Skip to main content



The Role of Operations Research in Public Policy

This paper reflects on the role Operations Research (OR) has and has not played in public policy. It is based on a plenary talk delivered on September 4,2002 to the International Conference on OR held in Klagenfurt, Austria. It concludes that OR clearly makes many important contributions to public policy making, but relatively speaking OR makes the smallest contribution to strategic issues for which authority is distributed and for which the “physics” of the underlying system are not central. One reason is that there are so many layers of people between the typical Operations Researcher and the key policy makers. OR models and model results are not effective from such a distance because they are distorted and diluted each time they are translated. An alternative path of influence would be to modify OR curricula so that people with OR training can take jobs “closer” to the key policy makers.

Jonathan P. Caulkins

Risk-Return Optimization of the Bank Portfolio

In an intensifying competition banks are forced to develop and implement enterprise wide integrated risk-return management systems. Financial risks have to be limited and managed from a bank wide portfolio perspective. Risk management rules must be accomplished from internal and regulatory points of view. Expected returns need to be maximized subject to these constraints, leading to a generalized portfolio optimization problem under different capital limits. We give a survey on a risk-return optimization model for the bank portfolio that maximizes the expected returns to the planning horizon with respect to internal and regulatory loss risk constraints. We derive consistent planning information that ensures efficient return targets and maximal capital use of the economic and the regulatory capital. The impact of the optimization is shown by an application example.

Ursula Theiler

Resource-orientated Purchase Planning and Supplier Selection - Models and Algorithms for Supply Chain Optimization and E-Commerce

With the internet being available everywhere companies are able to automate parts of their global business processes. In this paper we consider the problem of simultaneous supplier selection and purchase order sizing. Significant cost-savings are possible by integrating these problems in a software-based Business-to-Business environment.

Gabriele Reith-Ahlemeier

A Combinatorial Approach to Orthogonal Placement Problems

This article presents the main results of a PhD thesis that deals with two families of NP-hard orthogonal placement problems. We develop a common combinatorial framework for compaction problems in graph drawing and for labeling problems in computational cartography. Compaction problems are concerned with performing the conversion from a dimensionless description of the orthogonal shape of a graph to an area-efficient drawing in the grid. Map labeling is the task of attaching labels to point-features so that the resulting placement is legible. On the basis of new combinatorial formulations for these problems we develop exact algorithms. Extensive computational studies on real-world benchmarks show that our linear programming-based algorithms solve large instances of the placement problems to provable optimality within short computation time. Often, our algorithms are the first exact algorithms for the respective problem variant.

Gunnar W. Klau

Assigning Frequencies in GSM Networks

Mobile communication is a key technology in today’s information age. Despite the ongoing improvements in equipment design, interference remains a limiting factor for the use of radio communication. The author investigates in his PhD thesis how to largely prevent interference in GSM networks by carefully assigning the available frequencies to the installed base stations. The topic is addressed from two directions: first, new algorithms are presented to compute “good” frequency assignments fast; second, a novel approach, based on semidefinite programming, is employed to provide lower bounds for the amount of unavoidable interference.

Andreas Eisenblätter

Optimal Control of Methadone Treatment in Preventing Blood-Borne Disease

In this paper an optimal control model describing the influence of methadone maintenance treatment (MMT) on the spread of blood-borne diseases like Hepatitis C Virus (HCV) and Human Immunodeficiency Virus (HIV) among injection drug users is analyzed. The aim of the model is to find the optimal policy to minimize the discounted stream of the overall costs arising from MMT and the social costs caused by new infections from HIV and HCV.

Julia Almeder

Strategien zur Rüstzeitvermeidung in der Elektronikfertigung

Mit der Einführung der Surface Mount Technology wurde die Grundlage für eine hochautomatisierte Fertigungsabwicklung in der Elektronikindustrie geschaffen. Durch die Normierung und Miniaturisierung oberflächenmontierter Bauelemente (Surface Mounted Device, SMD) kann mit gängigen SMD-Bestückungssystemen eine Vielzahl unterschiedlicher SMD-Gehäuseformen verarbeitet werden (Pawlischek 1991). Dies führte zu einem Wechsel von der werkstatt-zur linienorientierten Fertigungssteuerung (Schweitzer 1994).

Claus-Burkard Böhnlein

Optimale Belegung von Stranggießanlagen mittels 2-dimensionaler Bin-Packing-Modelle

In integrierten Hüttenwerken wird in der Produktionsstufe Stranggießen Rohstahl, der in der Konvertennetallurgie in detenninistischen Pfannenmengen produziert wurde, zu kontinuierlichen Strängen vergossen. Im vorliegenden Beitrag wird das Problem der Belegung einer Stranggießanlage als zweidimensionales Bin-Packing-Problem mit variabler Stränglange und variabler Strangbreite modelliert. Als zu minimierende Zielfunktion wird hierbei auf die Anlagenleistung/Gießzeit (und darnit aufgrund der konstanten Gießgeschwindigkeit die Stränglange) zurückgegriffen. Es wird ein Lösungsverfahren vorgestellt, das mit kommerzieller OR-Software auf einem Standard-PC implementiert in der Lage ist, für praxisrelevante Problemgrößen gute Lösungen im Minutenbereich zu generieren.

Thomas Spengler, Oliver Seefried

Short-term Capacity Planning in Manufacturing Companies with a Decentralized Organization

Bottleneck stations have a production rate lower than the demand rate for parts they are producing. Therefore they limit the production system ’s output rate. Short-term capacity planning ’s purpose is to reduce the bottlenecks’ negative impact on the output rate. This paper discusses several opportunities to adjust capacity to the system ’s requirements in a decentralized production organization. Since these opportunities are related to additional costs, it is crucial to find a costefficient solution of this problem. For this purpose, a linear model is proposed which can be used to determine the optimal mix of short-term capacity adjustment measures. The model can be recalculated easily to take the actual output and capacity requirements into account if any deviations from the original plan occur.

Peter Letmathe

Performance Analysis of Make to Stock Supply Chains Using Discrete-Time Queueing Models

Manufacturing supply chains are formed out of complex interconnections among several vendors, manufacturing facilities, warehouses, retailers, logisticsproviders. In this paper, we concern ourselves with a Supply Chain Network(SCN) consisting of single manufacturing plant, one warehouse and several retail outlets. We present a discrete-time queueing model which can be used for evaluating the performance of a given SCN which processes customer orders at discrete time intervals. It can also be used for det ermining the optimal inventory level at the warehouse that minimizes total cost of carry ing inventory and back order cost associated with serving orders in the backlog queue, subject to service level constraint in terms of expected waiting t ime for customers. The model is analyzed by using discrete-time queues. We use CONWIP as th e inventory control model

Sandeep Jain, N. R. Srinivasa Raghavan

A New Optimal Demand Forecast Model

This paper presents a model for determining an optimum demand plan which has been developed whereby both demand figures from the past as well as the results of market research concerning price elasticity are known. The sales demand demonstrates a high seasonal fluctuation as well as a shortterm fluctuation, which is also relevant. The production capacity is restricted . The main idea of the model is first of all to approximate the past demand figures in order to predict future demand figures through extrapolation. In the second step, a non-linear production optimisation model is applied for calculation of the cumulative demand plan taking into account both the price elasticity and resource restrictions. The fmal predicted demand data is shown by an extrapolated curve corrected by a constant. The latter is determined by the fact that the integral of the demand curve has to be the cumulative demand plan. In addition to this result, a beta level of service and consequently the level of the safety stock can be calculated by means of a curvilinear regression analysis .

