| description abstract | An optimization model has been built with consideration of the required reliability, the minimum machining cost, and quality loss. The normal and the lognormal distributions of tolerances that depend on the production types of components are used in the reliability model. Cost tolerance data obtained from Bjørke are used to calculate the machining cost. The asymmetric quadratic quality loss model is used to calculate quality loss caused by the deviation and the mean-shift of distributions. Tolerance allocation of a sliding vane rotary compressor is optimized for the required reliability, the minimum cost and quality loss, and optimum tolerances of components are recommended. The results show that high accuracies of the slot length, the vane thickness, and the slot width are required. Hence, their tolerances are smaller than other components. The effects of the correlation coefficient of the bottom cover plate and the top cover plate and the correlation coefficient of the front cover plate and the rear cover plate to total cost are insignificant. Further, the cost of quality loss is reduced when the weighting ratio of the quality loss function weighting coefficient to the machining cost function weighting coefficient is increased. The total cost is increased because tight tolerance allocation increases the machining cost. | |
