Numerous studies have investigated the buffer allocation problems (BAPs) and the serial production line balancing problems (SPLBPs) to improve system performance. There are two main disadvantages in many of these studies: (1) the BAPs and SPLBPs have been approached separately; (2) the objectives are calculated in traditional steady-state, and their errors are great when the volume of a production run is small. This paper considers the two optimized problems simultaneously in serial production line with transient analysis. The production run is finite, and the stations are in Bernoulli reliability. The objective is to maximize profit that includes both revenue per unit of throughput and cost per unit of storage (WIP and buffer space). A computationally efficient algorithm based on aggregation is developed to approximate the objective function. Then, a genetic algorithm is proposed to find an optimal task assignment and buffer allocation. The results of extensive experimentation demonstrate that some un paced unreliable serial production lines can never be entirely balanced, and using buffer can improve line efficiency.