An evolutionary and interactive approach to simulation of assembly planning in virtual environment

This paper introduces a methodology for collaborative design and planning using an evolutionary and interactive approach which has been developed to address both assembly modeling and assembly sequence planning in an interactive virtual environment. Starting with product CAD data and feature model, it generates the best feasible assembly sequence. It generates assembly rules and precedence constraints from the CAD data that defines the connections between the product parts, modifies random sequences to feasible assembly sequences using the rules and constraints database, and finds an optimum assembly sequence by using a modified genetic algorithm. To evaluate the feasible assembly sequences, assembly operations are evaluated considering factors like connection types, directionality, stability and manipulability. An efficient user interface in the virtual environment increases the credibility of the system by allowing the user to interact with the system. The user, considering the manufacturing setup and other conditions assesses the cost of the assembly operations and gives feedback to the system. An industrial example is solved by this new methodology