Tolerance analysis is indispensable in modern manufacturing processes. Thus, practitioners have many methods and tools at their disposal. Unfortunately, differences in methods often result in different outputs – depending on the problem, some of the methods may be more convenient than the others. A case study has been performed, aiming to compare the performances of four different tolerance analysis methods. Two assemblies were used to compare the tolerance chart analysis, Monte Carlo simulation, vector loop, and unified Jacobian-torsor method. Each method was explained in detail, and the steps needed to carry it out were shown. The results have shown that unified Jacobian – torsor method is the most comprehensive, but also has the most complex procedure. Monte Carlo simulation is suitable for analysis of parts produced in large series, mostly due to its statistical approach. Lastly, vector loop and tolerance chart analysis displayed similar results, with the former being suitable for solving 2D and 3D problems.
Petar Kosec
2019
Master thesis
