How Does an Aircraft Happen ? (The industrial Design Loop)

A brief outline of the steps in the structured design process

1. Problem definition

The problem definition portion of the design process is very important step in that improper problem formulation leads to inefficient use of time, a tremendously valuable commodity. Usually, the problem definition step in an industrial design endeavor consists of a negotiation session between the engineer and his boss concerning the details of the "real" problem to be considered in the impending design effort.

2. Time allotment and schedule development

The design process is complex and lengthy with many perennial interactions. Time allotted for a design must be negotiated with management and schedules generated to ensure that each step contained in the design process is given a fair share of the time available. Short changing any of the steps can easily lead to an inferior product.

3. Payoff function definition

The end result of any design study must be a "best" solution. The process used to determine the "best" normally includes the generation of many nominal solutions from which a "best" is obvious. The payoff function is the mechanical used to compare solutions and judge one superior. Payoff functions are not always obvious and are quite often defined by means of an engineer/management negotiation session. The lack of a payoff function precludes a successful design process and different payoff functions usually lead to different solutions.

4. Establishment of design constraints

Constraints represent limitations placed on the set of admissible solutions in any design problem. The validity and practicality of a design process solution are ultimately governed by the constraints placed on the system. Constraints can be complex and numerous and must be carefully specified. Over constraining a system will decrease the flexibility of the design process leading to inferior solutions. Under constraining usually leaves a weakness that will be exploited by the design process yielding impractical solutions.

5. Understanding the design space

How big is the box within which you "live" during the design process? How much "rope" have you given yourself or how much variation in a nominal configuration can you make in the search for the "best" configuration? Understanding the design space usually ends up being a review session in which the designer and management decide if the proposed design problem formulation is properly scoped. Can it be completed in the allotted time? Is it too complex? Is it oversimplified. Control variables are carefully scrutinized in this step.

6. The nominal solution algorithm definition

In this step one generates an algorithm that can be used to generate arbitrary solutions in the design space. The first nominal solution represents a "first look" at what the ultimate solution will resemble and should never be used as the solution in a design procedure. Rather the first nominal solution is merely the beginning point for an interactive process that will culminate in a "best" solution.

7. The control variable/constraint boundary sensitivity study

Some control variables may be quite effective in terms of causing changes in the payoff function while others may generate but negligible changes. The control variable sensitivity study is simply a study in which one evaluates the effects of each of the control variables on the payoff function to determine their order of importance. "Weak" control variables are generally discovered and discarded in this step.

For design problem solutions that exist on a constraint boundary, it is important to determine the sensitivity of the payoff function to small changes in location of the constraint boundary. Perhaps it would be advisable to change the constraint boundary location if great improvements in payoff function can be realized.

8."Marching" from nominal to "Best" solution

During this step one formulates a technique that will allow systematic changes in the control variable such that each subsequent solution is better and simultaneously satisfies all constraints. The process begins with the first nominal solution and proceeds with subsequent solutions converging to a "best" solution

9. The presentation

One gets paid for that which he sells and nothing more ! The presentation to management is a very important step in the design process. If you can't sell your design, you've wasted a lot of time and money. Design reviewers (management side) are typically hard-to-sell types since they tend to lose much more than the designer in the case of inferior configurations! Hence, the presentation must be very carefully developed and prepared if success is to be realized.

The following schedule delineates the various groups involved, the extent of their involvement and the manpower required to complete each task. A detailed plot of the manpower loading is included. It is noted here that the example is taken from a real life case in the general aviation business.

	DJFMAMJJASOND	JFMAMJJASOND	JF
2-D Airfoil Design	**** 3.0
2-D Transonic Tests	*~** 1.5
Stability & Control Group	*** 2.5
Analytical 3-D Aerodynamic Group	****** * 6.0	************** 5.25
3-D Low Speed Wind Tunnel Group	**********~ 4.25	****~*******~** 5.0 6.0 3.0
3-D Transonic Wind Tunnel Group	***********~* 4.0	****~******~*** 5.5 5.5 3.0
Simulator Group		********* ** 7.5 6.0	**
Performance Group	******** 3.0	** 4.0	**
Preliminary Air Loads Group		** ** 1.5 1.5

Note: ~ implies wind tunnel entry period

Manpower loading for the (2+) - year period

5

*

*

*

*

4

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

3

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

2

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

1

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

0

D

J

F

M

A

M

J

J

A

S

O

N

D

J

F

M

A

M

J

J

A

S

O

N

D

J

F

Schedule Item	Configuration details for study and/or specification
2-D Airfoil Design	Wing root and tip airfoil geometries are generated analytically and tested in inexpensive subsonic wind tunnel tests to verify desired aerodynamic characteristics. The crucial information generated by this group consists of wing control station airfoil ordinate distributions which ultimately define chordwise camber distributions and max. t/c values.
2-D Transonic Wind Tunnel Test Group	Transonic wind tunnel tests on all wing control station airfoils are undertaken to ensure that desired transonic aerodynamic characteristics are realized. The most important characteristic is the drag divergence Mach No. Failure to generate the necessary Mach No. Performance requires continued airfoil design activity.
Stability and Control Group	The activities of this group include the following: (1) Type and size of roll control devices (spoiler, aileron, combination of spoiler and aileron ?) (2) Horizontal tail and elevator design (a) Surface size for each of horizontal stabilizer and elevator (b) Horizontal stabilizer location (tail length) (c) Horizontal stabilizer surface profile geometry (3) Vertical tail and rudder design (a) Surface size for each of vertical stabilizer and rudder (b) Vertical stabilizer location (c) Vertical stabilizer profile geometry
Analytic 3-D Aerodynamic Studies Group	(1) Details of the final 3-D wing planform (chord and t/c distributions along the span) (2) Wing twist distribution for good stall and high drag divergence Mach No. (3) Inboard wing decambering, incidence, and fillet studies for minimization of wing root interference flow problems. (4) Engine pod/fuselage or pod/wing shaping (matching) to minimize local flow field separation. (5) Other 3-D problems.
3-D Low Speed Wind Tunnel Group	(1) Build low speed wind tunnel model (2) Test model to investigate; (a) Subsonic aerodynamic characteristics (zero lift values for lift, drag, and moment coefficients;stability derivatives) (b) Max. lift coeff. (c) Control surface performance (d) Inlet and exit air flow fields (e) Flow field breakdown on wing at stall (f) Local flow field separation problems
3-D Transonic Wind Tunnel Group	(1) Build transonic model (2) Test model to investigate; (a) Transonic aerodynamic characteristics in general (b) Many drag divergence tests (c) Shock induced separation problems that occur (d) Transonic control surface performance (e) Transonic inlet and air flow fields
Simulator Group	(1) Handling qualities for the proposed aircraft (2) Ensure reasonable control for the aircraft at all points on the V-n diagram (3) Develop flow field conditions at all critical flight situations
Performance Group	Verify or predict new performance indices as data becomes available. Performance indices to include: (1) Max. Level flight velocity (2) Max. Static rate of climb (3) Max. Endurance (4) Max Range (5) Service ceiling (6) Takeoff performance (7) Etc.
Preliminary Air Load Group	Provide loads input to the structures group so that they may provide a minimum weight structure for the aircraft.