Fuzzy decision making system and the dynamics of business games

Abstract

Effective and efficient strategic decision making is the backbone for the success of a business organisation among its competitors in a particular industry. The results of these decision making processes determine whether the business will continue to survive or not. In this thesis, fuzzy logic (FL) concepts and game theory are being used to model strategic decision making processes in business organisations. We generally modelled competition by business organisations in industries as games where each business organization is a player. A player formulates his own decisions by making strategic moves based on uncertain information he has gained about the opponents. This information relates to prevailing market demand, cost of production, marketing, consolidation efforts and other business variables. This uncertain information is being modelled using the concept of fuzzy logic. In this thesis, simulation experiments were run and results obtained in six different settings. The first experiment addresses the payoff of the fuzzy player in a typical duopoly system. The second analyses payoff in an n-player game which was used to model a perfect market competition with many players. It is an extension of the two-player game of a duopoly market which we considered in the first experiment. The third experiment used and analysed real data of companies in a case study. Here, we chose the competition between Coca-cola and PepsiCo companies who are major players in the beverage industry. Data were extracted from their published financial statements to validate our experiment. In the fourth experiment, we modelled competition in business networks with uncertain information and varying level of connectivity. We varied the level of interconnections (connectivity) among business units in the business networks and investigated how missing links affect the payoffs of players on the networks. We used the fifth experiment to model business competition as games on boards with possible constraints or restrictions and varying level of connectivity on the boards. We also investigated this for games with uncertain information. We varied the level of interconnections (connectivity) among the nodes on the boards and investigated how these a ect the payoffs of players that played on the boards. We principally used these experiments to investigate how the level of availability of vital infrastructures (such as road networks) in a particular location or region affects profitability of businesses in that particular region. The sixth experiment contains simulations in which we introduced the fuzzy game approach to wage negotiation in managing employers and employees (unions) relationships. The scheme proposes how employers and employees (unions) can successfully manage the deadlocks that usually accompany wage negotiations. In all cases, fuzzy rules are constructed that symbolise various rules and strategic variables that firms take into consideration before taken decisions. The models also include learning procedures that enable the agents to optimize these fuzzy rules and their decision processes. This is the main contribution of the thesis: a set of fuzzy models that include learning, and can be used to improve decision making in business.