The match point for business enterprises nowadays is decided by speed. It is very important for fashion clothes to control time; therefore, the textile industry must shorten its cycling time to meet the demand for the apparel business. At past the production system of dyeing process was mass production that usually adopted the allotted dyeing vats in order to omit the time of setup cost. However, at present the main production is a make-to-order system; high indetermination will cause unbalance of the dyeing machine and lead to low efficiency. This research is to exchange the dyeing vats and apply the genetic algorithm to obtain the minimizing objective of tardiness and setup cost. The performance of the genetic algorithm is compared with the traditional dispatching rules, EDD and SPT. All the results show that the genetic algorithm used in this study is superior to the other approaches.