Theses
Business informatics is a discipline that represents the interface between the different sub-areas of business administration. As such, the application areas in which methods and concepts of business informatics are applied play a large and often determining role. For this reason, theses in the field of business informatics are often considered interdisciplinary: Based on a business problem in a subfield of business administration, methods and concepts of business informatics are applied to solve the given business problem. Therefore, almost all areas represented at the Faculty of Business Administration and Economics can be considered for a thesis.
Questions from the field of operations research, which is related to business informatics, can also be dealt with in the form of a thesis. The business informatics-specific part of the thesis can vary and is discussed individually with the respective supervisor.
The listed topics can be dealt with in the context of Bachelor's or Master's theses. There are no fixed dates for the theses and the preceding project seminar; they can be started at any time. For more detailed information, please contact the designated supervisor.
We currently offer theses on the following topics:
Bachelor / Master
Case study: Merchandise management system at a retailer
The aim of this thesis is to investigate which merchandise management systems are suitable for local owner operated retail outlets (LOORO). For this purpose, a comparison of common free and fee-based merchandise management systems will be carried out. After a suitable system has been identified and selected together with a practice partner, which we will provide, the implementation will take place at the practice partner. In doing so, hurdles in the implementation are to be addressed and a critical conclusion on the suitability is to be drawn. The aim of this thesis is to provide LOORO with a case study that they can use to help themselves with digitalisation.
Problems of order management in the steel industry and possible solutions from a practical point of view
Steel is an important raw material for many industries such as the automotive industry or aerospace. Order management in the steel industry faces the challenge that the orders planned in a period of time must be scheduled in their composition and sequence on the individual production levels as cost-effectively as possible. Here, fluctuating demand, high price fluctuations, high fixed costs and data deviations make optimisation difficult. The aim of the work is to identify the problems of order management more precisely in cooperation with a regional steel producer and to outline initial approaches to solutions.
Using text-mining techniques to extract relevant information from texts for business use.
There is a huge amount of data on the internet. One difficulty is to extract relevant information from the mass of data. Text mining is one of the methods that can be used for this purpose. In the context of this thesis, NLP (Natural Language Processing) and text mining methods are to be used to extract relevant information from financial news, for example, and to use it for price forecasts (oil price, gold price, shares). Other possible application scenarios could be, for example, innovation research or online reviews. A possible focus could be opinion mining, for example, but other - preferably creative, novel - approaches are also conceivable.
Development and validation of a securities investment strategy with machine learning methods using the example of association rules
For decades, research has been concerned with the analysis and forecasting of stock prices and their movements. Most of the work can be divided into two schools: Technical and Fundamental Analysis. In this paper, a slightly different approach will be taken. It can often be observed that when certain events occur, shares of similar companies move in the same or opposite direction. These correlations are to be investigated in this work and, if possible, used and evaluated for the development of an investment strategy.
Default prediction of peer-to-peer loans with neural networks
This thesis aims to model defaults in peer-to-peer lending and to investigate the use of profit scoring. The working basis for this is an existing database on such lending.
Source literature: Serrano-Cinca, C., & Gutiérrez-Nieto, B. (2016). The use of profit scoring as an alternative to credit scoring systems in peer-to-peer (P2P) lending. Decision Support Systems, 89, 113-122.Lending club statistics. Available at: https://www.lendingclub.com/info/statistics
Process comparison for automated data cleansing: Detection of erroneous data sets.
Primarily aimed at students with programming skills, as efficient design of the simulation study (vzgws. R) is required. This work is to model exemplary processes that lead to faulty data records and to compare common methods for detection in the literature with each other in a controlled manner. For example, a supervised procedure for the detection of inconsistent data (univariate, LOF, etc.) will be compared with specialised procedures.
Breunig, M. M., Kriegel, H.-P., Ng, R. T., & Sander, J. (2000). LOF: identifying density-based local outliers. In Proceedings of the 2000 ACM SIGMOD international conference on Management of data (Vol. 29, pp. 93-104).
Ilyas, I. F., & Chu, X. (2015). Trends in Cleaning Relational Data: Consistency and Deduplication.
Business Data Science
Primarily aimed at students with prior knowledge of machine learning who want to write a practical thesis on a topic specified by a company within the framework of a work-study placement/internship/etc. and are looking for a supervising chair.
Baecker, Julius; Böttcher, Timo Phillip; and Weking, Jörg, "How Companies Create Value From Data - A Taxonomy on Data, Approaches, and Resulting Business Value" (2021). ECIS 2021 Research Papers. 124. https://aisel.aisnet.org/ecis2021_rp/124
Ahmed Abbasi, David Dobolyi, Anthony Vance, Fatemeh Mariam Zahedi. "The Phishing Funnel Model: A Design Artifact to Predict User Susceptibility to Phishing Websites". Information Systems Research 32 (2) 410-436 https://doi.org/10.1287/isre.2020.0973
Simulation and optimisation of data management in distributed systems with the help of AI methods.
In order to cope with Big Data requirements, data stocks are distributed across several computer nodes. The central research question of the thesis is how the distribution is carried out in concrete terms, on which nodes which data sets should be stored, possibly also in copy. The goals are, on the one hand, a high level of failure safety and, on the other hand, the rapid availability of data. For this purpose, a corresponding decision/simulation model is developed and solved with the help of AI methods (e.g. evolutionary algorithms).
Cost-optimised scheduling of orders in multi-stage production systems with AI methods
Scheduling large order quantities over a planning period of several weeks is a complex planning task. In order to ensure cost-effective order execution, the delivery dates must be taken into account in addition to the permissible sequences of operations, as well as a sensible combination of orders in order to save set-up costs. Furthermore, the available resources with their capacity limits must of course be taken into account. The aim of the work is to develop a planning/simulation model for this, which calculates a cost-optimal solution on the basis of AI algorithms.
SAP Focus: Migration
SAP is often used as the system architecture for controlling business processes. Here, companies face challenges when it comes to migrating existing systems. The thesis to be assigned aims to present a migration process to S/4HANA or the re-implementation on-premise or cloud-based.
SAP Focus: Data Warehouse
With the help of the SAP Data Warehouse Cloud (DWC) and its predecessor Business Warehouse (BW), real-time decisions can be made in companies. The aim of the thesis to be assigned is to gain an overview of the possibilities for analysing and modelling data in one place.
Organisation
If you would like to write a Bachelor's or Master's thesis, first look around in the list above for a topic that interests you. If you have found a potential topic, you can contact the specified supervisor directly. Alternatively, you can also submit your own topic suggestions, which can then be carried out after consultation with the professorship. Student suggestions are always welcome.
Study programmes
All students of a degree programme whose examination regulations permit a thesis in Business Information Systems can also write a thesis at the professorship. Students of
- Economics
- Industrial Engineering
- Business Mathematics
- Computer Science
can in any case write a final thesis at the professorship, whereby a professor of computer science is required as an additional examiner for students of computer science.
Theses in companies
In addition to the classical, theoretically oriented theses, we offer the possibility to work on theses in cooperation with companies. The exact topic is then closely coordinated with the company. If you already know a company where you would like to write your thesis, you are welcome to contact an assistant in this regard. Since we work closely with companies, we can also refer you to a company if necessary.
Requirements
Previous knowledge in the field of business informatics and the respective application area is desirable but not a prerequisite. However, students who have not taken any modules in business informatics during their studies, but who would like to write a thesis in one of the above-mentioned areas, are also invited to find out about the possibilities of working at the professorship. Due to the different research areas of the assistants, thesis topics can be found for many interested students - just contact an assistant.
You do not have to have taken a seminar with us to be able to write a thesis.