at n = 4, the turn speed that has to be establish. Use Eq. (15.58) (derived in the next chapter) and evaluate at three speeds at Mach 0.5 (V = 558.25 ft s⁻¹), Mach 0.55 (V = 614.075 ft s⁻¹) and Mach 0.6 (V = 669.9 ft s⁻¹).

, on substituting the numeric values, the following is obtained.

On substituting the numerical values of ρ = 0.002378 slugs/ft³, k = 0.07 and C_DPmin = 0.0212.

where T is the installed thrust at maximum continuous rating.

Aircraft sizing thrust loading is expressed in terms of uninstalled sea level static thrust at maximum takeoff rating, T_{SLS_unins}. Maximum continuous rating is taken at 90% of maximum takeoff rating and installation loss is taken as 3.5%. That makes:

installed thrust at maximum continuous rating,

Tables 14.8–14.10 show the (T/W) values at three (W/S_W) cases of 50 lb ft⁻², 60 lb ft⁻² and 70 lb ft⁻² for the three speeds at Mach 0.5, 0.55 and 0.6.

W/SW (lb ft−2)	50	60	70
V2	311 587.2	311 587.2	311 587.2
	0.151 001	0.181 201	0.211 401 5
	0.155 794	0.129 828	0.111 281 1
T/W	0.306 795	0.311 029	0.322 682 6
TSLS/W	0.352 814	0.357 684	0.371 085

W/SW (lb ft−2)	50	60	70
	0.124 802	0.149 763	0.174 723 3
	0.190 006	0.157 082	0.135 718 6
T/W	0.314 808	0.306 845	0.310 441 9
TSLS/W	0.362 03	0.352 871	0.3570 082

W/SW (lb ft−2)	50	60	70
	0.104 874	0.125 849	0.146 824 1
	0.226 111	0.188 425	0.161 507 5
T/W	0.330 985	0.314 275	0.308 331 6
TSLS/W	0.380 633	0.361 416	0.354 581 4

Tables 14.8–14.10 give three lines with close values (T/W). Only the line for Mach 0.55 is plotted in Figure 14.5 as the middle one of the three. With the excess thrust available, AJT can turn at a higher n (see Chapter 15).

14.7.1 Single Seat Variants in the Family of Aircraft Designs

Military aircraft are no exceptions in offering variant designs depending on their mission role, in addition to the typical ‘payload‐range’ variation. The F16 and F18 have had modifications since they first appeared with an increasing envelope of combat capabilities. The F18 has increased in size. The BAE Hawk100 jet trainer has produced a single seat close support combat derivative, the Hawk200.

The CAS role aircraft is the only variant of the AJT aircraft (Figure 14.6). The details on how it is achieved with associated design changes are described next.

14.8.1 Civil Aircraft Sizing Studies

Aircraft sizing exercise freezes aircraft to a unique geometry with little scope to make variation as shown in Figure 14.3 for the Bizjet. In these kinds of studies, aircraft performance requirements from the Federal Aviation Administration (FAA) and specification (customer) are given as: (i) takeoff field length and climb gradients, (ii) initial en‐route climb, (iii) initial HSC capability and (iv) LFL and climb gradients at baulked landing are simultaneously satisfied with a matched engine and a unique solution emerges. To search for a compromising to offer a family of variants it depends on objectives, for example, the degree of component components can be retained.

It is a broad‐based analysis and could be extensive. Basically, it is an optimisation process to find the best value providing opportunities to make any compromise, if required. The easiest method is to take one variable at a time in a parametric search to find on how it is affecting other parameters. Normally, the variations are made in small steps to keep changes in the affecting parameters within the ranges of acceptable errors. In civil aircraft design, the study should continue to examining the objective function, say the DOC. Typically, the results are shown in carpet plots (Figure 14.7) studying up to as many as five variables [5]. This carpet plot shows four variables in one graph with mutual inter‐dependency. It shows how aircraft DOC is affected with aircraft drag change, fuel price and aircraft cost for an Airbus 320 class aircraft.

Apart from parametric method of searching there are theoretical methods available. The simpler ones are the Simplex and Gradient methods; those are used in the industry, but the parametric method is most popular.

