Agent-based modeling for competing firms: from balanced-scorecards to multiobjective strategies

Abstract

This paper proposes a novel method for agent-based modeling in business management domain. We will model competing companies with the balanced scorecards principle and examine their value proposition strategies for customers. The proposed method is characterized by (1) designing decision making agents or competing companies with strategic parameters to be optimized; (2) employing a multiobjective optimization framework with genetic algorithms to evolve the artificial simulated society; (3) grounding the simulation conditions and results with marketing survey data in the real world, and (4) validating the strategic parameters of the agents after simulation via statistical analysis of the individual genes. The objective of our agent-based modeling is to explore 'optimal' marketing strategies on given specific markets. Competing companies will thrive by choosing their customers, narrowing their focus, and dominating their markets. We will uncover which type of companies provide good value propositions for customers from their activities. Conventional research in business strategy literature states the importance of translating the strategy of a company into action to get the profit. In our study, on the contrary, we will observe agents' actions or companies' activities in the artificial society with given conditions and investigate the agents' or companies' strategy. From intensive experiments using our agent-based simulator in a television set market, we have observed that 1) the price and service are important for benefit and cash flow maximization; 2) there are few dominant strategies for share maximization; 3) price and time will affect borrowing strategy, and 4) time is an important factor in a radio cassettes market. These results have suggested that our framework works well to analyze behaviors and strategies of competing firms.