Simulation Approach to Decision-Making

A simulation model is a mathematical model that represents a simplified form of a complex system whose equations are solved by simulation. The main goal of simulation models is to provide mechanisms for experimentation and behavior prediction, the resolution of questions such as "What would occur if...?" and learning more about the system represented, among others. They allow one to understand how systems behave over time and to compare their performance under different conditions. One of the main advantages of simulation models is that they allow one to experiment with different decisions and to analyze the results obtained in systems in which the cost, the time, or the risk of performing real experiments is high. Besides, simulation models permit the analysis of complex systems that are not possible to be represented with analytical models.

Simulation modeling is very frequently used to support decision making in different business areas, and it provides solutions to a wide range of issues at strategic, operational and tactical level.

There are different simulation approaches, such as, state-based process models, discrete-event simulation, System Dynamics, agent-based simulation, Petri-net models, queueing models, Monte Carlo simulation, probabilistic simulation, and traditional mathematical simulation. The most appropriate approach depends on the nature of the problem to be solved. Specifically, System Dynamics mainly focus on strategic issues and policy analysis. This approach is considered appropriate when taking a distant perspective (meaning strategic) where events and decisions are seen in the form of patterns of behavior and systems structures.

This section provides an overview of the works that use simulation modeling to support decision making in the context of ITIL service strategy module. A search of various digital libraries and citation databases for papers that apply simulation modeling in this field has been performed. Our aim is not to offer an exhaustive study but to analyze if it is appropriate to apply simulation modeling in this scope.

The purpose of ITIL service strategy module is to design, develop, and implement service management as an organizational capability and as a strategic asset. This module processes are as follows: (a) strategy of IT services, (b) financial management for IT services, and (c) demand management. The objective of the strategy of IT services process is to ensure that service strategy is defined and maintained and achieves its purpose. It is responsible for defining strategic goals and the appropriate strategies for compliance. The purpose of the financial management for IT services is to secure the appropriate level of financing to design, develop, and deliver services that meet the strategy of the organization. Finally, the main aim of the demand management process is to understand, anticipate, and influence customer demand for services. This process works with the management capacity process to ensure that the service provider has enough capacity to meet this demand.

Below, we analyze the main application field of the papers referenced and associate them with the most suitable process of ITIL service strategy module.

2.1. Strategy for IT Services Process

Most works found in the scope of this process propose System Dynamics to help define the strategy of IT services. Gary et al. present an overview of System Dynamics contributions in the strategy field focusing on why some firms are more profitable than others. The ideas introduced in this work can also be applied in the context of IT service strategy. The authors focus on business objectives of organizations. Karapetrovic and Willborn propose System Dynamics to describe links in a service organization and represent the interrelationships between business objectives, resources, and processes. On the other hand, Folgueras et al. propose to analyze different business objectives strategies and to determine the most adequate using Systems Dynamics. Finally, the process modeling and simulation approach help evaluate performance metrics, define an IT investment strategy, and select the optimal scenario for business and IT governance alignment.

2.2. Financial Management of IT Services Process

The following works apply simulation to analyze the cost of services. Gebauer presents various System Dynamics models to study service behavior, resource allocation, customer perception, and reaction of competitors. The simulations results allow one to identify situations in which the cost of improving services exceeds the benefits. The framework for developing discrete-event simulation model helps estimate serviceability, costs, revenue, profit, and services quality. Popkov and Karpov introduce a simulation model that helps predict the costs of services, analyze the effects of business strategies, and define both IT infrastructure and services prices. The queuing model proposed by Villela et al. helps evaluate the resource allocation that maximizes the benefits of service providers and minimizes the costs of services failures. The authors of propose a probabilistic model that enables analyzing the expected change-related costs. Analytical models of service costs are presented. The authors of these works evaluate the effectiveness of their approach through simulation. Abrahao et al. introduce a cost model to analyze penalties due to violation and rewards received when the level targets are exceeded. The authors propose a model to analyze the costs of services considering different development teams and service providers. The model simulations help decide the optimal service provider.

Other more recent works study economic aspects of services in cloud computing environments. An analytical model of hybrid cloud costs is presented. In this model the costs of computing and data communication are considered. The authors propose a mathematical model for supporting cloud services selection across multiple sources considering mainly cost and risk. Finally, Goiri et al. discuss an analytical economic model of resource provision in a federated cloud. The effectiveness of the models is evaluated through simulation.

2.3. Demand Management Process

A service-demand-forecasting method that uses multiple data sources for improved accuracy is proposed. The authors present an advanced scenario simulation framework to analyze each customer service choice behavior and total service demand under an assumed condition. On the other hand, the authors focus on service multirequests from different users and propose a cooperative downloading strategy with multiclass request. The strategy provides different services according to users' expectations and helps service providers obtain the most benefit by making use of the repetition ratio of data and user-defined class. The simulation results indicate the benefits of the proposed scheme in terms of increasing service quality and benefit for service providers. Finally, Sen et al. propose an analytical model for SLA formulation that is responsive to demand fluctuations and user preference variance, with the objective of maximizing organizational welfare of the participants. This formulation features a dynamic priority based price-penalty scheme targeted to individual users. Simulations performed using data from an existing SLA to provide evidence that the proposed dynamic pricing scheme is likely to be more effective than a fixed price approach are presented.

Our first research question is aimed at whether simulation modeling is used to make decisions in IT service strategy. The analysis of the papers above shows that these techniques are widely used in this context.

The second research question is aimed at whether it is adequate to use System Dynamics to solve IT service strategy problems. The authors emphasize that System Dynamics focus mainly on strategic issues and policy analysis. Besides, although the referenced papers show that different simulation approaches have been used in this field, most of papers found in the context of strategy for IT service process propose System Dynamics to define service strategy, strategic goals, and analysis policies. Instead, in the context of the financial management of IT services and demand management processes, other simulation approaches such as analytical models or discrete-event simulation are more used.

In the following sections, we present a System Dynamics model built to help service providers in the decision-making process to define their strategic goals.