Due to the ever-increasing competition in today’s global markets, the cost of the product is rapidly emerging as one of the most crucial factors in deciding the success of the product. Decisions made during the design stage affect as much as 70–80% of the final product cost. Hence, a manufacturing cost estimation tool that can be used by the designer concurrently during the design phase will be of maximum benefit. A literature study of the available cost estimation tools suggests that a majority of these tools are meant for use in the later stages of the product development lifecycle. In the early stages of a product lifecycle, the only information that is available to the designer is related to geometry and material. Hence, the cost estimation methods that have been developed with the intent of being used in the early stages of design make use of the geometric information available at that stage of the design. Most of the earlier models that use parametric cost estimation and features technology consider the design features in their implementation. However, such models fail to consider “manufacturing based features” such as cores and undercuts. These manufacturing based features are very important in deciding the manufacturability and the cost of the part. The Engineering Cost Advisory System (ECAS) is a knowledge-based system that presents cost advice to the designer at the design stage after considering various design parameters and user requirements. Some of these design parameters can be extracted via standard Application Programming Interfaces (APIs). Moreover, ECAS uses innovative techniques of geometric reasoning and the hybrid B-rep-voxel model approach to extract manufacturing feature-based geometric information directly from the CAD input. By considering the manufacturing based features along with the design parameters, the ECAS architecture is applicable to a much wider variety of manufacturing processes. The complexity of the part, which is derived from the geometric parameters (manufacturing based and design based) and other non-geometric user requirements (e.g. quantity, material), is used to estimate the manufacturing effort involved in process specific activities. The final cost is then estimated based on this manufacturing effort and considering the hourly rates of labor and other contextual resources as well as material rates.
A Genetic Algorithm and Support Vector Regression based Hybrid Cost Estimation Model for Feature-based Plastic Injection Products
