New perspectives on organitzational design based on the analysis of complex communication networks

Abstract

The typical chemical company is large, usually with thousands of employees. According to data from the European Union, in 1990 almost 70% of the total turnover generated by the chemical industry corresponded to companies with more than 250 employees. The remaining 30% corresponded, in similar amounts, to small companies with less than 50 employees and to medium sized companies with 50 to 250 employees. Indeed, although some products have regional markets, chemical industry is essentially global and is dominated by large multinationals like Bayer, with 117,000 employees, BASF, with 93.000 employees, DuPont, with 79.000 employees, or Dow Chemical, with 50.000 employees. Specially for such large companies, organizational design and human capital management play a key role, as important, at least, as technology or management of material resources. A substantial part of the human workforce of such a company is devoted to information processing rather than to "make" or "sell" products in the narrow sense. However, most formal analysis of organizations have downplayed communication and information processing and focused on issues related to individual incentives. Only in the last decade, the importance of communication processes in organizations has started to be understood, mainly in the economics literature. Parallel to these efforts to understand the role of communication in organizations, the appearance and fast development of huge technology-based communication networks such as the Internet, as well as their inherent complex structure and dynamics, has contributed to awaken the interest of the scientific community in the so-called ``complex networks'. Actually, the study of networks was already a topic by itself in social sciences and in mathematics. However, recent studies on these technology-based communication networks as well as the discovery of surprising properties in big and complex networks in fields as diverse as biology, physics, computer science, engineering or economics, has generated a great interest. In particular, statistical physics has played a particularly important role in understanding some of the properties of such networks. The reason is that some of the tools derived to understand complex collective behavior in physical systems (that differ from the addition of the individual behaviors of the parts of the system) are applicable in the field of complex networks. The present work uses ideas from both the economics literature and complex networks literature to understand the role of communication processes in organizations. The problem is tackled from a double perspective: theoretical and empirical. From the theoretical point of view, we propose and study a general and simple model for communication processes. With the understanding obtained from the model, we address the problem of finding optimal communication networks. From the empirical point of view, we study the complex communication network of real organizations and we obtain information about the structure of the different communities in the organization. This information can be used as a quantitative and objective indicator of the status and evolution of the organization.