Joachim Althaler, Herbert Jodlbauer

A Multi-Product Batch-Available-to-Promise Model for Make-to-Stock Manufacturing

Available-to-Promise (ATP) comprises of a variety of tools that enhance the responsiveness of order promising and the reliability of order fulfilment. Based oncustomer requests (i.e. requested product, order quantity and delivery time window) they support “order quantity” and “order due date quoting” .

Richard Pibernik

Scheduling of Rolling Ingots Production

We consider a real-world scheduling problem arising in the context of a rolling ingots production. We review the production process and discuss peculiarities that have to be observed when scheduling a given set of production orders on the production facilities. We then describe a model for this scheduling problem using prescribed time lags between operations, different kinds of resources, and sequence-dependent changeovers. The basic principle of the solution procedure is to relax the resource constraints by assuming infinite resource availability. Resulting resource conflicts are then stepwise resolved by introducing precedence relationships among operations compet ing for the same resources. The algorithm has been implemented as a beam search heuristic enumerating alternative sets of precedence relationships.

Christoph Schwindt, Norbert Trautmann

Determination of Economic Production Quantity for a Multi-Stage Production System with Limited Storage Capacity

This paper describes a model for a multi-stage production system in which a uniform lot size is produced through all stages with a single setup and without interruption at each stage. Transportation of partial lots, called batches, is allowed between stages before the whole lot is completed. Although the batch size must be equal at any particular stage, the optimal number of equal-sized batches may differ across stages. Additionally, we assume that inventories between consecutive stages are constrained by a limited storage capacity. Considering setup costs, inventory holding costs, and transportation costs, an optimisation method is developed to determine the economic production quantity and the optimal batch sizes for each stage.

U. Buscher, G. Lindner

A Reverse Logistics Model with Integer Setup Numbers

A production-recycling system is investigated. A constant demand can be satisfied by production and recycling. The used items are bought back and then recycled. The not recycled products are disposed off. A model is examined with EOQ-type inventory holding costs and linear waste disposal, recycling, production and buyback costs. It will be shown that under these circumstances the mixed strategies are dominated by the pure strategies. The paper generalizes a former model proposed by the authors for the case of integer recycling and production batch.

Knut Richter, Imre Dobos

Lotsizing in a Production System with Rework and Product Deterioration

Producing new or recovering defective products often takes place on a common facility, with these activities carried out in lots. Consequently, there is a necessity to coordinate the production and rework activities with respect to the timing of operations and also with regard to appropriate lot sizes for both processes while completely satisfying a given demand. Thereby, it has to be taken into account whether the state of defective items that await rework worsens in the course of time or not. In this paper we present an EPQ model which addresses all of these aspects. Considering set-up and inventory holding costs, optimization algorithms are developed covering different planning situations.

K. Inderfurth, G. Lindner, N. P. Rahaniotis

Mathematical Programming Models for Strategic Supply Chain Planning and Design

Although facility location and configuration of global manufacturing and distribution networks have been studied formany years, a number of important real world issues has not received adequate attention in the literature. In this paper we propose realistic models for the strategic planning and design of global supply chains. Our main concern is to provide comprehensive models which explicitly capture the essential elements of many industrial environments. A generic mathematical programming model is described in detail which became part of Supply Chain Design, a module within mySApTM Supply Chain Management, developed by SAPAG. Moreover, extensions to the dynamic situation are discussed.

J. Kalcsics, M.T. Melo, S. Nickel

Ein dynamisches Verhandlungsmodell des Supply Chain Management

Im Fokus von Supply Chain Management (SCM) steht die untemehmensübergreifende Koordination der Material- und Informationsflüsse über den gesamten Wertschöpfungsprozess, von der Rohstoffgewinnung über die einzelnen Veredelungsstufen bis hin zum Endkunden (vgl. [8], S. 8). Die Koordination in der Supply Chain (SC) kann generell nach dem hierarchischen oder heterarchischen Prinzip erfolgen: Während das hierarchische Koordinationsprinzip dadurch gekennzeichnet ist, class eine übergeordnete Planungsebene Rahmenpläne entwirft, die untergeordneten Planungsebenen als Vorgaben dienen, treffen nach dem heterarchischen Prinzip die Akteure in der SC ihre Entscheidungen durch gegenseitige Ubereinkunft (vgl. [10], S. 5). Das hierarchische Koordinationsprinzip ist im Rahmen von SCM kritisch zu beurteilen: “Most supply chains are composed of independent agents with individual preferences. These agents could be distinct firms or they could even be managers within a single firm. (¡­) It is also reasonable to assume that each agent will attempt to optimize his own preference, knowing that all of the other agents will do the same.” ([2], S. 113). 1m Rahmen einer untemehmensiibergreifenden Koordination der Wertschöpfungsprozesse sind Planungsprobleme mit interpersonellen Mehrfachzielen zu lösen. Die Koordination erfolgt i. d. R. nach dem heterarchischen Prinzip, d. h., es erfolgt eine dezentrale Abstimmung der Akteure, die ihre interdependenten Entscheidungen durch unmittelbare Interaktion im Rahmen von Verhandlungen treffen (vgl. [3], S. 55-59).

Eric Sucky

Modeling the Interaction between Operational and Financial Decisions in the Inventory Pooling of Repairable Spare Parts Problem

Equipment-intensive industries such as airlines, nuclear power plants, various process and manufacturing plants using complex machineries are often confronted with the difficult task of maintaining a high system performance while controlling their inventory holding cost. An important type of components in such industries are called repairable items. The typical problem is to determine the optimal stocking level of spare parts. An insufficient stock of spare parts can lead to an excessive downtime cost. On the other hand, maintaining an excessive number of spare parts increases the cost of tying up capital in non-revenue-generating spare parts inventories.

Hartanto Wong, Dirk Cattrysse, Dirk Van Oudheusden

Die Schätzung von Markentreue, Nichtkäuferanteil und Marktpotenzial aus Handelspaneldaten

Eine Kernaufgabe des Marketings ist die Identifizierung der relevanten Kunden-bedürfnisse und die dementsprechende, spezifische Ausrichtung des Angebots.Häufig ist es dabei nicht mÖglich, den gesamten Markt in derselben Weise zu bearbeiten und so bedient man sich geeigneter Segmentierungstechniken. Eine wichtige Art der Segmentierung auf Konsumgütermlärkten bezieht sich auf verhaltensbezogene Variablen und hier spielen wiederum Kaufhäufigkeit sowie Markentreue eine wesentliche Rolle. Die dafür benÖtigten Informationen liefert der Marktforscher in der Regel auf der Grundlage von Haushaltspaneldaten oder von eigens dafür durchgeführten Primärerhebungen, Die vorliegende Arbeit beschäftigt sich mit der Ermittlung der Anteile an loyalen Kunden, Markenwechslern und Nichtkäufern, beschreitet zu deren Bestimmung aber einen anderen Weg, indem sie vorschlägt,aus Handelspaneldaten, also nichtpersonenbezogenen Angaben, individuelles Nachfrageverhalten abzuleiten. Der Vorteil einer solchen Datenquelle liegt darin, dass sie regelmäßig (zumeist mit Hilfe der modernen Scannertechnologie)erhoben, insbesondere fü r Logistikzwecke eingesetzt wird und daher häufig ohne Zusatzkosten zur Verfü gung steht.

