A Practical Implementation of Stochastic Programming: an Application to the Evaluation of Option Contracts in Supply Chains



mai 2004, vol. 40, n°5, pp.743-756

Départements : Informations Systems and Operations Management, GREGHEC (CNRS)

Stochastic programming is a powerful analytical method in order to solve sequential decision-making problems under uncertainty. We describe an approach to build such stochastic linear programming models. We show that algebraic modeling languages make it possible for non-specialist users to formulate complex problems and have solved them by powerful commercial solvers. We illustrate our point in the case of option contracts in supply chain management and propose a numerical analysis of performance. We propose easy-to-implement discretization procedures of the stochastic process in order to limit the size of the event tree in a multi-period environment.

Analyzing Variability in Continuous Processes

K. Rajaram, A. ROBOTIS

European Journal of Operational Research

juillet 2004, vol. 156, n°2, pp.312-325

Départements : Informations Systems and Operations Management

Author Keywords: continuous processes; production; variability; process design; operational improvement Abstract: We analyze the impact of variability on a continuous flow production process. To perform this analysis, we consider an n-stage, serial continuous process in which variability is introduced at each stage. We develop a continuous time model to capture the propagation of variability through the system and use this model to calculate the mean and the variance of the distribution of the output from this process. These results are then used to determine the optimal decisions for variability reduction when designing and operating these processes

Conception d'un projet de service dans une bureaucratie professionnelle : Le cas d'un projet immobilier et de soin


Sciences de Gestion

2004, n°45, pp.17-49

Départements : Informations Systems and Operations Management, GREGHEC (CNRS)

Efficient Scheduling RuleS in a Combined Make-to-Stock and Make-to-Order Manufacturing System

K. Youssef, C. VAN DELFT, Y. Dallery

Annals of Operations Reseach

février 2004, vol. 126, n°1-4, pp.103-134

Départements : Informations Systems and Operations Management, GREGHEC (CNRS)

In this paper, we consider a mixed MTS/MTO policy to manage a single manufacturing facility producing two classes of end-products. A few end-products have high volume demands, whereas a fairly large number of end-products have low volume demands. In this situation, it is appealing to try to produce the high volume products according to an MTS policy and the low volume products according to an MTO policy. The purpose of this paper is to analyze and compare the impact of the choice of the scheduling policy on the overall performance of the system. We consider two policies: the classical FIFO policy and a priority policy (PR). The PR policy gives priority to production orders corresponding to low volume products over production orders corresponding to high volume products. Under some simple stochastic modeling assumptions, we develop analytical/numerical solutions to optimise each system. We then provide insights regarding this issue with the help of numerical examples. It appears that for some range of parameters, the PR rule can outperform the FIFO rule in the sense that, to achieve the same service level constraint, the corresponding cost under the PR rule is much lower. This situation is encountered when the low volume products can be managed with an MTO policy under the PR scheduling rule, while they have to be managed according to an MTS policy under the FIFO scheduling rule. We also derive some theoretical properties that support our empirical findings.

Optimal server assignment for a multiple customer classes problem

J. Gonzalez, M. Norbis, L. KERBACHE

IMA Journal of Management Mathematics

juillet 2004, vol. 15, n°3, pp.195-210

Départements : Informations Systems and Operations Management, GREGHEC (CNRS)

The problem of assigning identical servers to different customer classes with the objectives of optimizing customer service and resource utilization is analysed. The problem that motivated this research is the assignment of batch jobs in a computer centre running the operating system MVS (Multiple Virtual Storage). A mathematical formulation of the problem is presented and its computational complexity discussed. A new schema for the definition of customer classes and a heuristic for the assignment of servers to classes are developed and applied to the problem. Numerical results show the efficiency of the procedure as compared with other previously utilized methods. The final part of the paper presents the conclusions and recommendations for further research

