Type:

Master Thesis Business Information Systems

Status:

in process

Tutor:

• Dr. Monika Kaczmarek-Heß

Abstract

Motivation

In recent years, research on Artificial Intelligence (AI) has put forward remarkable results. The public sphere has particularly recognized advances in the areas of autonomous driving, retail recommendation systems, political forecasting, and AIs for strategic games. For example, a few years ago the world champion in the strategic game of Go, one of the few strategic board games where humans used to be stronger than computers, was defeated by an AI system.

Artificial intelligence is also increasingly applied in organizations to support, or even automate, decision making at different levels. But despite growing capabilities of AI technologies, there are still numerous preconditions that have to be fulfilled before AI can be employed to deal with complex organizational problems. Some of these conditions can be traced back to issues that have already been recognized at an earlier stage of AI research in the 1970s, where enthusiastic expectations about, for example, an AI-based "General Problem Solver" (Simon, Newell) did not succeed. Given these difficulties, (IT) managers are facing the challenge of deciding where, and in what form, AI-based systems can be meaningfully introduced in the organization to improve the effectiveness of decision making.

Description

This thesis is meant to address three goals. First, the thesis should provide a structured and critical overview of key concepts, technologies, and assumptions of artificial intelligence research. This should also involve a brief account of the historical development of the research field. Second, the thesis should critically examine the potentials of AI for the support or even the automation of organizational decision making. This examination should be based on a proper review of insights about conditions and characteristics of real-world organizational decision processes. Finally, the thesis should develop a new framework to support organizations in the preparation or execution of projects meant to introduce AI for purposes of decision support or automation. The specific purpose and nature of the framework can be defined depending on the interests of the student and the focus of the Master's thesis. For example, it would be possible to develop a basic framework with checklists and heuristics to assess whether the introduction of AI makes sense in a given organizational setting. It would also be possible to develop a full-fledged domain-specific modeling language (DSML) to describe the purpose and desired effects of an AI project in the context of a larger enterprise model.

Literature

• Silver D, Huang A, Maddison CJ, Guez A, Sifre L, Driessche G van den, et al. (2016) Mastering the game of Go with deep neural networks and tree search. Nature 529(7587): 484–489.
• Marwala T (2014) Artificial Intelligence Techniques for Rational Decision Making. Cham: Springer.
• Frankish K and Ramsey WM (2014) The Cambridge Handbook of Artificial Intelligence. Cambridge: Cambridge University Press.
• Bonabeau E (2009) Decisions 2.0: The Power of Collective Intelligence. MIT Sloan Management Review 50(2): 45–52.
• Flasinski M (2016) Introduction to Artificial Intelligence. Cham: Springer.
• Russell SJ and Norvig P (2016) Artificial Intelligence. A Modern Approach. Boston: Pearson.
• Goertzel B and Wang P (2007) Advances in Artificial General Intelligence: Concepts, Architectures and Algorithms. Proceedings of the AGI Workshop 2006. Amsterdam: IOS Press.
• Dreyfus HL (1994) What Computers Still Can't Do. A Critique of Artificial Reason. Rev. Ed. Cambridge: MIT Press.
• Searle JR (1990) Is the Brain's Mind a Computer Program? Scientific American 262(1): 25–31.
• Frank U (1988) Expertensysteme: Neue Automatisierungspotentiale im Büro- und Verwaltungsbereich? Wiesbaden: Gabler.
• Zelewski S (1986) Das Leistungspotential der Künstlichen Intelligenz. Eine informationstechnisch-betriebswirtschaftliche Analyse. Band 1, 2 und 3. Zugl.: Dissertation, Universität zu Köln. Bonn: Wehle.
• Newell A and Simon HA (1972) Human Problem Solving. New York: Prentice-Hall.
• Newell A, Shaw JC and Simon HA (1958) Elements of a Theory of Human Problem Solving. Psychological Review 65(3): 151–166.