For assessing the lifetime of highly reliable and long lifetime products, the stepstress accelerated degradation test is an economical, quick and useful tool. To handle the problem of settings of sample size, measurement frequency and number of measurements for stepstress accelerated degradation test, a generalized inverse Gaussian process was used to describe the degradation path. A modified Monte Carlo method was applied to obtain the estimation of the parameters. Under the constraint about the total experimental budget, sample size and termination time, the optimal configuration for stepstress accelerated degradation test was determined by minimizing the asymptotic variance of the estimated quantile of the lifetime distribution of the product. Finally, an example was given to illustrate the proposed model and method.