Catálogo de publicaciones - libros

SCM is concerned with the coordination of material, information and financial flows within and across legally separated organizational units. Software vendors have developed so called Advanced Planning Systems (APS) to provide decision support at different hierarchical planning levels of an intra-organizational supply chain. However, since APS are based on principles of hierarchical planning further ideas and concepts are required to coordinate activities and flows between adjacent planning levels of each partner.

A major goal in any optimization or decision-making process is to identify a single or all best solutions within a set of feasible alternatives. In multiple objective programming (MOP), while it is theoretically possible to identify the complete set of efficient solutions, finding an exact description of this set often turns out to be practically impossible or computationally too expensive. Thus many research efforts during the last thirty years have focused on concepts and procedures for the efficient set approximation.

Support vector machines (SVM) from statistical learning theory are powerful classification methods with a wide range of applications including credit scoring. The urgent need to further boost classification performance in many applications leads the machine learning community into developing SVM with multiple kernels and many other combined approaches. Owing to the huge size of the credit market, even small improvements in classification accuracy might considerably reduce effective misclassification costs experienced by banks. Under certain conditions, the combination of different models may reduce or at least stabilize the risk of misclassification. We report on combining several SVM with different kernel functions and variable credit client data sets. We present classification results produced by various combination strategies and we compare them to the results obtained earlier with more traditional single SVM credit scoring models.

Finanzmarktakteure müssen ihren Investitionsentscheidungen Erwartungen bzgl. der zukünftigen Marktentwicklung zugrunde legen. Sie stehen vor einer Entscheidung unter Unsicherheit und sind bestrebt mittels Prognoseverfahren die zukünftige Kursentwicklung möglichst gut vorherzusagen.

Für marketingrelevante Fragestellungen und für die strategische Bewertung von Technologiealternativen ist die Untersuchung des Diffusionsverlaufs bei Netzeffektgütern, wie z.B. Telekommunikationsdiensten, von hohem Interesse. Es wird ein Entscheidungsunterstützungssystem vorgestellt, dass insbesondere Marketingvariable zur Ausgestaltung relevanter Merkmale von netzeffektbasierten Gütern unterstützt. Ausgehend von dynamischen Nutzenfunktionen wird ein Modellansatz auf Grundlage der Mastergleichung der Physik bzw. den Mittelwertgleichungen vorgestellt, der Wechselwahrscheinlichkeiten zwischen den Produktalternativen ableitet. Die Anwendungsrelevanz des Modellansatzes wird durch entscheidungsrelevante What-If-Analysen verdeutlicht.

Solving discrete optimization problems (DOP) can be a rather hard task. Many real DOPs contain a huge number of variables and/or constraints that make the models intractable for currently available solvers. There are few approaches for solving DOPs: tree search approaches (e.g., branch and bound), relaxation and decomposition methods. Large DOPs can be solved due to their special structure. Among decomposition approaches we can mention poorly known local decomposition algorithms using the special block matrix structure of constraints and half-forgotten nonserial dynamic programming algorithms which can exploit sparsity in the dependency graph of a DOP.

Monden [11] defined two goals for the mixed-model assembly line sequencing problem: (1) Leveling the load on each station on the line, and (2) Keeping a constant rate of usage of every part used by the line. To handle these problems, Goal chasing I and II (GC- I and GC- II ) were developed by Toyota corporation. Miltenburg [9]developed a nonlinear programming for the second goal and solved the problem by applying tow heuristic procedures. Miltenberg et al [10] solved the same problem with a dynamic programming algorithm. The objective considered by Bard et al [1] was the minimization of overall line length. Bard et al [2] used Tabu search (TS) algorithm to solve a model involving two objectives: minimizing the overall line length and keeping a constant rate of part usage. Hyun et al [4] addressed three objectives: minimizing total utility work, keeping a constant rate of part usage and minimizing total setup cost. This problem was solved by proposing a new genetic evaluation. Mcmullen [6] considered two objectives: minimizing number of setups and keeping a constant rate of part usage. He solved this problem with a TS approach. Mcmullen [7,8] has also solved the same problem by using genetic algorithm, and ant colony optimization, respectively.

This paper presents a fuzzy goal programming approach for solving a multi-objective mixed- model assembly line sequencing problem in a just-in-time production system. A mixed-model assembly line is a type of production line that is capable of diversified small lot production and is able to respond promptly to sudden demand changes for a variety of models. Determining the sequence of introducing models to such an assembly line is of particular importance for the efficient implementation of just-in-time (JIT) systems. In this paper, we consider three objectives, simultaneously: minimizing total utility work, total production rate variation, and total setup cost. Because of existence conflicting objectives, we propose a fuzzy goal programming based approach to solve the model. This approach is constructed based on the desirability of decision maker (DM) and tolerances considered on goal values. To illustrate the behavior of the proposed model, some of instances are solved optimally and computational results reported.

In the paper the m-layer planar 3-dimensional assignment problem is considered. It is an NP-hard modification of well-known planar 3-dimensional assignment problem. Approximation algorithm, proposed by Gimadi and Korkishko is analysed. Its asymptotical optimality for a special class of random instances is proved.

Mixed-model assembly line sequencing is one of the most important strategic problems in the field of production management. In this paper, three goals are considered for minimization; That is, total utility work, total production rate variation, and total setup cost. A hybrid multi-objective algorithm based on Particle Swarm Optimization (PSO) and Tabu Search (TS) is devised to solve the problem. The algorithm is then compared with three prominent multi-objective Genetic Algorithms and the results show the superiority of the proposed algorithm.