Design processes of complex products are riddled with uncertainties, making them notoriously difficult to plan. Processes inevitably involve task failure and iteration. Yet most design processes are planned as if each task had a fixed duration and number of expected iterations, largely the result of current design process planning techniques such as Gantt charts and PERT charts. This paper reviews alternative design processes modeling approaches. These include ‘signposting’ models that describe tasks in terms of the quality of their input parameters and the probability of reaching a certain quality of output parameters. Such models enable design process performance to be simulated, giving managers an in depth understanding of likely process behavior. However, simulations do not normally provide a clear schedule of tasks to assist subsequent planning. This paper addresses this issue and describes how information regarding task dependency and sequence, which is salient in a conventional process description, can be extracted from probabilistic simulations.Copyright © 2004 by ASME
Stored only when both papers are in the corpus — 34 citations exist upstream.