Le Global Sourcing est aujourd'hui en pleine expansion car il offre aux entreprises une source potentielle de compétitivité dans un environnement de plus en plus concurrentiel. Néanmoins, il génère aussi une complexification des flux et une fragilisation de la Supply Chain Globale. La problématique du Global Sourcing est vaste et touche les différents niveaux de décision de l'entreprise. Pour cela nous nous sommes focalisés dans ce travail sur les aspects tactiques et opérationnels de ce domaine. Nous avons abordé ainsi diverses questions : Quels leviers d'action pour un pilotage efficace des flux en approvisionnement lointain? Comment sécuriser les approvisionnements lointains dans le contexte industriel actuel ? Les politiques classiques de pilotage de flux sont-elles suffisantes pour les approvisionnements lointains ? En collaboration avec les partenaires industriels de la Chaire Supply Chain de l'Ecole Centrale Paris, nous avons abordé différentes facettes de cette problématique. Nous nous sommes intéressés tout d'abord à la prévision comme élément nécessaire au pilotage des flux lointains et nous avons proposé une méthodologie de sélection et de mise à jour de méthodes de prévision. Les délais longs en approvisionnement lointain font que les erreurs de prévision s'amplifient, ce qui nous a amenés à étudier l'erreur prévisionnelle. Nous avons proposé dans ce sens une modélisation fine de cette erreur et de son évolution en fonction de l'horizon temporelle de la prévision. Dans la dernière étape de ce travail, nous avons utilisé cette modélisation de l'incertitude pour piloter efficacement les flux lointains. Nous avons montré sur des cas réels issus de l'entreprise PSA l'efficacité de la méthode proposée en termes de respect du niveau de service avec un niveau de stock largement inférieur aux méthodes classiques. / Global Sourcing is becoming a common practice in industrial activities since it offers companies opportunities to improve its competitiveness in an increasingly competitive business environment. At the same time, it makes the flows more complex and the supply chain more fragile. Global Sourcing thus gives rise to a wide range of issues and impacts different levels of decision making. To address such a problem, we focus on tactical and operational decision making. We attempt to answer a variety of questions: What are possible actions for flow management in global sourcing? How to secure the procurement in the current industrial context? Are classical flow management policies also efficient in global sourcing? In collaboration with the industrial partners of the Chaire Supply Chain at Ecole Centrale Paris, we consider different problems. Firstly, we are interested in demand forecasting, an essential element for flow management in global sourcing and proposed a methodology to select an appropriate forecasting method and to update it dynamically. The fact that the lead times are long in global sourcing makes the forecast less reliable and less and less reliable when the forecast horizon increases, which requires an evaluation of the forecast accuracy. We propose a detailed model of the forecast accuracy and its evolution with time horizon involved. As the last step of the work, this forecast accuracy model is applied to a real life flow management problem in global sourcing. The case study carried out based on real life data from PSA demonstrates a clear superiority of the proposed method over existing ones in terms of both service level and inventory level.

Long-range demand planning and capacity management play an important role for policy makers and airline managers alike. Each makes decisions regarding allocating appropriate levels of funds to align capacity with forecasted demand. Decisions today can have long lasting effects. Reducing forecast errors for long-range range demand forecasting will improve resource allocation decision making. This research paper will focus on improving long-range demand planning and forecasting errors of passenger traffic in the U.S. domestic airline industry. This paper will look to build upon current forecasting models being used for U.S. domestic airline passenger traffic with the aim of improving forecast errors published by Federal Aviation Administration (FAA). Using historical data, this study will retroactively forecast U.S. domestic passenger traffic and then compare it to actual passenger traffic, then comparing forecast errors. Forecasting methods will be tested extensively in order to identify new trends and causal factors that will enhance forecast accuracy thus increasing the likelihood of better capacity management and funding decisions.

Studies show teenagers are influenced by television. This study tested the hypothesis that students can learn meteorology by viewing a weather forecast on Music Television (MTV). MTV was used because it is the network watched most by adolescents. Two surveys were administered to 175 high school students along with a DVD showing a weather forecast for MTV’s Spring Break. Half of the sample group was told the forecast was for MTV and the other was told it was for Channel One. Results showed that there was no statistically significant difference between MTV and Channel One in information recalled from the forecast. Results also showed the White student population recalled more information from the weather forecast format than the other races surveyed in the study. Additionally a series of concert forecasts was pilot-tested on 15 bands with very positive and promising feedback.

