| Abstract: | To reduce the uncertainty and reworks in complex projects,a novel mechanism is systematically developed in this paper based on two classical design structure matrix(DSM) clustering methods:Loop searching method(LSM) and function searching method(FSM). Specifically,the optimal working areas for the two clustering methods are first obtained quantitatively in terms of non-zero fraction(NZF) and singular value modularity index(SMI),in which the whole working area is divided into six sub-zones. Then,a judgement procedure is proposed for conveniently choosing the optimal DSM clustering method,which makes it easy to determine which DSM clustering method performs better for a given case. Subsequently,a conceptual model is constructed to assist project managers in effectively analyzing the network of projects and greatly reducing reworks in complex projects by defining preventive actions. Finally,the aircraft design process is presented to show how the proposed judgement mechanism can be utilized to reduce the reworks in actual projects. |
