Introduction
Today, the services industry provides the majority of all jobs in Western countries, and services tend to be delivered more and more using the Internet. The service economy is becoming increasingly dominant in developed economies, with knowledge assets playing a great role relative to physi -cal and financial assets. (Rouse & Baba, 2006)
Services are often characterized as intangible, perishable, experience-based, difficult-to-standardize products needing many interactions between customers and services providers. Gronroos (2001) identified three basic characteristics of services:
1. Services are processed consisting of activities or a series of activities rather than things.
2. Services are at least to some extent produced and consumed simultaneously.
3. The customer participates in the service delivery process.
All kinds of information and communication technology (ICT) are applied to support the creation of service networks. Service networks are constellations of independent organizations that work together in various configurations in order to deliver services. The provisioning of services can be viewed as a series of activities leading to some observable behavior between service providers (or service brokers) and service requesters. They are delivered using the Internet, accessible from any place at any time and often involve no direct human involvement of the service provider. The term e-services is typically used to describe a variety of electronic interactions ranging from basic services, such as the delivery of news and taking out an insurance policy, to more complex services, such as the delivery of context-aware, personalized services.
To understand a service network, both the network and the decision makers involved need to be understood. Stakeholder theory states that those who can effect change or be affected by it should be accounted for in the transformation process (Pfeffer, 1981). The diversity of key stakeholders and their interests makes evaluating the design and the efficiency and effectiveness of service networks very complex.
Often stakeholders are characterized by opposing interests, having heterogeneous systems and being part of multiple service networks. The effective management of such services network is key to success, which requires understanding each other’s interests, business processes, and information systems. Often organization network managers, a particular type of electronic intermediary (e-intermediary), specialize in coordinating such networks (Janssen & Verbraeck, 2005b). Design decisions are critical, as they determine the efficiency and effectiveness of the service networks. The development and growth of service networks requires the developer to carefully identify, evaluate, and understand the possible impact of the various design alternatives. A business engineering methodology can be of help in designing and developing service networks by providing insight into current network structure and potential structure, and by evaluating the implications of potential arrangements. Simulation can be used to compare the performance of the current and possible situations in a business engineering methodology. Simulation of service networks is much more difficult than physical networks, as the products often concerns intangibles. The objective of this article is to discuss research issues concerning the simulation of service networks to support business engineering.
background
Service networks can use a large variety of coordination mechanisms and structures to coordinate the activities of participants. Service networks consist of organizations cooperating together, and the management of the networks has a large impact on the total performance. Figure 1 shows a service network schematically.
The dynamic nature of service networks—that is, the changing number and/or types of partners, and the involvement in several networks—increases the difficulty and complexity to understand the dependencies in the network. The creation of flexible, temporary service networks results in the creation of business processes that are no longer self-contained within a single organization. The effectiveness of service networks depends more and more on the performance of external partners that are often unknown and viewed as black boxes (Tewoldeberhan & Janssen, 2007). Therefore, some organizations can be considered as white boxes and others as black boxes in the networks, as schematically depicted in Figure 1. Not all organizations must be involved in each service provisioning process. Organizations might be selected and dynamically assembled based on the services needed by the customers. A service network consists of multiple businesses having varying types of relationships.
Figure 1. Example of a service network
The core of a service network is the coordination of the various interdependent activities performed by autonomous organizations. There are two opposing views on coordination. In a coordination of tasks approach, the design of processes is dependent on the coordination mechanisms that manage the dependencies between tasks (Malone & Crowston, 1994). The coordination of commitments approach emphasizes networks of commitments that organizations establish through intentional acts of speech (Winograd & Flores, 1987). This coordination approach emphasizes the fulfillment of human commitments and describes activities in terms of contracts and promises. A traditional approach to supply chains is the coordination of tasks view. In a service network, both views apply, as the activities performed by the independent organization needs to be coordinated in order to agree on and fulfill commitments.
The requirements of an organization are not easily elicited and can demand innovative mechanisms or deliberate tradeoffs. The timely sharing of information among organizations is often a major issue (Christopher, 2003). Information sharing is necessary for efficient coordination of the service network and to optimize performance. The organizations making up the network often want to avoid that information is provided to other network members. Information might be used to negotiate lower prices or undermine competitive advantage, as competitors might learn from it. Another typical issue in the business engineering of service networks is the selection of coordination mechanisms, as members can have different and even opposing requirements. For example one organization strategy might be to minimize trading time, while another might want to minimize costs. The most conspicuous opposing requirement is that buyers want to have the lowest price at the best possible trading conditions while sellers want to have the highest possible price to maximize revenue.
Table 1. A list of business engineering issues
• Which information architectures and structures are most beneficial in which situations?
• Aligning mechanism with service and markets characteristics?
• Integration of the information systems of network members?
