sábado, 15 de mayo de 2010

MATHEMATICAL APPROXIMATION


MATHEMATICAL APPROXIMATION

Numerical approximation is defined as a figure reoresenta to a number whose exact value is the twelfth to the extent that the figure was closer to the exact value, it will be a better approximation of that number.






SIGNIFICANT FIGURES

The significant figures (or significant digits) represent the use of a level of uncertainty under certain approximations The use of these considers the last digit of approach is uncertain, for example, to determine the volume of a liquid using a graduated cylinder with a precision of 1 ml, implies an uncertainty range of 0.5 ml. It may be said that the volume of 6ml of 5.5 ml will be really to 6.5 ml. The previous volume is represented as (6.0 ± 0.5) ml. For specific values closer would have to use other instruments of greater precision, for example, a specimen finest divisions and thus obtain (6.0 ± 0.1) ml or something more satisfying as the required accuracy.



ACCURACY AND PRECISION



Accuracy refers to the dispersion of the set of values from repeated measurements of a magnitude. The lower the spread the greater the accuracy. A common measure of variability is the standard deviation of measurements and precision can be estimated as a function of it. Accuracy refers to how close the actual value is the measured value. In statistical terms, accuracy is related to the bias of an estimate. The smaller the bias is a more accurate estimate. When we express the accuracy of a result is expressed by the absolute error is the difference between the experimental value and the true value.





NUMERICAL STABILITY


In the mathematical subfield of numerical analysis, numerical stability is a property of numerical algorithms. Describe how errors in the input data are propagated through the algorithm. In a stable method, errors due to approximations are mitigated as appropriate computing. In an unstable method, any error in the processing is magnified as the calculation applicable. Methods unstable quickly generate waste and are useless for numerical processing.





CONVERGENCE



In mathematical analysis, the concept of convergence refers to the property they own some numerical sequences tend to a limit. This concept is very general and depending on the nature of the set where the sequence is defined, it can take several forms.





fuente
  • George E. Forsythe, Michael A. Malcolm, and Cleve B. Moler. Computer Methods for Mathematical Computations. Englewood Cliffs, NJ: Prentice-Hall, 1977. (See Chapter 5.)
  • William H. Press, Brian P. Flannery, Saul A. Teukolsky, William T. Vetterling. NumericalRecipes in C. Cambridge, UK: Cambridge University Press, 1988. (See Chapter 4.)

MATHEMATICAL MODEL





MATHEMATICAL MODEL


A product model is an abstraction of a real system, eliminating the complexities and making relevant assumptions, applies a mathematical technique and obtained a symbolic representation of it.





A mathematical model comprises at least three basic sets of
elements:

  • Decision variables and parameters
The decision variables are unknowns to be determined from the model solution. The parameters represent the values known to the system or that can be controlled.

  • Restrictions
Constraints are relations between decision variables and magnitudes that give meaning to the solution of the problem and delimit values feasible. For example if one of the decision variables representing the number employees of a workshop, it is clear that the value of that variable can not be negative.

  • Objective Function
The objective function is a mathematical relationship between variables decision parameters and a magnitude representing the target or product system. For example if the objective is to minimize system costs operation, the objective function should express the relationship between cost and decision variables. The optimal solution is obtained when the value of cost is minimal for a set of feasible values of the variables. Ie there to determine the variables x1, x2, ..., xn that optimize the value of Z = f (x1, x2, ..., xn) subject to constraints of the form g (x1, x2, ..., xn) b. Where x1, x2, ..., Xn are the decision variables Z is the objective function, f is a function mathematics.


HOW TO DEVELOP A MATEMATICAL MODEL

1. Find a real world problem.

2. Formulate a mathematical model of the problem, identifying variables (dependent and independent) and establishing hypotheses simple enough to be treated mathematically.

3. Apply mathematical knowledge that has to reach mathematical conclusions.

4. Compare the data obtained as predictions with real data. If the data are different, the process is restarted.




CLASSIFICATION OF MODELS

  • Heuristic Models: (Greek euriskein 'find, invent'). Are those that are based on the explanations of natural causes or mechanisms that give rise to the phenomenon studied.
  • Empirical models: (Greek empeirikos on the 'experience'). They are using direct observations or the results of experiments studied phenomenon.

Mathematical models are also different names in various applications. The following are some types to which you can adapt a mathematical model of interest. According to its scope models:

  • Conceptual models :Are those that reproduce by mathematical formulas and algorithms more or less complex physical processes that occur in nature.
  • Mathematical model of optimization :Mathematical optimization models are widely used in various branches of engineering to solve problems that are by nature indeterminate, ie have more than one possible solution.


CATEGORIES FOR ITS APPLICATION


For use commonly used in the following three areas, however there are many others such as finance, science and so on.

  • Simulation: In situations accurately measurable or random, for example linear programming aspects precisely when, and probabilistic or heuristic when it is random.
  • Optimization :To determine the exact point to resolve any administrative problems, production, or other status. When the optimization is complete or nonlinear, combination, refers to mathematical models little predictable, but they can fit into any existing alternative and approximate quantification.
  • Control: To find out precisely how is something in an organization, research, area of operation, etc..

fuente:

  • http://www.investigacion-operaciones.com/Formulacion%20Problemas.htm
  • Ríos, Sixto (1995). Modelización. Alianza Universidad.