No margin for error in Space.. (an article about chaos theory).

 
Chaos theory & Space
Read my online article on Chaos theory I welcome your comments. Thanks. Carl.

http://www.suite101.com/discussion.cfm/101_fun_stuff/32939/latest/2

Subject: Chaos theory and successful innovations
PREDICTING SUCCESSFUL INNOVATION FOR SPACE EXPLORATION AND COLONIZATION

by Carl Zimmerman

"...The opportunity to create and build our own worlds from scratch..."

Imagine this dialogue between a parent and teenager in the not too distant future:

Parent: Why does your generation want to leave "mother" Earth and colonize Mars? They will have to take great risks and pay enormous costs in lives lost and material destruction if initial attempts fail.
Teenager: Colonizing Mars gives us the opportunity to create and build our own worlds from scratch. We don't have to waste our lives dealing with problems inherited from previous generations on Earth. We believe that justifies the risks and costs.
Parent: What problems?
Teenager: Ecological disasters and people conflicts--racial, ethnic, sexual and personality--to name a few. In the Mars colonies, we can't afford these problems. Survival depends on innovation and cooperation based on contribution, not personal style. For example, when an airlock problem is detected, there's no time for debate. We'll listen to the autistic airlock expert who can do nothing but fix airlocks and has a 100% history of success. We can't afford failure.

This imaginary dialogue depicts the direction that the U.S. and many other developed countries are taking today--space exploration and colonization. Currently, the main goal is to develop new technologies that will improve the quality of life on Earth, but as the dialogue suggests, the colonists may "fall in love" with living on other planets. This will especially happen if the cost of solving major problems, such as ecological disasters, on Earth becomes prohibitive. Models which predict the innovations that are likely to succeed will improve the quality of life on Earth as well as for our colonists on distant planets.

In some previous innovations, such the airplane, the consequences of early failure were limited (e.g., a biplane crashing on the side of a barn). For innovations needed for space exploration and colonization, the costs of failure would be enormous, including, for example, ecological disasters during testing on Earth and destruction of entire colonies on other planets. Predicting changing needs and their impact on populations is required to predict successful innovations.

Ideally, in order to produce a successful innovation, we must predict the essential needs of the space colonies, and identify what does and does not exist today that will meet these needs. The "what does not exist today" will tell us which innovations will have to be developed. However, nnovations based solely on "guesstimates" of future needs and technologies run a high risk of costly failure if these conflict with actual future needs. Innovations that reflect only the present state-of-the-art without any assumptions of the future will almost certainly fail. Innovations should be based on the right balance between what we assume is needed tomorrow and what we know will work today. How can this be accomplished? The following
will help: If we depict a region and its components, including the population and, for example, their possessions and architecture, after classifying them an evolutionary model would emerge, suggesting a "cast of characters" and behavior over time. To understand the system and how its future may be affected by some event(s), choice(s) and/or policy(ies), we would build a mathematical model of the system as it exists at a particular time, and describe the processes that will increase or decrease its different components.

We could apply the traditional approach of physics, that is, identify the components and the interactions operating on these, both to and from the outside world. From this, we can predict the functioning of the system at that time based on the causal relations between components, which we presume are present.

However, the predictions can be correct for as long as the taxonomy (classifications) of the system remains unchanged (Aristotle fans, please take note). The mechanical model of deterministic equations for a given time cannot produce new types of objects and variables. Its predictions will only be valid until some moment, unpredictable within the model, when there is an adaptation to innovation, and new behavior emerges. Consequently, we need models that not only predict what future needs will be, but when they will change and what change in population behavior will result.

Equilibrium does not exist in the real world

The basis of scientific understanding has traditionally been the mechanical model. In addition to focusing on causal relations between components at a particular time, it is assumed that this system has run itself to equilibrium, so that the correspondence between object and model is made through balance of variables at equilibrium. In economics, for example, equilibrium is represented by optimal utility for the actors, where consumers minimize cost of goods and services, and producers maximize profits. This approach assumes that all the actors know what they want and how to get it, and are doing what they want given the choices available. However, we know that equilibrium does not exist in the real world.

Today, system dynamics is available to replace this oversimplified static approach, which is based on unrealistic assumptions, in developing models for rational prediction of population behavior, need and innovation. These are based on the following parameters:
Values of external factors, reflecting the "environment" of the system, such as temperature, climate and soils, and expected changes in these over time. Values corresponding to population "performance" due to their internal characteristics, such as technology, and knowledge level and application, and expected changes in these over time. Change comes from within the system "If it's not in the heart, it's not in the head." (Salesmanship proverb).

"The clouded mind sees nothing." (The Shadow, a fictional character)

Since system dynamics models required for predicting successful future innovations are concerned with evolution, these must be model systems in which adaptive and structural changes can occur. The internal characteristics of the actors change endogenously, and new variables and mechanisms of interaction can appear spontaneously within the system itself, leading to a changed taxonomy.

In order to make the model work, it must be simplified. This can be accomplished with two assumptions:

1. Events occur at an average rate.
2. All individuals of a given type are identical and of an average type.

Chaos equations are better predictors of successful future innovations than bell curve equations Errors introduced by assumption #1 (above) can be corrected by probabilistic dynamics, which assumes that all individuals are identical to an average type, but that events of different probabilities occur. Consequently, probabilistic dynamics includes sequences of events that correspond to runs of good or bad "luck" and their probabilities.

Systems with nonlinear interactions between individuals eliminate the concept of a simple, constant trajectory. Evolution of the system will be described by a probability distribution that gradually changes in shape from a single modal "bell curve" (sharply peaked and centered on a mean) to spreading and splitting into a multi-modal distribution with peaks that correspond to different attractors of the dynamics (attractor=position where the dynamic system converges). These attractors could be point or cyclic, but most likely, based on previous experience, will be chaotic.

Since unpredictable runs of good or bad "luck" will occur, a precise trajectory of the system does not exist for predicting future behavior. Also, these deviations from the average rate of events means that a system can "tunnel" through apparently impassable potential barriers, and can switch between attractor basins and explore the global space of the dynamic system in a manner that the system cannot predict.

Adaptations For Space Colonization

Since the innovations for future space exploration and colonization will be dynamic systems, the relationships of the system variables will be nonlinear. Consequently, we expect chaos equations to provide very useful descriptions of relationships for mission-critical biological, ecological and economic systems. Already electronics hardware such as solid state lasers, oscilloscopes and analog computers make extensive use of chaos equations. We've only scratched the surface of their potential.

In future issues, Carl Zimmerman will expand on the promising applications of chaos theory and its advantages for future pioneers.

About The Author

Carl Zimmerman is a techno-marketing writer in the U.S. for a major global manufacturer of animal nutrition and health products. He has a B.S. degree in chemistry and an M.B.A. in marketing.