• How to manage the service network?
• Should intermediaries be used to coordinate the service network?
• Level of coarse and fine-grained services
• Evaluation of implications of changes
• Pooling and sharing of services
• Conflicting interests of network members
• Incomplete information, ensuring information privacy
• Ensuring quality of product to buyer and payment to sellers
• Tracking and tracing
• Reducing transaction risk and increasing trust
• Use of software agents as assistance in search and evaluation
• Intelligent product and vendor matching mechanisms
• Product distribution/delivery
• Creating and disseminating product information
• Information processing and aggregation
• Open and closed networks
• Spot sourcing for dynamic networks vs. systematic sourcing for sustainable networks
• Type of management information and dissemination of management information
In short, a large number of other trade-offs and decisions need to be made before an efficient and effective service network can be established. A list of business engineering issues is shown in Table 1. Some limitations are coming from the state of the art of the technology and from market and/or service characteristics; others are coming from the opposing requirements and needs of the parties involved. Simulation for business engineering of service networks can help decision makers gain insight into these issues. This should support them in making deliberate choices without having to experiment in real life, which could be costly and even result in a loss of customers.
simulation-supporting business
ENGINEERING
Service networks are by nature complex, and analytic methods can only be applied in a limited way. Although these approaches contribute to insight into and design of service networks, they do not help decision makers evaluate the impact and support their decision making in practice. Ideally, the implications of changes should be evaluated prior to implementation on criteria such as costs, utilization, trading time, delivery time, number of bids, matching chance, and so on. Analytical approaches also do not grasp the time-dependent dynamics resulting from the interplay between actors executing business processes.
A business engineering methodology can be seen as a continuum of approaches to process change (Kettinger, Teng, & Guha, 1997). Such a methodology is often seen as a way to tackle issues from an engineering perspective, as well as from a social perspective. The analytical framework of Sol (1990) provides a suitable way to describe business engineering methodologies. This framework classifies design methodologies by ways of thinking, working, modeling, and controlling. The way of thinking describes the philosophy on which the design methodology is based. It provides the basic assumptions of the business engineering approach and should contain, for example, whether it is aimed at radically changing the service network or at step-wise improvement, and will outside experts be used to analyze the situation or will the business engineering approach be based on stakeholders’ participation? The way of working provides the subsequent steps that should be carried out to arrive at a new situation. An approach supporting relatively small steps and letting stakeholders participate so that they can gain the necessary knowledge to formulate their own incremental improvements or a radical approach can be taken. The way of modeling refers to the concepts that are used to abstract reality into models of the problem domain. Often models are used to support communication and evaluate the impact of changes. The way of controlling or management approach provides the components needed for the management of a design project. Often a project management approach is taken using milestones to guide the process.
Figure 2. Analytical framework for business engineering methodologies
The way of modeling provides the necessarily support for a business engineering approach. Simulation of business processes constitutes one of the most widely used applications of operational research, as it allows us to understand the essence of business systems, to identify opportunities for change, and to evaluate the effect of proposed changes on key performance indicators (Law & Kelton, 1991). The philosophy behind a business engineering approach is to develop a simulation model of the service network, experiment with this model, and experiment with alternative situations (Sol, 1982). An analysis of service networks needs to begin with an understanding of current processes and should investigate how conventional transaction methods are changed as a result of ICT adoption. One of the advantages of simulation is that what-if analysis can be carried out without changing reality at lower costs. These analyses often compare situations using time- and cost-based performance indicators such as delivery time, utilization of resources, cost of activities, and waiting time.
Animation is often a standard feature of simulation. An animation model is a graphical representation of a problem situation and includes visualization of the time-ordered dynamics of objects, a static background, an overview of performance indicators, and a user-interface (de Vreede & Verbraeck, 1996). The purposes of animation are to facilitate decision makers to acquire insight into the dynamic interactions of the modeled system, the performance of the ‘as is’ and ‘to be’ situation, and to facilitate communication between parties involved in a dynamic modeling study. Effecting enterprise-wide technology and business process change in the service networks is a massive and complex undertaking, and animation might help to create a shared vision and understanding among participants.
future trends
Business engineering methodology provides context to the simulation technique (Greasley, 2003). Business engineering approaches proposed by Streng (1994) and Giaglis, Paul, and Doukidis (1999) tackle the analysis of the added value of technology-enabled changes by making use of discrete-event simulation. Nikolaidou and Anagnostopoulos (2003) use a simulation approach for modeling distributed systems.
Dependent on the characteristics of the service network under study, certain issues shown in Table 1 dominate and should be simulated. A business engineering methodology should ensure the incorporation of the relevant business engineering issues, and simulation can be used to aid conscious decision making. A business engineering approach should help to focus on those issues most relevant to the particular situation and help to find solutions to those issues. A fruitful research direction seems to be which issues should be included in the business engineering for which types of situations.
