Research Goals and Strategies of the Chair of Business Administration, esp. Information Management

Information and communication systems are a prerequisite for basically every production, exchange, and coordination process. Having "grown up" with an ever increasing applicability of IT and recently telecommunication networks, we have developed in close cooperation with industry partners a variety of prototypes of novel information systems exploiting network effects and providing decision support for corporate planning and infrastructure decisions. This requires stable theoretical foundations which we continuously strive to improve and adapt to emerging phenomena associated with working in networks.

Nowadays, the network metaphor spans all virtual and physical systems of interdependent agents, one important network being the so-called information society focussing on the impact of ubiquitous connectivity of all areas of private and public life. A particular property of these networks is the existence of network effects describing the synergies between network actors. Thus, network effects imply a coordination problem that might be fundamental to solve for efficiently structuring information societies.
In recent years, particularly triggered by the "Internet lessons", the theory of positive network effects has emerged as a new area of research aiming at explaining the implications on market coordination and efficiency (project Economics of Standards). Especially IT products and standards are considered to be subject to strong network effects. We aim at contributing building blocks to a prospective unified theory of networks. While the overall goal must be the development of an axiomatic theory of network coordination, our particular research goal at the Chair of Economics, especially Information Management can be described as developing and evaluating different coordination designs for an Information Society.

Of all the coordination problems, our research focuses on "business problems", i.e. n-player games in a discrete decision world with transferable utility. Within that context, we focus on two classes of problems:

1. temporal resource dependencies
2. controlling infrastructure decisions

a) Mechanisms for controlling temporal resource dependencies
Prominent examples of temporal dependencies are job shop schedules: Which production steps (orders) are produced on which resource (e.g. machine) at what time?
Regularly, based on an agreed-upon schedule, gaining additional advantage for one customer (or for one order) by e.g. reducing its latency time results in additional costs for another customer (order) having to release "his" time slot. However, it might be the case that the costs of a delay of the pushed-away customer (order) momentarily is lowered significantly because, for example, he encountered a late delivery notice of one of his other suppliers. Generally speaking: we must develop mechanisms to exploit these types of temporal interdependencies and stochastic demand between actors.
We develop, evaluate, and deploy mechanisms to control temporal interdependencies between actors in three projects (as of July 2001):

DispoWeb (Protocols for distributed disposition of resources in supply webs): Given a schedule, how to find optimal temporal exchanges of time slots of scarce resources, e.g. by auctioning? We work in this field since 1995.
PRISE (dynamic pricing of bundles of goods or rights): How to combine auctions of interdependent goods (e.g. subject to intertemporal network effects)? We work in this field since 1989.
Yield Management Systems: How to determine optimal prices for products subject to tightly limited (perishable) production capacities, e.g. airplane seats or information? We work in this field since 1989.

b) Mechanisms for controlling infrastructure decisions
Some markets are determined by strong positive network effects (demand-side economies of scale), deriving from the need for compatibility. This means for example that the willingness to adopt a technology correlates positively with the number of existing (or expected) users. Popular examples are the information technology and telecommunication markets. The network effects in these markets mainly originate from two different areas, the need for compatibility to exchange information or data and the need for complementary products and services. In economic terms, ubiquitous networks and global competition have enterprises face new challenges finding adequate cooperation designs. The network metaphor describes efforts towards integrating partners and their data as well as their processes into common information infrastructures in order to facilitate the automated execution of business processes throughout entire value chains.
We develop, evaluate, and deploy mechanisms for supporting infrastructure decisions in five projects (as of July 2001):

IT standards and network effects: How to efficiently plan and control systems subject to network effects? We focus on: Economics of standards, technology diffusion and the exploitation of network effects, coordination of infrastructure decisions in communication networks, XML (Extensible Markup Language) / EDI (Electronic Data Interchange) and the integration of small and medium sized enterprises (SME), contributions to an interdisciplinary network effect theory. We work in this field since 1995.
Information logistics: How to optimize the relation between physical and virtual network layers? Our research comprises all information infrastructure, e. g. router, server, software, integration of supply web partners complementary to the respective (physical) supply web in order to exploit yet uninternalized network effects. We work in this field since 1996.
Enterprise Application Integration (EAI) and E-Security: What are efficient technological and organisational integration strategies? EAI is based on technological as well as organisational process synchronization, and a secure EAI infrastructure, for example building on EJB-security and integration middleware, is fundamental. Since 2000 we run the Center of Excellence for E-Security which bridges technological and economic aspects in developing secure EAI infrastructure concepts and components. Since 1997, our Competence Center XML provides information and resources concerning XML/EDI and business integration.
Moreover, we strive to evaluate novel information and communication systems, the underlying infrastructure imperatives and related business models e.g. concerning wireless systems. (new chair of M-Commerce)

c) Fundaments
Striving for novel solutions for both classes of problems, our research is based on common theoretical fundaments and knowledge description and exploitation systems:

a) Common theoretical fundaments: We use existing theories, identify lacks of their explanatory power, develop extensions to fill the gap, and test the new approaches in practice.

We build our research on game theory as a theory of social interaction. Game theory provides the basis for explicitly modelling the actors' (interdependent) micro-behaviour, enabling the development of computational models of economic systems (http://www.econ.iastate.edu/tesfatsi/ace.htm).To simulate the dynamics of complex and adaptive systems, we employ software agents which contain economic states and induce transitions (actions). The methodological paradigm is the emerging field of Agent-based Computational Economics (ACE) (seminar).

Another fundament of our research is network effect theory. Discrete decisions, path-dependent properties of network effect systems and bounded rationality on behalf of the deciding actors require simulation approaches. In particular, micro-simulations are a promising means to analyze dependencies of different actors striving for individual objectives and their implications on system behaviour.

b) Common knowledge description and exploitation: We focus on using existing concepts and products and develop extensions to fill conceptual and functional gaps.

XML has quickly gained a reputation as lingua franca of the web. It allows the separation of presentation formats from descriptions of the semantics of the data presented. We run the XML competence center. We work in this field since 1997.

OIL: The Ontology Inference Layer OIL is a proposal for a web-based representation and inference layer for ontologies, which combines the widely used modelling primitives from frame-based languages with the formal semantics and reasoning services provided by description logics. It is compatible with RDF Schema (RDFS), and includes a precise semantics for describing term meanings (and thus also for describing implied information). We work in this field since 1997.