Similar questions

State the effect of initialization bias in steady state simulation and how one can reduce this effect?

Initialisation bias in steady state simulations:

• Initial conditions may be artificial or unrealistic.

• Methods to reduce the point estimator bias include:

Intelligent Initialization:

• Initailize(start) the simulation in a state that is near expected of steady state( long run) conditions.

• Simulation takes some time to stabilize.

• There are two ways to specify intelligently the initial conditions

1. Observations of real system: It requires lot of efforts to collect large data. If the system is different to the existing one then it is impossible to implement.

2. If the system does not exist, use any data on similar systems or build a simplified model that is mathematically solvable and collect data from it.

• Divide each simulation run into two phases

1. First: Initialization phase from time 0 to time T0 .

2. Second: Data collection phase from time ‘T0’ to ‘T0 + TSE’.

• The selection of T0 is an important issue as the system state I at time T0, is proper representation of steady state behaviour than the original condition (I0) at time 0.

• Also, the duration of data collection phase, TSE, should be long enough so that the steady state behaviour must be sufficiently precise estimates.

• The system state I is a random variable. The probability distribution of the system state at time T0, is sufficiently close to the steady state probability distribution which makes the bias point estimates of the response variables negligible. So we can say that the system is approximately in steady state.