14.8.1.1 Military Aircraft Sizing Studies

Aircraft sizing exercise freezes aircraft to a unique geometry with little scope to make variations as shown in Figure 14.3 for the Bizjet. The aircraft sizing exercise freezes aircraft to a unique geometry with little scope for variation as shown in Figure 14.5 for the AJT.

In aircraft design sensitivity can be made on determining individual component geometries, for example, for wing examine how its area, AR, thickness to chord ratio, sweep and so on, affects aircraft capabilities as shown in Table 14.11 for the AJT. An example of AJT wing geometry sensitivity study is given next involving what happens with small changes in wing reference area, S_W, aspect ratio, AR, aerofoil thickness to chord ratio, t/c, and wing quarter chord sweep, Λ_¼.

A more refined analysis could be made with a detailed sensitivity study on various design parameters, such as other geometrical details, materials, structural layout and so on, to address cost versus performance issues to arrive at a ‘satisfying’ design. This may require some local optimisation with full awareness that the global optimisation is not sacrificed. While a broad‐based MDO is the ultimate goal, dealing with large number of parameters in a sophisticated algorithm may not prove easy. It is still researched intensively within academic circles, but industry tends to use MDO in a conservative manner if required in a parametric search, tackling one variable at a time. Industry cannot afford to take risks with any unproven algorithm just because it is elegant bearing promise. Industry takes a more holistic approach to minimise costs without sacrificing safety but may compromise on performance if it pays.

14.9.1 The AJT

Military aircraft serves only one customer, the Ministry of Defence (MoD)/United States Department of Defense (DoD) of the nation that designed the aircraft. Front line combat aircraft incorporates the newest technologies at the cutting edge to stay ahead of potential adversaries. Its development cost is high and only few countries can afford to produce advanced designs. International political scenarios indicate a strong demand for combat aircraft even for developing nations who must purchase from abroad. This makes the military aircraft design philosophy different from civil aircraft design. Here, the designers/scientists have a strong voice unlike in civil design where the users dictate terms. Selling combat aircraft to restricted foreign countries is only one way to recover some finance.

Once the combat aircraft performance is well understood over its years of operation, consequent modifications follow capability improvements. Subsequently, a new design replaces the older design – there is a generation gap between the designs. Military modifications for the derivative design are substantial. Derivative designs primarily come from revised combat capabilities with newer types of armament along with all round performance gains. There is also a need for modifications, seen as variants rather than derivatives, to sell to foreign customers. These are quite different to civil aircraft variants, which are simultaneously produced for some time, serving different customers, some operating all the variants.

Advanced trainer aircraft designs have variants serving as combat aircraft in CAS. AJTs are less critical in design philosophy in comparison with front line combat aircraft, but bear some similarity. Typically, AJTs will have one variant in a CAS role produced simultaneously. There is less restriction to export these kinds of aircraft.

The military infrastructure layout influences the aircraft design and here the life cycle cost (LCC) is the prime economic consideration. For military trainer aircraft design, it is better to have a training base close to the armament practice arena, saving time. A dedicated training base may not have as long a runway as major civil runways. These aspects are reflected in the user specifications needed to start a conceptual study. The training mission includes aerobatics and flying with onboard instruments for navigation. Therefore, the training base should be far away from the airline corridors.

The AJT sizing point in Figure 14.7 gives a wing loading, W/S_W = 58 lb ft⁻² and a thrust loading, T/W = 0.55. It may be noted that there is a large margin, especially for the landing requirement. The AJT can achieve maximum level speed over Mach 0.88, but this is not demanded as a requirement. Mission weight for the AJT varies substantially; the NTC is at 4800 kg and for armament practice it is loaded to 6600 kg. The margin in the sizing graph covers some increase in loadings (the specification taken in the book is for the NTC only). There is a big demand for higher power for the CAS variant. The choice for having an up‐rated engine or to have an afterburner depends on the choice of engine and the mission profile.

Competition for military aircraft sales is not as critical as compared to civil aviation sector. The national demand would support the national product for producing a tailor made design with under manageable economics. But the trainer aircraft market can have competition, unfortunately sometimes influenced by other factors that may fail to bring out a national product even if the nation is capable of doing so. The Brazilian design Tucano was re‐engineered and underwent extensive modifications by Short Brothers plc of Belfast for the RAF in the UK. The BAE Hawk (UK) underwent made major modifications in the USA for domestic use.