Rockwell-automation Arena Contact Center Edition Users Guide User Manual download pdf (Page 23)

2 • INTRODUCTION TO SIMULATION

• • • • •

2 • Introduction to Simulation

Later, after we have developed the model, verified its correctness, and validated its

adequacy, we again need to consider the final strategic and tactical plans for the execution

of the experiment(s). We must update project constraints on time (schedule) and costs to

reflect current conditions, and we must impose these constraints on the design. Even

though we have exercised careful planning and budget control from the beginning of the

study, we must now take a hard, realistic look at what resources remain and how best to

use them. At this point, we adjust the experimental design to account for remaining

resources and for information gained in the process of designing, building, verifying, and

validating the model.

The design of a computer simulation experiment is essentially a plan for acquiring a

quantity of information by running the simulation model under different sets of input

conditions. Design profoundly affects the effective use of experimental resources for two

reasons:

 The design of the experiment largely determines the form of statistical analysis that

can be applied to the results.

 The success of the experiment in answering the questions of the experimenter (with-

out excessive expenditure of time and resources) is largely a function of choosing the

right design.

We conduct simulation studies primarily to learn the most about the behavior of the system

for the lowest possible cost. We must carefully plan and design not only the model but also

its use. Thus, experimental designs are economical because they reduce the number of

experimental trials required and provide a structure for the investigator’s learning process.

Input data

Stochastic systems contain one or more sources of randomness. The analyst must be

concerned about data related to the inputs for the model such as the contact-volume

forecasts, contact-arrival patterns, and contact-handle times. Although data gathering is

usually interpreted to mean gathering numbers, this interpretation addresses only one

aspect of the problem. The analyst must also decide what data is needed, what data is

available, whether the data is pertinent, whether existing data is valid for the required

purpose, and how to gather the data.

The design of a stochastic simulation model always involves choosing whether to

represent a particular aspect of the system as probabilistic or deterministic. If we opt for

probabilistic and if empirical data exist, then we must make yet another decision. Will we

sample directly from the empirical data, or will we try to fit the data to a theoretical

distribution and, if successful, sample from the theoretical distribution? This choice is

fundamentally important for several reasons.

1 2 ... 18 19 20 21 22 23 24 25 26 27 28 ... 203 204

No comments

Rockwell-automation Arena Contact Center Edition Users Guide User Manual Page 23