Many modern information applications spread dynamic data objects and process complex queries for retrieving multiple data objects. This paper studies the broadcast scheduling design for dynamic information dissemination. We formulate the problem of scheduling sequential data objects in an on-line fashion and derive the lower and the upper bounds of mean service access time. Accordingly, we further present several heuristic functions to approximate the schedule optimization. The experimental results show that the proposed design is able to on-line generate a broadcast schedule whose mean service access time is very close to the theoretical optimum.