A method for managing uncertainty levels in design variables during complex product development

Abstract

Uncertainty is a constant concern in the design and development of complex systems. Managing it is important for organizations since it fosters design process improvement and optimization. This paper presents a method aiming to support large organizations understanding, quantifying and communicating uncertainty levels that normally arise during the early stage of the design process. This stage is particularly affected by decision-making uncertainty because designers and engineers have not yet acquired sufficient knowledge or possess sufficient confidence to decide what is the precise value that should be assigned to design variables in order to satisfy the customer's needs in an optimal manner. This type of uncertainty has been coined in literature as imprecision. The method proposed in this paper relies on the statistical characterization of the typical level of imprecision that should be expected by designers and engineers based on the collection of historical records of change from past product development processes. We quantify the level of imprecision based on two proxy variables: The time between changes and the magnitude of change that can be expected in new projects. In addition, our method proposes a Matrix of Imprecision that is capable of communicating uncertainty levels to the participants involved in new product development projects.