Heribert Reisinger, Udo Wagner, Matthias Schuster

A Conjugate Direction Frank-Wolfe Method with Applications to the Traffic Assignment Problem

We present a version of the Frank-Wolfe method for linearly const rained convex programs, in which consecutive search direction are made conjugate to each other. We also present preliminary computat ional studies in a MATLAB environment. In these we apply the pure Frank-Wolfe, the Conjugate Direction Frank-Wolfe(CDFW) and the “partanized” Frank-Wolfe to some classical Traffic Assignment Problems. CDFW compares favorably to the other methods in this study

Maria Daneva, Per Olov Lindberg

Online-Algorithmus zur Steuerung von Verkehrslichtsignalanlagen

Es wird ein Algorithmus entworfen, der laufend verkehrsabhängige Daten einer Kreuzung und strategische Vorgaben des Verkehrsrechners verarbeitet, urn den jeweils nächsten Umlauf einer Ampelanlage zu planen.

Klaus Ladner

Optimal Sorting Machine Allocation in the Postal Distribution Network

The postal distribution system constitutes many-to-many distribution system with several transshipments [1], [3]. At the beginning, consignments are transported by vehicles from post offices to transit centers. Another fleet continuous the transport from transit centers. This one finishes at so-called sorting center, which serve a given cluster of the transit centers. Rough sorting of the consignments coming from the individual transit centers of the cluster is done in the sorting center to separate the consignments according to destination sorting centers and each class is divided according to the destination transit centers. Having completed the rough sorting, each bulk of consignments is transported to the destination-sorting center. Fine sorting is performed there to separate the consignments in accordance to destination post offices. The fine sorting is completed after all bulks have been sorted. Then the sorted consignments are loaded on vehicles and transported to destination transit centers to be there up to the time, when distribution to the destination post offices starts. It can be noticed that the earliest time of consignment bulk departure from its original transit center results from given times of departures from the transit centers. The similar situation occurs at the opposite end of the transport chain. The arrival time of a bulk at the destination transit center is fixed for distribution fleet to be able to complete its work in time. It follows that the time of bulk departure from destination sorting center is given as well.

Jaroslav Janáček

A Combined Approach to Solve the Pickup and Delivery Selection Problem

In this article we propose a model which decides upon the execution of transportation requests in order to maximize their contribution to the overall profit. Positive contributions of requests are obtained by composing appropriate routes for the vehicles involved. The composition of routes is hindered by limited capacity of vehicles and time windows associated to requests. A hybrid approach consisting of a parallel path construction heuristic which seeds a Genetic Algorithm is presented. We assess its capability for a set of suitable benchmark instances for pickup and delivery problems with time windows

Jörn Schönberger, Herbert Kopfer, Dirk C. Mattfeld

VRP with Interdependent Time Windows — A Case Study for the Austrian Red Cross Blood Program

This work shows our endeavors to assist the fleet management division of the Austrian Red Cross blood programme by developing an algorithm for the VRP with interdependent time windows. Usually, blood donation campaigns are operating all day long and the collected blood must be transported from campaign locations to a central blood processing facility, where it has to be further processed within a certain amount of time. By considering this constraint, each location must be visited for a number of times during the course of a day. Certain vehicles are scheduled for picking up the stored blood donations and delivering it to the processing facility within the given tim e window. Time windows for each location show a dynamic behaviour as they depend on previous pickup times. In order to solve the problem we develop a savings-based constructive procedure for the VRP with interdependent time window const raints. First, our algorithm calculates the minimum number of pickups for each location and then it determines possible combinations when constructing a tour. Preliminary results on real life data show that we can easily decrease overall tour lengths.

Karl Doerner, Manfred Gronalt, Richard F. Hartl, Marc Reimann, Kerstin Zisser

Incident Management Based on Real Time Simulation

The main function of a traffic simulation which is used to evaluate the current traffic situation is reflected in the spatio-temporal interpolation of the situation between the stationary detector points of any road network.

Jürgen Zajicek, Martin Linauer, Katja Schechtner

Online-Dispatching of Automobile Service Units

We present an online algorithm for a real-world vehicle dispatching problem at ADAC; the German Automobile Association.

Martin Grötschel, Sven O. Krumke, Jörg Rambau, Luis M. Torres

Multi-Class User Equilibria under Social Marginal Cost Pricing

In the congested cities of today, congestion pricing is a tempting alternative. With a single user class, already Beckmann et al. showed that “system optimal” traffic flows can be achieved by social marginal cost (SMC) pricing. However, different user classes can have wildly differing time values. Hence, when introducing tolls, one should consider multi-class user quilibria, where the classes have different time values. With SMC pricing, Netter claims that multi-class equilibrium problems cannot be stated as an optimization problems. We show that, depending on the formulation, the multi-class SMC-pricing equilibrium problem (with different time values) can be stated either as an asymmetric or as a symmetric equilibrium problem. In the latter case, the corresponding optimization problem is in general non-convex. For this non-convex problem, we devise descent methods of Frank-Wolfe type. We apply the methods and study a synthetic case based on Sioux Falls.

Leonid Engelson, Per Olov Lindberg, Maria Daneva

School Bus Routing and Scheduling Problem

We consider th e school bus rout ing and scheduling problem,where transportation demand is known and bus scheduling can be planned in advance.We first propose a modeling framework which aims to optimize a level of service for a given number of buses. Then,we describe a procedure building first a feasible solution, and subsequently improving it,using a heuristic. After the performance analysis of three types of heuristics on real and synthetic data,we recommend the use of simulated annealing exploring infeasible solutions,which performs slightly better than all others. More importantly, we find that the performance of all heuristics is not globally affected by the choice of the parameters.This is important from a practitioner viewpoint,as the fine tuning of algorithm parameters is not critical for its performance.

Michela Spada, Michel Bierlaire, Thomas M. Liebling

Covering Population Areas by Railway Stops

We study the installation of new stops (or stations) along existing links in public transportation, e.g. railway systems. This improves the coverage level of the system, i.e., the number of people living near some stop. On the other hand, additional cost is incurred and travel times tend to increase. We model this as a network location problem, where the network corresponds to the transportation links. Our main contribution is to model the population distribution by a system of compact subsets in the plane, the population areas. The goal is to cover all these areas with as few stops as possible along the railway tracks.We present an efficient algorithm for finding an optimal solution in the case of a single edge and show its applicability to real world problems.

Anita Schöbel, Michael Schröder

Innovative Lösungen im bimodalen Transport Straße / Binnenwasserstraße

Im Bereich des Güterverkehrs weisen die prognostizierten Entwicklungen flür die kommenden Jahre einen deutlichen Anstieg aus, wobei die Zuwächse sich im wesentlichen auf den Straßengüterverkehr beziehen. Die Ursachen hierfür sind sehr vielschichtig, u.a. bedingt durch strukturelle Veränderungen im Güteraufkommen und dem Steigen der Transportnachfrage aufgrund einer zunehmenden Internationalisierung in der Fertigung und im Handel (s. u.a. Daduna 2001). Daein Anwachsen des Straßengüterverkehr mit den verkehrspolitischen Zielvorstellungen in keiner Weise zum Einklang zu bringen ist, u.a. aus ökologischen U Überlegungen aber auch aufgrund der weitgehend fehlenden Möglichkeiten für Kapazitätserweiterungen im Straßennetz, wird nach Auswegen aus dieser Situation gesucht. Regulierende Maßnahmen staatlicher Institutionen zur Reduzierung der (Straßen-)Güterverkehre bzw. zur Beeinflussung von Verkehrssystementscheidungen erweisen sich in der Regel als wirtschaftspolitisch kontraproduktiv, auch mit Blick auf die Bedeutung der Mobilität als eine der wesentlichen Determinanten für ein nachhaltiges Wirtschaftswachstum. Geeignete Lösungen müssen daher mit Hilfe anderer Ansätze gefunden werden, wobei neben den Informations- und Kommunikations(IuK)technologien (s. u.a. Daduna/VoB 2000 bzw. Daduna 2001) die Bildung bi- bzw. multimodaler Transportketten im Kombinierten Ladungsverkehr (KLV) (s. u.a. Seidelmann 1997) im Vordergrund stehen.

