With the increasing aging population, the nursing homes have developed into the serious needs of the society. Nursing work has the characteristics of “multi shift, high time-varying demand, hierarchical and collaborative”, Multi shift means that there can be multiple shifts in a day covering the same time period; high time-varying demand means that the number of nurses needed within 24 hours of a day will change greatly according to different time periods; hierarchical means that there are different levels of nurses’ abilities; collaborative means different levels of employees cooperate to serve the elderly. Our work devotes to the shift design problem(SDP) and the hierarchical staffing problem, i.e. how to design the time window of each shift to cover the demand periods, and how to allocate the hierarchical staff to the shifts. Our work will solve the problem by the two-stage method. In the first stage, according to the characteristics of high time-varying demand, we design the schedule of shifts and determine the number of personnel needed for each shift. In the second stage, we further determine the number of personnel needed for each shift at each level according to the constraints of hierarchical coordination and cooperation. The sensitivity analysis of different demand fluctuations is carried out and some meaningful conclusions are drawn. Our research outputs can provide efficient decision tools and methodology in daily staffing to the nursing homes which faces fast-rising demands and staff shortage.