8.9 Some Final Thoughts
This chapter presented how to apply simulation to a realistic problem. The solution involved many of the simulation modeling techniques that you learned throughout the text. In addition, and its supporting tools were applied to develop a model for the problem and to analyze the system under study. Finally, experimental design concepts were applied in order to recommend design alternatives for the problem. This comprehensive example should give you a good understanding for how to apply simulation in the real world.
There are a number of other issues in simulation that have not been given full description within this text. However, based on the material in this text, you should be well prepared to apply simulation to problems in engineering and business. To continue your education in simulation, you should consider reading (Chung 2004) and (Banks 1998). Chapter 13 of (Chung 2004) provides a number of examples of the application simulation, and Chapters 14-21 of (Banks 1998) illustrate simulation in areas such as health care, the military, logistics, and manufacturing. In addition, Chapters 22 and 23 of (Banks 1998) provide excellent practical advice on how to perform and manage a successful simulation project.
Finally, you should get involved in professional societies and organizations that support simulation modeling and analysis, such as:
American Statistical Association (ASA)
Association for Computing Machinery: Special Interest Group on Simulation (ACM/SIGSIM)
Institute of Electrical and Electronics Engineers: Systems, Man, and Cybernetics Society (IEEE/SMC)
Institute for Operations Research and the Management Sciences: Simulation Society (INFORMS-SIM)
Institute of Industrial Engineers (IIE)
National Institute of Standards and Technology (NIST)
The Society for Modeling and Simulation International (SCS)
An excellent opportunity to get involved is through the Winter Simulation Conference. You can find information concerning these organizations and the Winter Simulation Conference easily through the internet.
As a final thought you should try to adhere to the following rules during your practice of simulation:
To only perform simulation studies that have clearly defined objectives and performance metrics.
To never perform a simulation study when the problem can be solved through simpler and more direct techniques.
To always carefully plan, collect, and analyze the data necessary to develop good input distributions for simulation models.
To never accept simulation results or decisions based on a sample size of one.
To always use statistically valid methods to determine the sample size for simulation experiments.
To always check that statistical assumptions (e.g. independence, normality, etc.) are reasonably valid before reporting simulation results.
To always provide simulation results by reporting the sample size, the sample average, and a measure of sample variation (e.g. sample variance, sample standard deviation, standard error, sample half-width for a specified confidence level, etc.)
To always perform simulation experiments with a random number generator that has been tested and proven to be reliable.
To always take steps to mitigate the effects of initialization bias when performing simulations involving steady-state performance.
To always verify simulation models through such techniques as debugging, event tracing, animation, etc.
To always validate simulation models through common sense analytical thinking techniques, discussions with domain experts, real-data, and sensitivity analysis techniques.
To always carefully plan, perform, and analyze simulation experiments using appropriate experimental design techniques.
By following these rules, you should have a successful and professional simulation career.