Queuing Networks and the Topological Design of Supply Chain Systems

L. KERBACHE, J. MacGregor Smith

International Journal of Production Economics

2004, vol. 91, pp.251-272

Départements : Informations Systems and Operations Management, GREGHEC (CNRS)

Manufacturing firms are linking their internal processes to external suppliers and customers. The resulting supply chain is often a very large network of activities and resources. Modeling and optimization of such complex systems is a new area of research for which exact analytical results are obviously very difficult if not impossible.We develop a methodology based on analytical queueing networks coupled with nonlinear optimization to design supply chain topologies and evaluate various performance measures. The results obtained from small network configurations as well as from the case study discussed in the paper demonstrate that our approach is a very useful tool to analyze congestion problems and to evaluate the performance of the network topologies

Savoirs d'interaction et recomposition des filières de conception

C. Abecassis-Moedas, S. BEN MAHMOUD-JOUINI, T. PARIS

Revue Française de Gestion

juin 2004, vol. 30, n°149, pp.69-84

Départements : Informations Systems and Operations Management, GREGHEC (CNRS)

Selectionism and Learning in Projects with Complexity and Unforeseeable Uncertainty

S. SOMMER, C. Loch

Management Science

2004, vol. 50, n°10, pp.1334-1347

Départements : Informations Systems and Operations Management

Companies innovating in dynamic environments face the combined challenge of unforeseeable uncertainty (the inability to recognize the relevant influence variables and their functional relationships; thus, events and actions cannot be planned ahead of time) and high complexity (large number of variables and interactions; this leads to difficulty in assessing optimal actions beforehand).There are two fundamental strategies to manage innovation with unforeseeable uncertainty and complexity: trial and error learning and selectionism. Trial and error learning involves a flexible (unplanned) adjustment of the considered actions and targets to new information about the relevant environment as it emerges. Selectionism involves pursuing several approaches independently of one another and picking the best one ex post.Neither strategy nor project management literatures have compared the relative advantages of the two approaches in the presence of unforeseeable uncertainty and complexity. We build a model of a complex project with unforeseeable uncertainty, simulating problem solving as a local search on a rugged landscape. We compare the project payoff performance under trial and error learning and selectionism, based on a priori identifiable project characteristics: whether unforeseeable uncertainty is present, how high the complexity is, and how much trial and error learning and parallel trials cost. We find that if unforeseeable uncertainty is present and the team cannot run trials in a realistic user environment (indicating the project's true market performance), trial and error learning is preferred over selectionism. Moreover, the presence of unforeseeable uncertainty can reverse an established result from computational optimization: Without unforeseeable uncertainty, the optimal number of parallel trials increases in complexity. But with unforeseeable uncertainty, the optimal number of trials might decrease because the unforeseeable factors make the trials less and less informative as complexity grows.

Spreadsheet-Based Professional Modelling


INFORMS Transactions on Education

janvier 2004, vol. 4, n°2

Départements : Informations Systems and Operations Management

Time to market vs. Time to delivery. Managing speed in Engineering Procurement and construction projects

S. BEN MAHMOUD-JOUINI, C. Midler, G. Garel

International Journal of Project Management

juillet 2004, vol. 22, n°5, pp.359-367

Départements : Informations Systems and Operations Management, GREGHEC (CNRS)

The time-to-market in NPD projects is a key factor in the competition between innovative firms. Research on concurrent engineering has shown that time can be managed as well as a delay and as a speed. Our concern in this paper is to study the time factor in the case of Engineering, Procurement and Construction (EPC) projects, where a customer initially contracts for a project from a contractor on the basis of specifications, budget and delay. Is time-to-delivery a key factor? Does its reduction represent a competitive advantage for the client and/or for the contractor in EPC projects? Is project speed a key variable to be managed, or does it result from other project parameters? We first define an analytical model to characterize a speed profile in EPC projects. We implement this model for six major construction projects developed by a large, international firm. A variety of speed profiles result. We conclude by showing the relevance of NPD project speed management in EPC projects