The ability to create flexible alliances with partners to form supply chains or business networks becomes more and more important for businesses. Almost all of the current attention seems to be geared toward how static service networks can be configured, almost completely neglecting the dynamism of the environment in which a service network operates. Little attention is given to how these service networks evolve and emerge over time, or how environmental changes that require the service network to adapt can be dealt with. Various types of changes and their possible impact on the complete network and also on the individual organizations should be captured by further research.
In the early stages of service networks, management resembled like traditional business relationships, and the network participants where often modeled using black-box approaches. Most studies assume that all data was available, however in many service networks this might not be a valid premise. Many organizations will be reluctant to share data about their mission-critical business processes with others. Thus, the efficiency of service networks will be hampered by the fact that organizations often demand that their data and inner business processes remain hidden from the other organizations in the supply chain. Future research should also include the modeling of service networks for situations where the supply chain is not completely known and limited data is available.
Web services technologies are more and more used to support the creation of service networks and especially the use of Web service orchestration technology to coordinate the interactions between network participants. Fast adoption is often hampered by the need for experimentation to make efficient use of this technology (Tewoldeberhan & Janssen, 2007). As such, more insight in business engineering, simulation, and the effect of Web service technology is a research direction. Especially to evaluate the implications of new technologies, simulation of a service network should on emulation, which means that actual software is used and embedded in the simulation experiment. This combines a simulation and prototyping approach into one modeling approach.
A large number of independent organizations with their own strategies and sometimes even opposing aims carry out business with each other. The relations between organizations can change during the trading process, as organizations can enter or leave the playing field. The most powerful abstractions are the ones that minimize the semantic gap between the units of analysis that are intuitively used to conceptualize the problem and the constructs present in the modeling approach. Ideally, an organization should be simulated as distributed systems, where each system is represented by an autonomous entity. Software agents are autonomous entities that can be used for simulating organizations. The so-called agents are autonomous entities with their own interests and goals so they can decide to enter or leave a trading situation. Agent-based simulation has appeared for modeling organizations within electronic markets (Ramat & Preux, 2003; Janssen & Verbraeck, 2005a). Janssen and Verbraeck (2005a) developed an agent-based simulation technique for electronic markets. Agent-based simulation approaches view systems in terms of autonomous agents that engage in interactions to coordinate their activities. The coordination problem in a system consisting of a number of agents is analogous to the coordination problem of independent organizations trading with each other. This approach might also be suitable for a service network and seems to be a feasible research direction.
Communication over the Internet using a Web-based simulation is preferable for supporting communication between the researcher and persons involved in the design process. Another research issue is to use distributed simulation so multiple participants could interact with a simulation environment at the same time. In this way the participants can gain experiences with a hypothetical service network. This might lead to an increase in insight into the problem situation by participants and might help designers to make better decisions.
conclusion
Organizations cooperate and compete more with each other in service networks, and the efficient and effective management of such networks determines success. Business engineering is aimed at creating change by simultaneously considering organizational and technical aspects, and simulation can make change possible by supporting the business engineering approach. During the business engineering of service networks, numerous trade-offs and decisions must be made, influencing the performance and possible adoption. Business engineering using simulation should help decision makers focus on the most relevant issues and make appropriate decisions and trade-offs without having to experiment in real-life situations. It is still unclear what the most relevant issues are to focus on when engineering service networks.
Ideally, service networks should be modeled representing reality as close as possible. All types of electronic intermediaries can support the coordination of the dependencies among organizations in the service network. A fruitful direction seems to be the use of a distributed, agent-based simulation, making use of emulated mechanisms. In this way emergent behavior and changes can be modeled and their impact evaluated. Ideally, decision makers of various organizations should be able to view the animation of the simulation over the Internet, manipulate parameters, and view the consequences of their actions. The accomplishment of this ideal needs ample research attention in the domain of business engineering methodologies, distributed simulation, agent-based simulation, Web-based animation, and emulation of mechanism.
KEY TERMS
Agent-Based Simulation: Simulates organizations as interacting autonomous entities with their own interests and goals.
Animation Model: A graphical representation of a problem situation which can consist of a visualization of the time-ordered dynamics of objects, a static background, an overview of performance indicators, and a user interface.
Business Engineering: The integral design of both organizational structures and information systems.
Coordination of Commitments: The actions by humans leading to the completion of work. Coordination is described in terms of contracts and promises consisting of recurring loops of requesting, making, and fulfilling commitments.
Coordination of Tasks: The management of dependencies between tasks.
Discrete-Event Simulation: Models a system by changing the systems state at discrete points in time.
Emulation: Actual software is written to execute something, instead of simulating it.
Service Network: Constellations of independent organizations that work together in various configurations in order to deliver services.
