Locator Placement Optimization for Minimum Part Positioning Error During Machining Operation Using Genetic Algorithm

Fixture design is one of the main factors which affect the final product quality. Proper design of fixture plays an important role in ensuring the required tolerance of the product. Proper placement of locators is one of the prominent factors in fixture design. Locators are elastic: they deform under clamping and machining forces causing rigid body displacement of the workpiece which in turn affects the part quality. In this article, a 3-2-1 type of fixturing system having elastic locators around a considerably rigid rectangular workpiece is considered. A genetic algorithm is proposed, which uses a fitness function that evaluates the positioning error of the workpiece under external forces and torque. Among several variables, 12 variables, which define the placement of locators, are chosen to be optimized while minimizing the positioning error of the workpiece at the point of action of machining force. The proposed algorithm optimizes the 12 interlinked variables, within the specified region, for machining force and torque at a single point. However, when the cutting tool moves to any other point on the workpiece, it is observed that either the workpiece loses its contact with any one of the locators or the positioning error increases by a large value. To overcome this issue, the proposed algorithm is further modified for placement optimization to cater for multi-point machining, and the isostatism of the workpiece is ensured by checking the magnitude and direction of displacement (of what?) at each point of workpiece-locator contact. Finally, the original and modified GA algorithms are explained through a case study where the single point optimized placement shows loss of contact when machining force is applied at other points. The placement optimized from the modified algorithm shows that the isostatism of the workpiece remains intact while all four positioning errors are converged towards the same value. The results obtained from the proposed and modified algorithm are verified using ANSYS simulation.