DISTRIBUTION AND TRANSPORTATION

Author: GUERRERO GARCÍA CARLOS.
Year: 2004.
University: MÁLAGA [www.uma.es].
Place of defense: ETSI INFORMATICA.
Place of preparation: ETSI INFORMATICA.

Summary: The work is framed within the context of the theory of localization centers not deseados.En first instance is studied this problem from the standpoint of dominance, allowing us to reduce the region practicable. We have studied a model of localization centers containing unwanted, as individual cases, the criteria studied to ahora.Dicho model introduces weights and considers any norma.Para this approach is characterized a series where there is necessarily a solution to the problem in terms of equidistancias Weighted to poblaciones.Este package turns out to be finite when particularizamos to euclídea distance. It includes examples of the proven results and codes of the routines in MATLABA employed.

Author: RISCO MARTIN JOSE LUIS.
Year: 2004.
University: COMPLUTENSE DE MADRID [www.ucm.es].
Place of defense: FACULTAD DE CIENCIAS FISICAS.
Place of preparation: FACULTAD DE CIENCIAS FISICAS.

Summary: The model for optimization problems is a complex process composed of multiple phases, which include an extensive period of time to obtain workable solutions, especially when sisitema resolve is large in scale (tens of thousands of variables and constraints). Currently there is a lack of generic modeling tools in terms of optimization problems are concerned. In this thesis are defined formalism to build a graphical modeling system optimization problems that include:. Graphical Modeling optimization problems. With a modeling problem, automatically generate database that will hold the data of the problem, generate fictitious data that will test models and algorithms for solving these, as well as algorithms cleansing if the solver is not feasible solution . . Editors automatically generate charts that allow graphically the problem data modeling, allowing modify these neatly. Finally these formalities have been used to implement a publisher of optimization problems, including the above points and based on XML and Java. The implemented system has been used successfully over the past two years in different departments logistics company Repsol- YPF, managing and optimizing its logistics network, transport or more specifically:. Repsol Butano, Spain. Wholesale Trade YPF in Argentina. Gas YPF in Argentina. Lipigas in Chile.

Author: VENTURA MOLINA INMACULADA.
Year: 2004.
University: SEVILLA [www.us.es].
Place of defense: ESCUELA SUPERIOR DE INGENIEROS INDUSTRIALES.
Place of preparation: ESCUELA UNIVERSITARIA POLITÉCNICA.

Summary: Due to the great interest and awareness of the society on environmental issues, location of unwanted service carries a very important role in today. Taking into account these concerns, this report presents geometric optimization problems associated with the location of unwanted service, where the study of the geometrical properties of objects suggests using computational geometry tools for the design of algorithms that allow the efficient resolution of problems. It begins with a vision paronamica of existing results on some problems of continuous non-point when using the criterion optimization maximin, which allows minimizing the undesired effect of the service on the people most affected. It explores the optimal location of unwanted structures on the plane, as segments of fixed length and a polygonal chains elbow and lower bounds are looking for such problems. It introduces a diagram voronoi kind abstract diagram voronoi analcado; considers its version dynamics. Good topological properties of the diagram allows its construcicón with deterministic algorithms. This structure allows efficient solutions to problems of optimum location of connections between points and algebraic curves.

Author: MARTÍN-ANDINO BENITEZ RAMON JULIAN.
Year: 2005.
University: PONTIFICIA COMILLAS [www.upcomillas.es].
Place of defense: ESCUELA TÉCNICA SUPERIOR DE INGENIERÍA.
Place of preparation: E.T. SUPERIOR DE INGENIERÍA.

Summary: The "whip effect" is a phenomenon that hampers management of supply chains and that is a distortion of the growing demand conveyed by the players involved in managing the flow of products as we move away from the market. This paper uses engineering controls applied to models that simulate the comprotamiento of agents in inventory management. Similar methods have already been used to illustrate the key role that carries inventory management in loyalty demands conveyed to the players earlier in the chain. The model is referred to an agent who performs functions brokerage Do you buy and sell? To which employs a system OUT replenishment of inventory, which must determine the amount requested by considering that revieses its holdings in each period and that deliveries of its supplier suffer a delay known beforehand. According to the agent must calculate this, in each period, their needs in accordance with demand expected in the coming period, whether stationary or not, while maintaining a stock equivalent to the projected demand for the delay Provider . In analyzing the behavior of demand conveyed by this agent has been used various values of delay, as well as various methods of prediction (exponential smoothing, moving average and minimum mean square error) and model demand (autorregresivo order one, AR (1) and autorregresivo and moving average of order one, AR; A (1.1), in order to compare different situations. These results indicate that the distortion in demand, apart from the influences due to the formation of lots of purchase, it is essential to the form of inventory management. order to avoid, therefore, the spread of a growing demand distortion, should be established as a rule that the players do not seek immediate recovery of lost inventory as a result of an unexpected increase in demand. alternative that few managers dare do for the risks involved. One way to solve this problem is to consider replenishment orders received but not yet Do you channel inventory logistics) as part of the inventory available and allowing compensate the fall of physical inventory. The findings of the study show that when considering this form of management and by adjusting the parameters of predicting demand, an agent can maintain in the medium or long term demand for its suppliers, which for practical purposes, you can consider as that received its clients. The study is complemented by a criticism of how to measure the "whip effect" as a relationship of the variances of the demands transmitted and received, for the reason that this phenomenon manifests itself not only as an extension of the variance in demand, but it is causing other problems, not reflected in this relationship, such as amendments to the seasonality of demand and delays in responses by agents, aspects which may have serious consequences on the decisions taken by the agents.