A methodology for analysing large-scale concurrent engineering systems

A key problem in large-scale concurrent engineering systems is the control and management of the large amount of information required. Constraint networks that are used to represent this information can become so large that structural and operational problems are difficult to identify and rectify. This paper proposes a new approach, called Meta-nets, to modelling large-scale concurrent engineering problems. Meta-nets provide an abstraction of an underlying constraint network. The overall result is a compact and powerful representation of the information flow. The abstraction and analysis of Meta-nets gives several important benefits. First, by dealing with network structure based on information flow, Meta-nets can assist in network navigation. Second, Meta-nets give the dependence relationships between variables in the concurrent engineering system. Third, Meta-nets effectively represent parallelism (i.e. the carrying out of activities separately but with some measure of cohesion), which is an essential characteristic of concurrent engineering. This can form the basis for parallel processing. Fourth and finally, Meta-nets provide a methodology for breaking large concurrent engineering networks into more manageable sub-networks. These advantages become especially important for large-scale concurrent engineering problems.