This paper proposes sheet thickness determination in manufacturing of laminated dies as an optimization problem. The aim of this optimization procedure is finding the best set of thicknesses which minimizes the volume deviation between actual computer‐aided design (CAD) model and assembled slices.
This works uses a modified version of genetic algorithms for the optimization purpose. Each set of thicknesses that can cover the whole CAD model surface is considered as a chromosome. Genetic operators such as crossover and mutation have to be modified to be used in this application.
A new method for finding the total volume deviation between assembled slices and the actual CAD model was developed in this research. On the other hand, the results show how the program can automate the slice plane locations search process.
Premature convergence does not allow the algorithm to search the entire solution space before getting trapped in a local optimum. Even the mutation operator cannot postpone this untimely convergence.
The proposed method is a good substitute for the manual methods that are currently used in industry. These experience‐based methods are mostly based on the decision made by a well‐trained technician on picking up the thicknesses for a specific CAD model.
This is the first attempt at optimizing the slicing method in laminated tooling. Other methods are mostly based on rapid prototyping (RP) and they are not applicable in the laminated tooling process since, despite RP, here not all optimization outputs can be used in practical procedure.
