مطالعه تجربی از شبکه های سیستم کار در محیط تولید

In the context of distributed manufacturing, existing models for the management of orders and work systems assume that the entirety of information about the state of the production environment is known. Instead, this paper studies the implications of networking in an environment in which no element possesses information about the entire state. An experiment in the form of a production simulation game is designed and carried out to explore the emergence and behaviour of such a network. Network structure and dynamics as well as the social aspects of networking are discussed and evaluated.

Global competition and the increased complexity of the production world are forcing companies to become more flexible and responsive. As a consequence, companies are organising into production networks–cross-company cooperations enabling mutual use of resources and joint planning of the value adding process [1]. While the operation of production networks has been studied extensively, the underlying mechanisms of networking remain relatively poorly understood for three reasons. First, networking is not only a technical but also a social process involving human subjects and is therefore difficult to model. Second, real-world experiments are almost impossible to conduct because they represent risks for the companies involved. And third, real networks often behave as complex adaptive systems [2] and [3], where the phenomenon of emergence is difficult to describe mathematically. In this paper, we apply the methods of experimental economics to the study of the networking process in a production environment. Specifically, we design and carry out an experiment in the form of a simulation game to study decision making, coordination, and social preferences involved in the networking process of work systems [4]. We test two hypotheses related to social aspects of networking and network robustness. We show that social preferences significantly influence networking and that network performance is minimally affected by external disturbances. The results highlight the potential of the approach for use in the study of various production phenomena.

A production experiment in the form of a multiplayer game shows that work systems interconnected in a network are more resistant to perturbation than a group of disconnected work systems. A network of work systems is both a competitive and cooperative production environment where work systems can also support each other. The work systems connections are dynamic and are affected by the opportunism, fear, and trust of individuals. In difficult or uncertain times, connections are predominately based on social preferences. Participants of the experiment demonstrated with their behaviour that people instinctively understand the advantages and disadvantages of connecting in networks. Future work should be composed of more experiments on larger populations and for longer durations. Such laboratory experimentation may provide a path towards real world experimentation and practical applications of the dynamic work system networking.