Tekhné

6

Authors: César Pérez, Luis Gutiérrez

Un área de aplicación de los modelos de redes es la planificación, programación y gestión de proyectos. Cuando las actividades que componen el proyecto tienen duraciones conocidas y deterministicas existe una técnica conocida para calcular la duración total del proyecto. Para ello se aplica la técnica del camino crítico (CPM por sus siglas en inglés). Cuando las actividades del proyecto tienen duraciones inciertas se calcula su duración por medio de la "Técnica de revisión y evaluación de programas" (PERT por sus siglas en ingles). Sin embargo la metodología que se acostumbra util izar cuando hay incertidumbre en las duraciones de las actividades conduce, con frecuencia, a errores en la estimación de la duración del proyecto, pues deja a un lado el hecho que el resultado final es una distribución de frecuencias. Otro supuesto que se suele hacer es que las actividades tienen duraciones independientes. Este trabajo muestra un caso de estudio sobre el tipo y magnitud de los errores de estimación de la duración de un proyecto cuando la duración de las actividades es estocástica y dependiente. El procedimiento seguido hace uso de la simulación de eventos discretos para representar la duración de las actividades y su relación de dependencia. La conclusión es_que los resultados que se obtienen dependen dt:' lós supuestos que se hagan en el modelo de red e~ .. del proyectq y afecta la ' . . . __ t ,, ll· estimación de la duración del proyecto

13

Authors: Mauricio Paletta

T e prediction of chaotic systems has been and continues to be a subject with great scientific interest. This is because the benefits that can be obtained knowing the future behaviors of this kind of systems based on previous known data. However, the search ior solution strategies with an acceptable error margin prese ts sorne levels of complexity that must be treated convenie tly, based on the differe ce between the predicted values and the real values. Enoug is to say, for instance, specific problems like climate and seismic predictions and the stock market, to realize of ti e benefit that would mean to have a good strategy based on prediction. Since these problems are related to managing knowledge, most attempts to obtain a predictive tool with acceptable results are related to technical heuristics. In this reg ard, Neural Networks is o e of the most commonly used, specifically by configuring Multi-layer Perceptron and Radial Based Functions models. We must ask ourselves then: Which one of these models is most appropriate to salve this kind of problems? Can we guarantee a tool of prediction with low error margins using this kind of technique? This article presents a comparative study oi patterns of Neural Networks above, responding to the previous two questions as well as providing concl usions that are i tended to assist the scientific world to address a specific prediction problem with this technique.

15

Authors: David Sánchez

Veh icle traffic studies use Nagei-Schreckenberg (NaSch) based simulations that produce significantsize information files. In arder to save that information, common compression techniques are not enough to achieve high compression in mini mal time. The present article demonstrates how to compress those syntheticvehicul ar-traffic flows using binary representation (data modeling) and .zip codification (generic data modeling with entropy-coding) to get a 12:1 compression ratio.

24

Authors: Antonio José Sucres Salas, Jorge Molero, Leonardo Contreras

Production processes are diverse, varied and complex; but all of them have generally the same pattern outlined as follows: Input (Raw Material) - Process of Conversion (products in process) - Output (finished products). [01] Production processes represent the competitive advantage of the organizations. Their control and the certainty of their quality assurance are essential at all stages of development and operation. The continuous qualitvimprovement and the qua/ity manaqement are STRA TE G/ ES OF PROGRESS BY EXCELLENCE. lnformation systems and information technologies provide different tools to maximize production and allow the satisfaction of collective needs for consumers as well producers of goods and services. The requirements generated by the continuous improvement and the ones generated by the clients promote changes, especially in production processes and control mechanisms, in this way the organizations demand for faster maintenance of the processes, at low cost and without significant or traumatic impacts in their operations. In the present investigation is designed an objectoriented prototype (OOP) of an information system to assess the product quality based on the design specifications of the product, generated by a QFD integrated with a process statistical control method; the system is completely configurable and general, such that, whatever the production process to apply QFD, it could be set and get functionality for assessing the quality regarding of that production process, without modifying or creating a new system or by modifying the production process, the system of quality control that controls it only requires new parameterizations and non maintenances that imply new developments in the information system.