Joachim R. Daduna, Johannes Schröter

Optimal Routing of Snowplows-A Column Generation Approach

In countries with heavy winters, winter road maintenance, i.e. snow removal, salting etc, is an important problem. In Sweden the government and municipalities together spend close to 0.3 GEUR every year for winter road maintenance. Approximately half of this is snow removal cost, which mainly is the cost for snowplows, which in turn mainly depends on the routing of the snowplows. In this paper we study optimal routing of snowplows after snowfall. One has then to design a set of routes starting and ending at given depots, such that each road segment gets plowed within a prescribed time window, depending on the class of road segment. Our solution approach is based on Dantzig-Wolfe decomposition with column generation. In order to obtain an integer solution to the master problem we have applied two heuristic procedures, one using branch-and-bound on a subset of the columns and the other one a greedy procedure.

Nima Golbaharan, Per Olov Lindberg, Maud Göthe Lundgren

Savings Based Ants for Large-scale Vehicle Routing Problems

In this paper we present a modified version of our Savings based Ant System for large-scale instances of the Vehicle Routing Problem (VRP). The main idea is to speed up the search by letting the ants solve only sub-problems rather than the whole problem. This is particularly necessary, when one has to solve large real world instances, for which the computation times of classic meta-heuristics are prohibitive. Our model is also inspired by Taillard ’s work on th e VRP ([1]), where the Tabu Search procedure employed solves only some sub-problems.

Marc Reimann, Karl Doerner

Single Machine Scheduling Problems with Exponentially Start Time Dependent Job Processing Times

The paper deals with single machine scheduling problems, where job processing times are given by exponentially start time dependent functions. We consider decreasing and increasing functions of job processing times. For the increasing functions, we proved ordinary and strong NP-hardness of the makespan minimization problem under ready times restrictions. We obtained the same results, i.e., ordinary and strong NP-hardness, for the maximum lateness minimization problem, but for the decreasing functions of job processing times. We proved also ordinary NP-hardness for the total weighted completion time minimization, under the assumption that job processing times are characterized by decreasing functions.

Aleksander Bachman, Adam Janiak, Mikhail Y. Kovalyov

Scheduling Problems with Optimal Due Interval Assignment Subject to Some Generalized Criteria

The paper deals with two problems of scheduling jobs on identical parallel machines, in which a due interval should be assigned to each job. Due interval is a generalization of well known classical due date and describes a time interval, in which a job should be finished. In the first problem, we have to find a schedule of jobs and a common due interval such that the sum of the total tardiness, the total earliness and due interval parameters is minimized. The second problem is to find a schedule of jobs and an assignment of due interval to all jobs, which minimize the maximum of the following three parts: the maximum tardiness, the maximum earliness and the due int erval parameters. We proved that the considered problems are NP-hard and outlined some methods how to solve them approximately as well as optimally.

Adam Janiak, Marcin Marek

A New Exact Resource Allocation Model with Hard and Soft Resource Constraints

In this paper we present a new bilevel resource allocation model with hard and soft resource constraints for projects. By definition, a hard resource constraint is not resolvable within the given planing horizon, but a soft resource conflict may be managed by a flexible “hiring-firing” strategy.

Ferenc Kruzslicz

Minimizing Total Weighted Tardiness on Parallel Batch Process Machines Using Genetic Algorithms

Recently, the electronics industry has become the largest industry in the world. A key aspect of this industry is the manufacturing of integrated circuits. In the past, sources of reducing costs were decreasing the size of the chips, increasing the wafer sizes and improving the yield, simultaneously with efforts to improve operational processes inside the wafer fabrication facilities (wafer fab). Currently, it seems that the improvement of operational processes creates the best opportunity to realize the necessary cost reductions [6].

Lars Mönch, Hari Balasubramanian, John W. Fowler, Michele E. Pfund

Sorting with Line Storage Systems

We consider a problem that arises in car production. In particular, we focus on the sorting of car bodies with respect to their designated enamel colors for a paint shop. Our objective is to sort a given car body sequence so that the number of color changes within the sorted sequence is minimized. Current technology is to perform the realignment of a sequence by the use of a line storage system.

Thomas Epping, Winfried Hochstättler

On Solvability of the Project Scheduling Problem with Accumulative Resources of an Arbitrary Sign

A project scheduling problem (PSP) with precedence constraints, deadlines and resource constraints is considered. It was known before that PSP is polynomially solvable if all resource constraints are of so called accumulative type (please, don’t mix the notion with completely different notions cumulative and cumulatives lately introduced in the literature and being nothing but a kind of renewable resource constraints!), provided that both resource availabilities and resource requirements are non-negative. Now we show that PSP with accumulative resource constraints remains polynomially solvable if we allow resource availabilities of an arbitrary sign, while all resource requirements remain non-negative. On the other hand, it is shown that in the case of resource requirements of an arbitrary sign the problem becomes NP-hard in strong sense.

Edward Gimadi, Sergey Sevastianov

Cost Optimized Layout of Fibre Optic Networks in the Access Net Domain

During the last two years European network-carriers have invested 7.5 bi1lions Euro in the expansion of the core and the distribution net domain (backbones, city backbones and metropolitan area networks). However, investigations have shown that about 95% of the total costs for the implementation may be expected for the area-wide realization of the last mile (access networks). In order to achieve a return on investment, carriers wi1l be forced to link access net nodes, like corporate clients, private customers or communication nodes for modern mobile services (e.g. UMTS) to their city backbones. We present the methodology behind the planning tool NETQUEST-OPT for the computation of cost optimized and real world laying for fiber optic access networks. Bases on detailed geoinformation data, cluster strategies, exact and approximation algorithms of graph theory, combinatorial optimization and ring closure heuristics form the optimization kernel. Beside the methodology, we present results of a benchmark project which was rocessed with one of our industrial partners.

Peter Bachhiesl, Gernot Paulus, Markus Prossegger, Joachim Werner, Herbert Stögner

Ein Algorithmus zur sicheren elektronischen Stimmabgabe über das Internet

In einem grundlegenden Beitrag identifizieren Nurmi, Salomaa und Santean [1] zwei Dienste im Rahmen eines E-Voting-Systems: den Registrator, bei dem eine elektronische Wahlberechtigung eingeholt wird, und die Stimmabgabestelle (elektronische Urne). Im folgenden wird die Eignung von aus der Literatur bekannten Verfahren diskutiert und ein eigener Vorschlag vorgestellt.3

Alexander Prosser, Robert Müller-Török

Accelerated MILP-Strategies for the Optimal

Operation Planning of Energy Supply System

In this paper a MILP-based opt imization algorithm for the optimaloperation planning of energy supply systems is presented. For this purpose, the maintasks of optimal operation planning are discussed and th e Priority-Based DynamicSearch Strategy (PBDSS) for the solution of the optimization problem is developed taking into account different acceleration strategies. Exemplary results comparingstandard branch-and-bound algorithm to the new search strategy demonstrate the significant improvement of the optimization process.

Peter Häcklander, Johannes F. Verstege

On On-line Systems for Short-term Forecasting for Energy Systems

The paper describes experiences with developing on-line computer systems for short-term forecasting of wind power production and heat consumption in district heating networks. The computer systems are briefly described and some general aspects regarding system modeling with the purpose of forecasting are discussed. One consequence of the approach used is that the stochastic properties of the forecast errors can not be inferred from the models generating the forecasts. With the purpose of using the stochastic properties as input to formal OR-models we discuss how these can be modeled.

