Increasing economic competition urges most companies to improve their marketing, sales, and customer support. Many firms are trying to create one-to-one relationships with their customers. They aim to make it easier for the customer to do business with them rather with a competitor and to estimate the lifetime value of each customer. The change in focus from broad market segments to individual customers requires learning from the data from and about the customers instead of simply gathering it or not gathering it at all. This is the goal of database marketing. Its essence is knowledge discovery in databases, i.e. "the non-trivial process of identifiying valid, novel, potentially useful, and ultimately understandable patterns in data". 

Data on customers is often exact, e.g. a customers age , the number of people which he or her shares the household with or the salary. On the other side there are vague concepts that lack in unambiquity, exactness or context independency like the customer's favourite dish, the state of his marriage or professional qualification. Furthermore, even exact data is often not necessary for an anlysis; a mail order business is not interested whether a customer earns 3,567 or 3,612 EURO a month but if his or her income is high or rather medium. An appropriate means for reflect such uncertain and vague concepts is fuzzy logic which was founded by Lotfi Zadeh in 1965. 

We were particularly interested in generating fuzzy rules from data and have written related articles, which were accepted for presentation at some international conferences. My work on this field is part of a project which a software house and an insurence broker are involved in. The project investigates methods for gaining new customers more effectively and cross selling.

One of the most important goals in marketing is to realize the highest profit by applying appropriate means to acquire customers, such as mailings, letters, telephone calls, offers and presentatons. In the past this business was the realm of men and women who "had a nose" for selling goods or services. More recently it has attracted also computer scientists, who support marketers by software tools for planning and organizing marketing campaigns or by more sophisticated methods like decision support systems or intelligent broker agents.

This study has been motivated while we are involved in improving the commercial software tool AkquiSys. The graphic user interface of AkquiSys is shown below.

A detailed description of the application can be gathered from the following publications

  1. L. Cromme, K. Weber, S. Würll: Entwurf einer Erweiterungskomponente für das Software-Werkzeug "AkquiSys" zum Auswerten von Akquiseplänen. (Outline of an extension of the software tool "AkquiSys" for analysing plans of acquiring customers. Report. Brandenburg Technical University at Cottbus, Chair for Numerical and Applied Mathematics, July 1999, Cottbus.
  2. L. Cromme, K. Weber, S. Würll: Entwurf einer Erweiterungskomponente für das Software-Werkzeug "AkquiSys" zum Simulieren und Optimieren von Akquiseplänen. (Outline of an extension of the software tool "AkquiSys" for simulating and optimizing plans of acquiring customers.) Report. Brandenburg Technical University at Cottbus, Chair for Numerical and Applied Mathematics, July 1999, Cottbus.
  3. L. Cromme, K. Weber, S. Würll: Einsatz eines Markov-Ketten-Modells zur Optimierung der Kundenakquisition - mit einem Fallbeispiel aus der Versicherungsbranche. (Application of a Markov chain model for optimizing process of acquiring customers - with a case study from the insurance branch) Report based on a talk at the DMV (German National Mathematical Society) annual conference, September 5-11, 1999, Mainz, Germany. 
  4. Klaus Weber, Sun Zhaohao: Fuzzy stochastic dynamic programming for process of acquiring customers. Accepted for 7th Zittau Fuzzy Colloquium, September 8-10, 1999, Zittau, Germany.
  5. Gavin Finnie, Zhaohao Sun, Klaus Weber: A Multiagent-based Intelligent Broker Architecture for Bargaining Processes. Submitted for The Australian Workshop on AI in Electronic Commerce, December 6, 1999, Sidney, Australia.
  6. Klaus Weber, Stephan Würll: Stochastic Multistage Process Simulation for Decision Support in Marketing. Submitted for 16th IMACS World Congress 2000 on Scientific Computation, Applied Mathematics and Simulation, August, 21-25, 2000, Lausanne, Switzerland.
  7. Stephan Würll, Klaus Weber: Optimal Management of Acquiring Customers by Means of Fuzzy Linear Programming. Submitted for 16th IMACS World Congress 2000 on Scientific Computation, Applied Mathematics and Simulation, August, 21-25, 2000, Lausanne, Switzerland.
  8. Stephan Würll, Klaus Weber: Optimal Customer Relationship Management by Means of Fuzzy Linear Programming. Submitted for 2nd International ICSC Symposium on Engineering of Intelligent Systems, June 29 - July 2, 2000, Paisley, U.K.
  9. Klaus Weber, Zhaohao Sun: Fuzzy Stochastic Dynamic Programming for Marketing Decision Support. Submitted for International Journal of Intelligent Systems, October 1999.
  10. Finnie, G.; Sun, Z.; Weber, K.: An Intelligent Broker-Centered Multiagent Architecture for Bargaining Processes. Submitted for The Fourth International Conference on Multiagent Systems, Boston, USA, July 7-12 2000.

Publications to this and other fields of research can be found at the papers & preprints page.

Co-operation: FOR/MAT Technology GmbH

Contact:  Prof. Dr. rer. nat. habil. Ludwig J. Cromme 
                                  Dipl.-Math. techn. Klaus Weber
                                  Dipl.-Math. Stephan Würll