Production and changeover control policies of a class of failure prone buffered flow-shops

This article deals with a stochastic optimal control problem for a class of buffered multi-parts flow-shops manufacturing system. The involved machines are subject to random breakdowns and repairs. The flow-shop under consideration is not completely flexible and hence requires setup time and cost in order to switch the production from a part type to another, this changeover is carried on the whole line. Our objective is to find the production plan and the sequence of setups that minimise the cost function, which penalises inventories/backlogs and setups. A continuous dynamic programming formulation of the problem is presented. Then, a numerical scheme is adopted to solve the obtained optimality conditions equations for a two buffered serial machines two parts case. A complete heuristic policy, based on the numerical observations which describe the optimal policies in system states, is developed. It will be shown that the obtained policy is a combination of a KANBAN/CONWIP and a modified hedging corridor policy. Moreover, based on our observations and existent research studies extension to cover more complex flow-shops is henceforth possible. The robustness of such a policy is illustrated through sensitivity analysis.     

Fatigue management by truck drivers in real life situations: some suggestions to improve training

Truck driver fatigue is a major safety issue for truck drivers and the public in general. Although training prepares drivers to effectively operate a truck, it tends to minimize the importance of working constraints faced daily on-the-job and thus reduces its impact on safety and effectiveness. With experience, drivers develop skills to combat fatigue. Documenting these skills can contribute to improved training of apprentices. An ethnographic approach was used to better understand the real-life fatigue management skills of truck drivers. Participant observation was used to analyze the activity of apprentices in training and the activity of truck drivers at work. Observations indicated that training focused on time management and regulations, but did not prepare trainees to manage real-life constraints. Experienced drivers were not merely managing time; rather they were managing working constraints (including time) as a whole. To do so, they used two strategies: managing psycho-physical transformations and dynamic work planning. By integrating psycho-physical preoccupations into all aspects of work and by preparing future drivers to face real-life constraints, we could better train and prepare apprentices. Drivers do develop effective skills to combat fatigue which can improve training and better prepare future drivers to face daily constraints. These improvements can have a significant impact on fatigue and safety in the transportation industry.     

Global Robustness with Respect to the Loss Function and the Prior

We propose a class [I,S] of loss functions for modeling the imprecise preferences of the decision maker in Bayesian Decision Theory. This class is built upon two extreme loss functions I and S which reflect the limited information about the loss function. We give an approximation of the set of Bayes actions for every loss function in [I,S] and every prior in a mixture class; if the decision space is a subset of , we obtain the exact set.     

Economic Modeling Triggers More Efficient Planning: An Experimental Justification

Consider a firm as an organization that needs to efficiently coordinate several specialized departments in an uncertain environment. Decision making involves collective planning sessions and decentralized operational processes. In this setting this paper explores the role of economic modeling through an experimental game. Results support the idea that economic modeling favors higher performance. Economic modeling facilitates the emergence of common knowledge and the decomposition of a group decision problem into individual decision problems that are meaningfully interrelated. This paper was presented at the Porquerolles Summer School on Cognitive Sciences, September 2001.     

Nonparametric particle filtering approaches for identification and inference in nonlinear state-space dynamic systems

Most system identification approaches and statistical inference methods rely on the availability of the analytic knowledge of the probability distribution function of the system output variables. In the case of dynamic systems modelled by hidden Markov chains or stochastic nonlinear state-space models, these distributions as well as that of the state variables themselves, can be unknown or untractable. In that situation, the usual particle Monte Carlo filters for system identification or likelihood-based inference and model selection methods have to rely, whenever possible, on some hazardous approximations and are often at risk. This review shows how a recent nonparametric particle filtering approach can be efficiently used in that context, not only for consistent filtering of these systems but also to restore these statistical inference methods, allowing, for example, consistent particle estimation of Bayes factors or the generalisation of model parameter change detection sequential tests. Real-life applications of these particle approaches to a microbiological growth model are proposed as illustrations.     

Resilience of organisations and territories: The role of pivot firms

How can organisations and territories contribute to maintain a high level of innovation while at the same time adapting to a turbulent environment? We give an answer to this question by mobilizing the concept of resilience. In order to do this, we develop a conception of resilience based on two dimensions. First, an organisational dimension which refers to the capacity of an organisation to manage a disturbance to its environment and to develop a new pathway and, a territorial dimension which refers to the collective capacity of the actors to contribute to facilitate the development of territorial responses to external disturbances. We illustrate this double dimension of resilience by focusing on the role of pivot firm as major actor in a territory.     

Small-sample results for maximum-likelihood estimation in branching processes

In this article, small sample properties of the maximum-likelihood estimator (m.l.e.) for the offspring distribution (pk) and its mean m are considered in the context of the simple branching process. A representation theorem is given for the m.l.e. of (Pk) from which the m.l.e. of m is obtained. The case where p0 + p1 + p2 = 1 is studied in detail: numerical results are given for the exact bias of these estimators as a function of the age of the process; a curve fitting analysis expresses the bias of m? as a function of the mean and the variance of the offspring distribution and finally an “approximate m.l.e.” for (pk) is given.     

Contribution empirique au probleme d'agregation de fonctions autoregressives: Le cas de la politique de distribution de dividendes

This paper deals with the aggregation problem of autoregressive functions. The macrocoefficients of the simplified specification are separated into three components, namely, a true value, an aggregation bias, and an implicit sampling error. Relative importance of these components is studied by using empirical data on a sample of 40 Canadian firms. The notion of aggregation gain of Grunfeld and Griliches is also examined.