Henrik Aalborg Nielsen, Torben Skov Nielsen, Henrik Madsen

Combining Bottom-up and Finance Modelling for Electricity Markets

The high price volatility observed on liberalised electricity spot markets requires detailed modeling to successfully manage price risks. For example, in December 2001 prices at the German power exchange LPX exceeded 1000 €/MWh for certain hours, whereas the yearly average is around 22 €/MWh. In the USA even price spikes with more than 9,000 €/MWh have been observed. Also outside extreme events, price changes of 20% from one day to another are not unusual. In order to cope with this volatility, price models are needed which reflect the particularities of the electricity market, notably the non-storability of electricity. On the other hand, also the experience from other commodity markets has to be integrated in any model for electricity prices. Therefore, in the following a model is presented which combines several approaches.

Christoph Weber

Gestaltung von Stoffstrom-Netzwerken zum Produktrecycling

Im vorliegenden Beitrag erfolgt die Vorstellung eines Konzepts zur 6konomische Bewertung von alternativen Stoffstrom-Netzwerken zum Elektro(nik)altgeriiterecycling. Dabei finden investitionsabhangige Kosten, variable Stofffluss-sowie Prozesskosten und sonstige Gemeinkosten Beachtung. Fiir die Ermittlung der Stofffluss-und Prozesskosten wird ein Mengengeriist auf Basis eines linearen Optimierungsmodells erstellt. Es erfolgt eine Anwendung am Beispiel der Unterhaltungselektronik fiir das Bundesland Niedersachsen.

Thomas Spengler, Grit Walther

Ein Ansatz zur Bewertung von Remanufacturingstrategien

Die Verschärfung der europäischen Umweltgesetzgebung wie sie sich etwa in der Richtlinie des europäischen Parlaments über Elektro-und Elektronikaltgeräte zeigt - und hierbei insbesondere die Erweiterung der Produktverantwortung auf den gesamten Produktlebenszyklus - stellt für Hersteller neue Herausforderungendar. Dabei kommt dem Schließen der Kreisläufe sowohl auf der Material- (Recycling) als auch auf der Komponentenebene (Remanufacturing) eine zentrale Bedeutung zu. In diesem Kontext ist eine Weiterentwicklung der Instrumente des Supply Chain Management im Hinblick auf die Integration von Rückflüssen erforderlich.

Axel Tuma, Baptiste Lebreton

Environmental Coordination of Supply Chain Networks Based on a Multi-Agent System

Following the actual discussion concerning modem production concepts, an increasing trend to network organizations can be identified. In interconnected production systems materials and different forms of energy are provided, converted, stored and transported. Environmental impacts can be identified at any production unit. In this context the allocation of workload to the different production units, considering simultaneously both ecological and economical goals, is of special interest. In this way it is possible to use available resources more efficiently and to reduce emissions and by-products caused by the production process. To achieve this goal a model of the production system has to be created. This model has to include the input and output streams. Due to the modelling requirements of supply chain networks (modelling of more or less independent production units, dynamic behavior) an agent-oriented simulation seems to be adequate. Secondly evaluation methods for the modelling of the economic and ecological goals have to be integrated. In order to address the multi-criteria structure of such a decision problem, a goal programming approach is discussed.

Axel Tuma, Jürgen Friedl

Decision Support for the National Implementation of Emission Reduction Measures bythe Dynamic Mass Flow Optimisation Model ARGUS

A large number of industrial sectors is affected by enforced environmental requirements, and therefore a cost-optimal transformation is essential for the economic and environmental efficiency both of companies and regions. Techno-economic dynamic optimisation models - like ARGUS - have been developed and implemented for various European Countries. The resulting cost functions for various scenarios are essential for the multi-national allocation of emission reductions. Cost-discounting effects and the temporal pathway of the implementation of emission reduction options within a given planning horizon (up to 2020) are considered. The results can be used to analyse the cost-effectiveness of environmental legislation and for emission projections and inventories.

Jutta Geldennann, Nurten Avci, Stefan Wenzel, Otto Rentz

Fuzzy Scheduling for the Dismantling of Complex Products

In this paper, an approach is presented for the solution of decisionmaking problems in short-term dismantling planning. Weak data resulting from uncertain dismantling time parameters are modeled as fuzzy sets and integrated into a scheduling model that supports deriving optimal work plans. Case studies from the construction industry and the electronic scrap recycling show the application of the approach.

Frank Schultmann, Otto Rentz

Group Decision Making Versus Expert Opinion in the Multi-Objective Analysis of Ecosystem Management

Ecological issues with demands for preserving the nature and multipurpose use of land have, along with existing economic criteria, become a key part of the modem concept of ecosystem management. Therefore, on one hand, the land owners and experts are faced with the land use decisions which maximize the profit and refer to ecological objectives, while on the other hand, the public, who benefits from the amenity value of the ecosystem, specifically derives its own scenario of decisions. As such, an ecosystem management problem is a satisfactory attainment of multiple, but conflicting, objectives (Zadnik 200 I, 2002).

Lidija Zadnik Stirn

Capital Market Efficiency — An Empirical Analysis of the Dividend Announcement Effect for the Austrian Stock Market

This study investigates the effects of dividend announcements using data from the Austrian stock market. Abnormal returns are established as the difference between actual returns and predicted returns generated from time series models. We use the model of expected dividends, where expectations are based on prior dividends. Our results provide evidence that announced dividend changes convey new information to the market as stock prices move in the same direction as dividends. In addition, the speed of stock price reaction to the new information provides a test of the semistrong form of the efficient capital market hypothesis.

Roland Mestel, Henryk Gurgul, Christoph Schleicher

On Tail Index Estimation and Financial Risk Management Implications

Estimation bias is a critical issue when drawing inferences about the tails of the return distribution of risky assets. Risk management applications which rely on methods of extreme value theory must consider the statistical properties of the tail index estimator used; see for example recent simulation studies by Gomes and Oliveira (2001), Matthys and Beirlant (2000), and Wagner and Marsh (2000). The present contribution outlines potential effects of bias on quantile estimation thereby considering error sensitivities within the widespread Value-at-Riskapproach. The results show that particularly inference far out in the distribution tails is sensitive to bias. The paper further gives an overview of recent literature documenting small sample bias in tail index estimation and points out some new approaches aiming at its reduction.

Niklas Wagner

Project Risk Management by a Probabilistic Expert System

Efficient applications of expert systems to project risk management problems are seldom, if not unusual. In this paper we overcome this lack by using the probabilistic expert system shell SPIRIT. The rule-based shell ’s power in conditioning, inference and reasoning under incomplete information will work well on risk estimation and classification. A key characteristic of SPIRIT is the possibility to integrate project objectives into the risk management model. So known dependencies between risk variables can be modelled by the user if known beforehand, whereas hidden dependencies might be detected by the proper system. Because of the novelty of projects they suffer from incomplete information and it is this incompleteness which SPIRIT handles at high information fidelity. Furthermore undirected inference is possible, due to the undirected graphical structure in which knowledge is acquired and processed. So, in an early-state risk management situation - where the final model in terms of certain variables and/or their respective dependencies is not yet available - preliminary risk analyses and even recommendations for adequate risk treatment measures are possible, too. A middle size product developement example, including 12 binary variables and 34 rules, shows the inferential power of SPIRIT.

André Ahuja, Wilhelm Rödder

Regulatory Impacts on Credit Portfolio Management

Efficient credit portfolio management is a key success factor of bank management. Discussions of the new capital adequacy proposals by the Basle Committee on Banking Supervision enlighten the necessity to consider the credit risk management both from the internal and the regulatory point of view. We introduce an optimization approach for the credit portfolio that maximizes expected returns subject to internal and regulatory risk constraints. With a simplified bank portfolio we examine the impact of the regulatory risk limitation rules on the optimal solutions.