There are a large number of investors and companies reliant upon analyst earnings forecasts. Missing analyst forecasts can have a massive impact on share price and investors often look to these values to make decisions regarding future investment decisions. However, there has been a great deal of speculation about these forecasts and especially the error associated with them. With the threat of an impending recession, it is important to know the reliability of forecasts during times leading up to recessions. More specifically, this study aims to see how the level of error associated with analyst earnings forecasts change leading up to recessions and whether or not they should be relied upon as heavily during these times.

Transportation infrastructure improvement projects are typically huge and have significant economic and environmental effects. Forecasts of demand of the facility in the form of traffic level help size the project as well as choose between several alternatives. Inaccuracy in these forecasts can thus have a great impact on the efficiency of the operational design and the benefits accrued from the project against the cost. Despite this understanding, evaluation of traffic forecast inaccuracy has been too few, especially for un-tolled roads in the United States. This study, part of a National Cooperative Highway Research Program (NCHRP) funded project, bridges this gap in knowledge by analyzing the historical inaccuracy of the traffic forecasts based on a database created as part of the project. The results show a general over-prediction of traffic with actual traffic deviating from forecast by about 17.29% on an average. The study also compares the relative accuracy of forecasts on several categorical variables. Besides enumerating the error in forecasts, this exploration presents the potential factors influencing accuracy. The results from this analysis can help create an uncertainty window around the forecast based on the explanatory variables, which can be an alternate risk analysis technique to sensitivity testing.

In this paper two different models for forecasting the number of monthly departing passengers from Sweden to any international destination are developed and compared. The Swedish transport agency produces forecasts on a yearly basis, where net export is the only explanatory variable controlled for in the latest report. More profound studies have shown a relevance of controlling for variables such as unemployment rate, oil price and exchange rates. Due to the high seasonality within passenger flows, these forecasts are based on monthly or quarterly data. This paper shows that a seasonal autoregressive integrated moving average model with exogenous input outperforms the benchmark model forecast in seven out of nine months. Thus, controlling for oil price, the SEK/EUR exchange rate and the occurrence of Easter reduces the mean absolute percentage error of the forecasts from 3,27 to 2,83 % on Swedish data.

The purpose of this research project was to evaluate the techniques outlined in the National Cooperative Highway Research Program Technical Report 255 Highway Traffic Data for Urbanized Area Project Planning and Design (NCHRP-255), published in 1982 by the Transportation Research Board. This evaluation was accomplished by using a regional travel demand forecast model calibrated and validated for the year 1990 and developing a highway forecast for the year 2000. The forecasted volumes along the Capital Beltway (I-495/I-95) portion located in the State of Maryland were compared to observed count data for that same year. A series of statistical measures were used to quantitatively evaluate the benefits of the techniques documented in NCHRP-255. The primary research objectives were: • To critically evaluate the ability of a regional travel demand forecast model to accurately forecast freeway corridor volumes by comparing link forecast volumes to the actual count data. • To evaluate and determine the significance of post-processing techniques as outlined in NCHRP-255. The most important lesson learned from this research is that although it was originally written in 1982, NCHRP-255 is still a very valuable resources for supplementing travel demand forecast model output. The "raw" model output is not reliable enough to be used directly for highway design, operational analysis, nor alternative or economic evaluations. The travel demand forecast model is a tool that is just part of the forecasting process. It is not a turn-key operation, and travel demand forecasts cannot be done without the application of engineering judgment. / Master of Science

This study develops a method for estimating confidence intervals surrounding futures based forecasts of natural gas prices. The method utilizes the Barone-Adesi and Whaley model for option valuation to "back-out" the market's assessment of the annualized standard deviation of natural gas futures prices. The various implied standard deviations are then weighted and combined to form a single weighted implied standard deviation following the procedures outlined by Chiras and Manaster. This option implied weighted standard deviation is then tested against the more traditional "historical" measure of the standard deviation. The paper then develops the procedure to transform the weighted standard deviation and futures price into a price range at the option expiration date. The accuracy of this forecast is then tested against 15 and 30 day average forecasts. / Master of Arts

This thesis compares the simple Autoregressive (AR) model against the k- Nearest Neighbor (k-NN) model to make a point forecast of five energy commodity prices. Those commodities are natural gas, heating oil, gasoline, ethanol, and crude oil. The data for the commodities are monthly and, for each commodity, two-thirds of the data are used for an in-sample forecast, and the remaining one-third of the data are used to perform an out-of-sample forecast. Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) are used to compare the two forecasts. The results showed that one method is superior by one measure but inferior by another. Although the differences of the two models are minimal, it is up to a decision maker as to which model to choose. The Diebold-Mariano (DM) test was performed to test the relative accuracy of the models. For all five commodities, the results failed to reject the null hypothesis indicating that both models are equally accurate.

No description available.