Ishii et al. [15,40,41,42] have developed design-compatability analysis tools to aid in designing products for various life-cycle considerations. In their approach, a set of design elements is defined for each life-cycle application. While the designer interactively identifies these elements in a proposed design, she is prompted to provide information about user and functional requirements. Their system uses a compatability knowledge-base to evaluate tradeoffs between various design elements and functional requirements. A compatibility knowledge-base is a collection of domain-dependent rules used to calculate a compatibility index. If a design attribute receives a poor compatibility index, the system offers advice by illustrating predefined cases that result in good compatability.
Ishii and his colleagues have built a number of design advisory systems using this approach. This task of designing parts and specifying interactions among the features can be tedious and the system can be improved in that regard. Also the methodology does not appear to be suitable in cases where the interaction among features are very high or the design might have multiple interpretation.
The work of Huh and Kim [43] describes a system for supporting concurrent design for injection molding. Their interactive expert system encodes rules for different molding materials and supports the synthesis of supplementary features to be put on to the initial design. The system aides the designer when performing tasks such as rib requirement, rib cross-section, rib frequency and in design of bosses. Both function and manufacturability are considered when providing help for these decisions. Interactive feedback is provided to the designer in two forms. First is the probability of having different forms of manufacturing defects, such as sink marks, warpage, or ejection difficulty. The second type of feedback is in the form of a warning message which suggests possible problems for the designer to avoid. The feedback is quantitative, in the sense that it gives probability of occurrence of common manufacturing defects. However this information is hard coded in the rules and the numbers that are calculated can only reflect the cases considered by the system.
For net shape manufacturing operations (e.g. casting, stamping, injection molding, sheet metal working) several manufacturability evaluation systems have been developed. Most of these systems use rule-based approaches to examine the violation of design-for-manufacturability heuristics. For example, Lazaro et al. [44] have developed a system for finding violations of design-for-manufacturing rules for sheet-metal parts. Rosen et al. [17] have developed a system for injection molding and die casting.
More work on general net-shape manufacturing is reported in three parts by Wozny et al. [45,46,47]. Their approach is broad and more complete than most others and considers multiple manufacturing processes when evaluating components. Evaluation is done hierarchically during the configuration and detailed design stages. In addition, they consider the functionality of the parts, tolerance information and also provide redesign suggestions. Finally, they also consider assembly of the components. Their approach integrates many phases of the design and manufacturing process.
Dissinger et al. [48] have developed a three-dimensional modeling system for designing powder metallurgy components. The design process follows the basic characteristics of a part created by powder metallurgy: the part is created layer by layer and, with the addition of each layer or a component to a layer, checks are made for possible manufacturing rule violations. The system is interactive, alerting the designer of the rule violations and giving suggestions for modifications. Finally the system allows only the design of manufacturable components.
Bourne [49] reports work for an intelligent bending workstation. Being developed in the same line as intelligent machining workstation developed earlier, they are implementing an open architecture model for the bending controller to overcome the common difficulties posed by closed controllers in NC machines. This system will be customizable and extendable for future addition of other modules.
Balasubramaniam et al. [50] proposed a general method for developing producibility metrics for process-physics dominated production processes such as extrusion, injection molding etc. Their approach predicts the likelihood of common manufacturing defects based on different physical characteristics of the design. As an example they developed metrics for various types of defects in extruded aluminum components for aircraft. They conducted experimental and statistical verification of the metrics based on actual vendor data.
There are additional works reported by researchers on various types of net shape manufacturing, including injection molding [51,52,53,54,55], die casting [56], sheet metal work [57,58], casting [59], powder metallurgy [60], extrusion [61] and stamping [62].