Ursula Theiler, Vladimir Bugera, Alla Revenko, Stanislav Uryasev

Verfahren zur Risikokapitalallokation im Eigenhandel von Banken

Der Value-at-Risk (VaR) als MaB zur Quantifizierung von Marktpreisrisiken bietet neue Moglichkeiten der Erfolgs-und Risikosteuerung, speziell im Eigenhandelsgeschiift der Bank. Die bisherige wissenschaftliche Auseinandersetzung konzentrierte sich v.a. auf Modelle zur Schiitzung des VaR. Der niichste Schritt besteht in der Konstruktion eines auf dem VaR-Modell aufbauenden VaRLimitsystems. Das ist zum einen ein aufsichtsrechtliches Erfordemis und zum anderen eine okonomische Konsequenz.

Mario Straßberger

Process Optimization via Conventional Factorial Designs and Simulated Annealing on the Path of Steepest Ascent for a CSTR

This work determines the efficiency of sequential algorithms for automatic optimization of a chemical process. A method of steepest ascent and an integrated approach between the method of steepest ascent and Simulated Annealing, are compared on a simulated continuous stirred tank reactor (CSTR) with various levels of signal noise. The results suggest that the method of steepest ascent seems to be the most efficient on the CSTR surface at the lower levels of noise. However, the integrated approach with the Simulated Annealing element works well when the standard deviation of the noise is at higher levels. Although the average, the standard deviation of the greatest actual concentration of the product and percentage of sequences ended at the optimum from the integrated algorithm are better, it needs more runs, on average, to converge to the optimum when compared.

Pongchanun Luangpaiboon

Optimization on Directionally Convex Sets

Directional convexity generalizes the concept of classical convexity. We investigate aC-convexity generated by the intersections of C-semispaces that efficiently approximates directional convexity. We consider the following optimization problem in case of the direction set of aC-convexity being infinite. Given a compact aC-convex set A, maximize a linear form L subject to A. We prove that there exists an aC-extreme solution of the problem. A Krein-Milman type theorem has been proved for aC-convexity. We show that the aC-convex hull of a finite point set represents the union of a finite set of polytopes in case of the direction set being finite.

Vladimir Naidenko

Meta-Heuristiken in virtuellen Lernumgebungen

In der Lehre gewinnt der Einsatz von virtuellen Lemumgebungen an Bedeutung, wobei fUr das Operations Research aufgrund der Anforderungen an eine praxisorientierte Ausbildung ein erhohter Bedarf an qualitative hochwertigen Lemangeboten besteht. Anhand von konkreten Lerninhalten soll aufgezeigt werden, auf welche Weise eine Verkniipfung der theoretischen Grundlagen mit praktischen Anwendungen hergestellt werden kann. Hierzu erarbeiten die Lemenden grundlegende Begriffiichkeiten und Zusammenhange, die in einem weiteren Schritt anhand von realen Problemstellungen aus einer neuen realitatsbezogenen Perspektive betrachtet und dadurch vertieft und gefestigt werden.

Torsten Reiners, Imke Sassen, Stefan Voß

An Evolutionary Algorithm for Bayesian Network Triangulation

The problem of triangulation (decomposition) of Bayesian networks is considered. Triangularity of a Bayesian network is required in a general evidence propagation scheme on this network. Finding an optimal triangulation is NP-hard. A local search heuristic based on the idea of evolutionary algorithms is presented. The results obtained using existing and proposed approaches are compared on a basis of a computational experiment.

Tomasz Łukaszewski

Approximation Algorithms for the k-center Problem: An Experimental Evaluation

In this paper we deal with the vertex k-center problem, a problewhich is a part of the discrete location theory. Informally, given a set of cities, with intercity distances specified, one has to pick k cities and build warehouses in them so as to minimize the maximum distance of any city from its closest warehouse. We examine several approximation algorithms that achieve approximation factor of 2 as well as other heuristic algorithms. In particular, we focus on the clustering algorithm by Gonzalez, the parametric pruning algorithm by Hochbaum-Shmoys, and Shmoys’ algorithm. We discuss several variants of the pure greedy approach. We also describe a new heuristic algorithm for solving the dominating set problem to which the k-center problem is often reduced. We have implemented all the algorithms, experimentally evaluated their Quality on 40 standard test graphs in the OR-Lib library, and compared their results with the results found in the recent literature.

Jurij Mihelič, Borut Robič

MaxFlow-MinCut Duality for a Paint Shop Problem

Motivated by an application in car manufacturing we consider the following problem: How can we synthesize a given word from restricted reservoirs of colored letters with a minimal number of color changes between adjacent letters? We focus on instances in which each letter occurs exactly twice, once in each of two given colors. In this case the problem turns out to be the dual of a MinCut problem for one point extensions of a certain class of regular matroids. We discuss consequences of the MaxFlow-MinCut duality and describe algorithmic approaches.

Thomas Epping, Winfried Hochstättler, Marco E. Lübbecke

From Edge Decomposition Formulae to Composition Algorithms

For some graph invariants simple decomposition formulae are known. Unfortunately, a straight forward implementation of such formulae usually leads to algorithms exponential in the number of edges or nodes of the graph even if the graph has a structure that wouldallow polynomial algorithms. This paper describes a way to derive polynomial algorithms for graphs of bounded pathwidth from edge decomposition formulae by using a variant of the composition method originally proposed by the author for the computation of graph invariants in graphs of small bandwidth.

André Pönitz

The Complexity of Some Problems on Maximal Independent Sets in Graphs

Let mi(G) be the number of maximal independent sets in a graph G. A graph G is mi-minimal if mi(H) < mi(G) for each proper induced subgraph H of G. As it is shown in [6], every graph G without duplicated or isolated vertices has at most 2k-1 +k - 2 vertices, where k = mi(G) > 2. Hence the extremal problem of calculating m(k) = max{IV(G)1: G is a mi-minimal graph with mi(G) = k} has a solution for any k ~ 1 We show that 2(k -1) ~ m(k) ~ k(k -1) for any k ~ andconjecture that m(k) = 2(k - 1). We also prove NP-completeness of some related problems.

Igor Zverovich, Yury Orlovich

Testing Solution Quality in Stochastic Programs

We describe a statistical procedure for testing the quality of a feasible candidate solut ion for an important class of stochastic programs. Quality is defined via the so-called optimality gap and th e procedure’s output is a confidence interval on this gap. We review a multiple-replications procedure for constructing the confidence interval. Then, we present a result that allows the procedure to be computationally simplified to a single-replication procedure.

David P. Morton

Scenario Updating Method for Stochastic Mixed-integer Programming Problems

In this paper, we propose an approximation scheme to solve large stochastic mixed-integer programming (SMIP) problems with fixed recourse. We refer to this as the Scenari o Updating Method. The algorithm is based on solving instances of the problem, which cont ain only a subset of the scenarios in the scenario tree. At each iteration, th e subset of scenarios is updated by adding only those scenarios which suggest a significant potential for change in the objective function value. The algorithm is terminated when the potential for change is insignificant.Different selection and updating rules are discussed.

Guglielmo Lulli, Suvrajeet Sen

Pricing of Multidimensional Resources in Revenue Management (Multidimensional Dynamic Stochastic Knapsack Problem)

Revenue management deals with selling a limited amount of a perishable resource. This resource becomes valueless at a known point of time (deadline). There are many papers dealing with one-dimensional resources (e. g. seats or hotel rooms) using different approaches as linear programming or Markov decision processes which consider the problem as a dynamic stochastic knapsack problem (DSKP).

