Monday, November 29, 2010

What Limits a System's Performance & What we might do?

 As a child I loved watching the trailing vortices of beautiful white plumes in the wake of commercial aircrafts flying majestically in the sky. I watched them as long as I could till the plumes gradually dissolved their presence into the blue sky.

Such vortices are strong, long and lingering. But, what might be the effect of such beautiful vortices on the system’s behaviour?

By the way, what system are we talking about?

The aircraft is a part of a much bigger system than we might immediately notice. The bigger system consists of the airport, the number of planes ready to take off or land, number of runways, air traffic control, signals, lights etc. These elements are well within our control or almost. However, some of the elements like atmospheric temperature, humidity, visibility are not within our control at all. 

We can easily understand that all these elements are interdependent on each other and produce a range of possible interactions between them. This we call as the "essence" of a system. The presence of this essence produces a range of behavior of a system that emerges in somewhat unpredictable or random manner. We call such possible and uncertain range or types of behavior (since we can't precisely predict) the 'emergence' of the system. Because of the 'essence' the 'emergence' takes place.

Obviously the essence and the emergence of a system effectively determines the behavior of the system and places an upper limit on a system’s performance.

Let us see how that might happen in the case of vortices generated by the interaction of an aircraft's movement and the wind?

These vortices stretch for miles across the sky trailing behind a large aircraft and endanger the planes that follow it or go across it by inducing drastic rolling moments, which might lead to the failure of the aircraft that follows or crosses the path of these vortices. Thus persistence and length of such vortices would govern the frequency of take off and landing of other planes in an airport. So, the 'essence' and the 'emergence' of the system determine the airport capacity and the limits of the system’s functioning and performance.

Now, if we want to improve the productivity of the airport in terms of number of take offs and landings we immediately recognize the inherent ‘imperfection’ in the given system. Recognition of such inherent imperfection (in terms of the objecive we would like to achieve) would automatically lead us to innovate or design solutions to either eliminate the ‘imperfections’ or avoid them in case imperfections can't be eliminated totally or monitor them if we have no other way but to live with them.

Such solutions might take the form of a concentrated research work to alleviate these swirling wakes or shorten the length of the wake or we might think of building or using alternative runways instead of relying on one or we might think of increasing the air traffic corridors etc.  In fact, most major airports in the world have now increased the number of runways and the air traffic corridors to improve system performance.

That might be all well for running commercial aircrafts. But what might we do to save aircrafts from accidents during an air show?


During such airshows aircrafts would be forced to fly parallel to each other to avoid the dangerous trap of vortices (as shown in the photograph).Perhaps we need very skilled and experience pilots to perform such feats.

Summary:

The steps that we followed were::

1. Observe the system as a whole (what are the elements that make the total system)

2. Understand the range of possible 'essence' (number of possible interactions happening over time)

3. Understand the set of related 'emergences' or behavior of the system (in relation to the essence)

4. Understand the present objective function or purpose of the system as of now and in the future.(since objective function/purpose changes over time)

5. Identify the inherent 'imperfection(s)' that limits the system to achieve the present objective.

6. Find or design solutions to eliminate the imperfectons or avoid the imperfection if the imperfection can be partially eliminated or monitor the imperfection if it is not possible to eliminate imperfection(s) at all.(all imperfections can't be solved and might have to be lived with, but we may do that in a better way rather than give up).

7. Observe changes in the system over time (in terms of essence, emergence and objectives)

8. Set new objectives and goals to harness uncertainty for productive use.

9. Respond with new solutions and check performance and/or behavior of the system over time.

Posted via email from dibyendu's posterous

No comments: