Efficient computing budget allocation for a single design by using regression with sequential sampling constraint

In this paper, we develop an efficient computing budget allocation rule to run simulation for a single design whose transient mean performance follows a certain underlying function, which enables us to obtain more accurate estimation of design performance by doing regression. The sequential sampling constraint is imposed so as to fully utilize the information along the simulation replication. We formulate this problem as a c-optimal design problem based on some common assumptions in the field of simulation. Solutions are generated for some simple polynomial, logarithmic, and sinusoidal functions. Based on the numerical solutions, we develop the Single Design Budget Allocation (SDBA) Procedure that determines the number of simulation replications we need to run, as well as their run lengths, given a certain computing budget. Numerical experimentation confirms the efficiency of the procedure.