Jens Feller

A Note on Quantitative Stability and Empirical Estimates in Stochastic Programming

The paper deals with a stability of stochastic programming problems considered with respect to a probability measure space. In particular, the paper deals with the stability of the problems in which the operator of mathematical expectation appears in the objective function, constraints set is “deterministic” and the probability measure space is equipped with the Kolmogorov or the Wasserstein metric. The stability results are furthermore employed to statistical estimates in the stochastic programming problems. Some results on a consistence and a rate of convergence are presented.

Vlasta Kaňková, Michal Houda

Splitting and Localization of the Epi-Topology Combined with Randomness

The paper introduces an extension of the epi-convergence, the lower semicontinuous approximation and the epi-upper semicontinuous approximation of random real functions in probability. The new notions could be helpful tools for sensitivity analyzes of stochastic optimization problems. The research is evoked by [4] and continuous the research started in [6], [7] and [5].

Petr Lachout

Standards für Modellierung und Simulation

Simulation ist das systematische Durchspielen des Verhaltens von geplanten, sich in der Entwicklung befindlichen oder bereits existierenden Systemen. Dabei wird ein Simulationsmodell zugrundegelegt, das die für die Simulation relevanten Aspekte des betrachteten Systems nachbildet. Simulation ist somit das zielgerichtete Experimentieren an Modellen, die der Wirklichkeit nachgebildet sind (Oberweis et al. 1999; Bossel 1992).

Claus-Burkard Böhnlein

System Dynamics (SD) — An Approach within Corporate Planning

Increasing speed in business, engagement on international markets, as well as rising costs for mismanagement and wrong decisions require that effects of changes, both endogenous and exogenous, can be foreseen. At least, it is important to have the possibility to rewrite existing action plans according to the current situation almost instantly. The use of models that are built and validated according to the companies current situation (including market information etc.) not only allows playing with the status quo but is the basis for simulations that provide results on possible future outcomes.

Peter Bradl

Optimal Decision Rules in a Monetary Union

This paper develops a dynamic game model to study strategic interactions between the decision-makers in a monetary union. In such a union, governments of the participating countries pursue national goals when deciding on fiscal policies, whereas the common central bank’s monetary policy aims at union-wide objective variables. Considering the example of a negative demand shock, we show how different solution concepts for the dynamic game between the common central bank and the national governments can be used as models of a conflict between national and supra-national institutions (noncooperative Nash equilibrium) and of coordinated policy-making (cooperative Pareto solutions).

Doris A. Behrens, Reinhard Neck

Impact of Feedback Loop on Group Decision Process when Applying System Dynamics Simulators

Influence of feedback information on group decision process supported by the application of system dynamics simulators is determined in present study. Experiment with decision groups was conducted under different experimental conditions: a1) determination of business strategy without application of formal model, a2) determination of the strategy with application of formal system dynamics model and a3) determination of the strategy with additional application of the group feedback information. The hypothesis that model application and group feedback information positively influence the convergence of the decision process and yield higher values of criteria function was confirmed. Described group decision process was subsequently analyzed in order to determine its frequency and deviation.

Andrej Škraba, Miroljub Kljajić

The Management Game SINTO-Market — Report on Some Recent Experiments

The management game SINTO-Market which was originally developed and programmed by Otwin Beckerand Reinhard Selten was recently performed at the University of Kalmar/Sweden and University of Graz/Austria several times. The decisions were made by groups of students, each group representing one of the three firms. The electronic management game SINTO-Market puts the players in a competitive situation in the branded food product sector. We will give a short description of the structure of the game. The game runs via internet and can be seen at the web-page of our department. It is interesting to see that much can be learned about human behavior in complex economic decision situations from experimental research with management games. Surprisingly the results show some differences between genders

Otwin Becker, Tanja Feit, Vera Hofer, Ulrike Leopold-Wildburger, Susanne Lind-Braucher, Jörg Schütze, Reinhard Selten

Bounds & Likelihood Procedure Revisited

In an experiment subjects are asked to predict the next value of an univariate time series on the basis of the past observations. The average forecasts of the subjects can be well described by a surprisingly simple rule, which is called the “Bounds and Likelihood procedure”(Becker/Leopold 1996). In a new series of the experiment, conducted 2002 in Graz with 72 participants, the behaviour and score of the subjects is further analysed with the Selective Attention Test (subset of D2, Brickenkamp 1962) and Intelligence Test for Reasoning (subset of PSB, Horn 1969).

O. Becker, J. Leitner, U. Leopold-Wildburger, J. H. Schuetze

On the Allocation of Excesses of Resources in Linear Production Problems

In this paper we consider non-centralized linear production situations. In one of those situations, each producer iof a setNhas an optimal production plan x0i for a linear production problem given by max {cixi : Aixi ≤bi,xi≥0}. In of resource (b1−A1x01,...,bn−Anx0n). We study the games which describe this situation when players cooperate and side payments are possible, when players do not cooperate and when players cooperate and side payments are not possible.

F. R. Fernández, G. Fiestras, I. García-Jurado, J. Puerto

Simulation eines CO 2-Zertifikatenhandels und algorithmische Optimierung von Investitionen

Mithilfe der am Zentrum für Angewandte Informatik entwickelten Software TEMPI (Technology Emissions Means Process Identification) steht eine Modellierungsumgebung zur Verfügung, die es ermöglicht, finanzielle Investitionen und ihren Einfluss auf CO2-Minderungsmassnahmen mithilfe eines zeitdiskreten Modells vergleichend gegenüberzustellen und zu optimieren.

Silja Meyer-Nieberg, Stefan Pickl

Indirect Expenditure Functions and Shephard’s Lemma

Based on preferences on the normalized price space, an indirect expenditure function will be defined. We will study the properties of the inverse demand function and of the indirect expenditure function following from hypotheses on normalized prices. It will also be shown that Shephard’s lemma holds without assuming transitivity and completeness of the underlying preference relation or differentiability of the indirect expenditure function.

S. Fuchs-Seliger

Bayesian Estimation of the Heston Stochastic Volatility Model

The goal of this article is an exact Bayesian analysis of the Heston (1993) stochastic volatility model, where different parameterizations of the latent volatility process and the parameters of the volatility process will be used to improve convergence and the mixing behavior of the sampler. We apply the sampler to simulated data and to DM/Us$ exchange rate data.

Sylvia Frühwirth-Schnatter, Leopold Sögner

Forecasting with Leading Economic Indicators — A Neural Network Approach

There is variety of important issues associated with the problem of business cycle forecasting, especially regarding forecast methodology and forecast evaluation. Overall, we can say that macroeconomic forecasting has a fairly poor reputation (Granger 1996). Still, even with the recognition that forecasting business cycles is a very difficult task, we find some hopeful signs for future progress. Our research on forecasting has focused on development of new approach in forecasting with classical NBER leading indicators by applying neural networks.

Timotej Jagric

Estimating Multivariate Conditional Distributions — An Application to the Truck Sales Forecast

A concept for forecasting the conditional multivariate distribution has been developed. It allows the forecast of the joint distribution of target variables in dependence on explaining variables. The concept can be applied to general distribution families such as stable or hyperbolic distributions. The conditional distribution parameters are estimated by a global optimization method, using neural networks for functional approximation. The information about a complete distribution of forecasts can be used to quantify the reliability of the forecast. A comparison with conventional forecasting concepts is done and the additional benefit of forecasting conditional distribution in general, and of hyperbolic distribution in particular is shown. The concept is illustrated on a case study concerning the future truck demand. In this application, th e distribution parameters are conditional on properties of the product and information about existing orders.

