In this thesis we consider a multi-class single-server queueing system, in which there are two types of customers with independent Poisson arrivals. Customers arrive at the system must wait for the service, and the single-server starts its service when the number of customer arrived is equal to the fixed-size service capacity. All customers are served in a batch by the single-server with exponential service times. Typical examples of the queueing system considered include public transportation service systems such as: shuttle bus transit system, railway transportation system, etc. The objective of this thesis is to solve the server scheduling problem for the queueing system considered in this thesis. There are two parts in this thesis: first, we use balance equations to derive the steady-state probabilities for the queueing system; second, we develop a cost model to find the optimal fixed-size service capacity in order to minimize the total system cost including the waiting cost of customers and service cost of the single-server.