Prescriptive Analytics: The Final Frontier for Evidence-Based Management and Optimal Decision Making by Dursun Delen

Author:Dursun Delen [Dursun Delen]
Language: eng
Format: epub
Publisher: Pearson FT Press
Published: 2019-10-20T16:00:00+00:00

Conclusion

This chapter provided an overview of simulation modeling—one of the most popular enablers of prescriptive analytics, only second to optimization. Simply stated, simulation is the art and science of imitating and replicating the real-world systems and processes in computers for the purpose of conducting experiments and what-if scenarios. Whereas Monte Carlo simulation is a simple yet useful technique to address stochastic/probabilistic business and scientific problems, discrete event simulation is a technique to model and study highly complex stochastic business processes. Because simulation allows for rich representation of the reality, including the imprecise/stochastic/probabilistic nature of the actual systems, it is suitable for complex systems that do not lend themselves to optimization-type prescriptive analytics methods. Compared to optimization, simulation is more descriptive than prescriptive; it is not a tool capable of providing the optimal solution. However, simulation is an excellent technique to describe the nature of the real-world systems, capable of producing much-needed information at a granular level to support timely and accurate decisions. Simulation is often used when an optimization (mathematical programming type) solution is not feasible.

Due to its versatility, demand for simulation software products has been increasing, resulting in a rich collection of tools and service/consultancy-providing companies. In the analytics market, one can find narrowly defined simulation tools (specific to an industry or a problem type) as well as generalized broad-spectrum software tools that claim to have the capability to address the situation. Common software tools include Simio, Arena, ProModel, AnyLogic, GoldSim, and SAS Simulation Studio, among others. Modern-day simulation modeling tools employ graphical and intuitive user interfaces that make it easy to model complex systems; however, as is the case in optimization, the secret sauce to great simulation modeling is in the way we characterize and represent the real system into a proper abstraction. Moving from a real system/subsystem/problem to an accurate and rich representation/abstraction as a computer simulation model is still more of an art than science. It’s one that requires diligent studying and in-depth understanding of the underlying real-world system, acquisition/collection of all related data/information, and meticulous representation of the underlying components and their logical relationships.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.