Adaptable, flexible and evolvable manufacturing systems and warehouses are so complex to manage that control systems have to be divided into several aspects. One of these is internal transportation, which has to do with all tasks involved in fulfilling a set of so-called transportation orders, i.e. commands to collect and deliver material from origin to destination spots. A common approach to design the controllers for these applications begins by modeling them as multi-agent systems and continues to final deployment through a cascade of transformations. To minimize development costs of internal transportation controllers, we have proposed a model of construction that includes components that synchronize the events from reality simulation and the ones from actual reality. By using these synchronizers, further development is required only for those parts of the initial multi-agent controller models with real counterparts. In this paper, we review the model and the architecture of the proposed internal transportation system controllers and we illustrate the whole design process through the development of a controller for an automated laboratory. Indirectly, we prove the validity of the architecture and of its key component, the synchronizers.
|Translated title of the contribution||Running Agent-based-models Simulations Synchronized with Reality to Control Transport Systems|
|Number of pages||14|
|Publication status||Published - 2016|