In modern banks, establishing efficient and quality customer services plays a crucial role to achieve the business success. There are a large number of customers to be served daily; however, required service resources (in particular, customer service representatives) can be rather limited. Therefore, customer service representative scheduling is a challenging problem for bank managers. In the present paper, we focus on a customer service representative scheduling problem considering the stochastic customer arrival time and service duration to minimise the total cost including two parts: (i) the loss concerning the customer waiting and (ii) cost of providing service representatives. We formulate this scheduling problem as an integer programming problem and employ an adapted water wave optimization (WWO) algorithm to solve this problem efficiently. We adapt the propagation and refraction operators of WWO according to the solution space of this problem. The results of the conducted computational experiments demonstrate that the adapted WWO algorithm outperforms the considered state-of-the-art evolutionary algorithms in solving a set of test instances.