Eric A. Stützle, Tomas Hrycej

Application of Techniques of Functional Data Analysis to Spectroscopic Data

A classification and identification problem is solved in the case of functional data. Functional data are data records considered as functions rather than vectors of observations at different values of the argument. This research deals with the spectral lines of rocks, being treated with light in the visible and mid-infrared regions. In a first step Principal Component Analysis in the functional data case will be applied to spectral lines of rock samples of various rock types to find out, whether the spectral lines can be separated.

Vera Hofer

Integrating Exchange Rate Theory in Data Mining

This paper focuses on an integration of exchange rate theory in a data mining process for the purpose of forecasting. The applied approach is centred by a Genetic Algorithm (GA) and Neural Networks (ANN), which allows the identification of relationships that are not describable by economic theory. As experience showed, relationships derived by data mining are often not convincing as regards their correctness and effectiveness. Most data mining approaches do not contribute much to persuade otherwise. In this work, it is tried to remove parts of this limitation by combining economic theory with data mining. However, usually the role of economic theory in exchange rate forecasting is to identify a list of relevant variables to be included in the analysis, with possibly and plausible signs of their coefficients. Previous research documents the failure of this way of analysis and thus for exchange rate theory in forecasting exchange rates at frequencies up to one month. Consistent with these results, in this paper the role of economic theory is extended as it is implemented in data mining as a framework in which and among which the possibilities of data mining are exploited. Other findings include: (i) During the years 2000 and 2001 countries relative economic growth is most significant in forecasting exchange rates one period ahead, (ii) the fmancial market is of major concern when explaining exchange rate movements as it may be used to proxy real economic activity as well as it maps massive capital flows between countries, which in turn affect exchange rates. (iii) fundamental forecast ing is more effective on lower frequencies. The approach is illustrated in some detail for five exchange rates on a daily, weekly and monthly frequency.

Bernd Brandl

The Ability of Artificial Neural Networks to Exploit Non-Linearities by Data Mining Models Compared to Statistical Methods

This paper discusses the question whether Artificial Neural Networks(ANNs) have the capacity to exploit (additional) non-linear information on the basis of exchange rate forecast models selected on linear criteria. This includes a comparison of exchange rate forecasts between ANNs and linear statistical methods and it is asked whether linear relationships serve as an approximation. The focused forecast models are selected by a data mining approach which combines fundamentals from economic theory, respectively building blocks from economic theory, with “fundamentals” derived solely by statistical criteria. This combination of theoretical and statistical relationships in data mining makes sure that both long run determinants on exchange rate behaviour as well as current influences are integrated in the forecast models. The results are evaluated for five exchange rates on a monthly frequency. The results favour the use of ANNs as they slightly improve the out-of-sample performance of the forecast models. What is more, linear as well as non-linear methods can be applied and the advantages from both methods can be used, which means that statistical methods allow a more detailed analysis of the results whereas ANNs offer a slightly better forecast performance.

Lutz Beinsen, Bernd Brandl

Preference Measurement with Conjoint Analysis and AHP: An Empirical Comparison

Conjoint analysis (CA) and analytic hierarchy process (AHP) are common methods for measuring preferences with CA dominating marketing research and practice and AHP becoming more and more relevant as a tool of decision analysis.

Roland Helm, Laura Manthey, Armin Scholl, Michael Steiner

Further Development of MADM-Approaches in China and in Germany

This paper seeks to give an overview and compare the main stream of thought in Multi Attribute Decision Making (MADM) theory and practice in China and Germany. MADM approaches are suitable for evaluation of a set of discrete alternatives. Widely applied approaches, including Multi Attribute Utility Theory (MAUT), Analytic Hierarchy Process (AHP) and the outranking methods ELECTRE, PROMETHEE, TOPSIS, find applications in the strategic production planning, site selection or technique assessment under economic, ecological and technical attributes. In China, MAUT and AHP are predominantly applied, while in Germany outranking methods are more popular. As further development,the integration of Fuzzy theory and artificial intelligence in MADM are discussed,thus imprecise information can be considered.

Jutta Geldermann, Kejing Zhang, Otto Rentz

Wissensrevision in einer MaxEnt/MinREnt-Umgebung

Die Abbildung menschlichen Wissens und seine maschinelle Verarbeitung sind ein wichtiger Forschungsgegenstand auf dem Gebiet der Künstlichen Intelligenz (KI). 1m allgemeinen unterliegt das Wissen zeitlichen Veränderungen, so dass zu einem säpteren Zeitpunkt nach Erfahren neuer Sachverhalte bisher bekannte nicht mehr gültig sind und das “alte” Wissen revidiert werden muss. Zur Formulierung von Sachverhalten in einer Domäne werden in diesem Aufsatz pr obabilistische Konditionale verwendet, mit denen eine Sprache zur Kommunikation zwischen Mensch und Maschine zur Verfügung st eht, die eng der menschlichen Denkweise angelegt ist. Das auf diese Weise formuliert e Wissen wird dabei informationstreu durch das Entropieprinzip verarbeitet. In dem vorliegenden Aufsatz wird gezeigt, wie der Wissensrevisionsprozess in einer ent ropieopt imalen Umgebung durchgeführt wird. Anhand eines ökonomischen Beispiels wird der Wissensrevisionsprozess näher illustriert; sämtliche hierfiir erforderlichen Berechnungen warden in der Expertensyst em-Shell SPIRIT vollzogen.

Elmar Reucher, Wilhelm Rödder

Innovation, Operations Research & Decision Support in the Military

Operations Research as a scientific discipline initially started some decades ago within the Military, focussing that time on logistics and questions about effective distribution of resources.

Heiner Micko

Von der Prädikatenlogik zur unternehmerischen Entscheidungsunterstützung

Mit der Expert ensystemshell SPIRIT steht ein mächtiges Instrument zur unternehmerischen Entscheidungsunterstützung auf der Basis probabilistischen konditionalen Wissens zur Verfügung. In mehreren Arbe iten - wie zum Beispiel [5], [3]-wurde über ihren professionellen Praxiseinsatz berichtet. Wegen der hohen Komplexität der Entscheidungsmechanismen bedarf es zumeist intensiver Gespräche bis zur Akzeptanz der zugrunde Iiegenden Theorien; letzte Zweifel werden jedoch häufig nicht ganz ausgeräumt. Mit diesem Beitrag wird nun für die Verwendung von SPIRIT ein neues Einsatzgebiet gewähit und damit ein noch nicht da gewesener Zugang geschaffen.Konditionalmodelle sind auch zur Lösung prädikatenlogisch er Aufgaben geeignet,wie sie beispielsweise in sogenannten Logikrätseln gestellt werden.Damit wird gleichsam für derartige Aufgaben eindrucksvoll nachgewiesen,dass die Fähigkeit des Menschen zur Deduktion von Wissen aus Wissen auf dem Computer formalisierbar ist.Nach der verbalen Formulierung einer beispielhaft ausgewählten prädikatenlogischen Aufgabe wird diese in der Shell SPIRIT kausal modeIIiert und fiir verschiedenartige FragesteIIungen eindeutig gelöst.1m letzten Teil des Beitrags wird durch Aufzeigen von Analogien deutlich, dass der Weg von »Logeleien« zu ernsthaften ökonomlschen Anwendungen nicht weitist.Mit dem Verstandnis des Mechanismus zur ModeIIierung und Lösung prädikatenlogischer Aufgaben kann so vielleicht der Widerstand gegen die Verwendung robabilistischer, mathematisch fundierter Konzepte abgebaut werden.

Friedhelm Kulmann, Wilhelm Rödder
Weitere Informationen