Evolution-generated Communications in Digital Business Ecosystem Juan Wang College of Information Engineering Shenzhen University, China Email: juanwang@szu.edu.cn Philippe De Wilde School of Mathematical and Computer Sciences Heriot-Watt University, UK Email: pdw@macs.hw.ac.uk Abstract—This paper discusses a new approach to model a complex evolving system, a Digital Business Ecosystem (DBE). We present a geography-based two coupled networks model. The upper layer is the business network layer regarded as an information ecosystem with multiple evolving agents. The lower layer is the underlying P2P communication layer to support communications. The two networks in the system are modelled as evolving over time and interacting with each other. In the business ecosystem, agents exchange services with each other. Through evolution processes, fit services reproduce and weak services die out. We study how these evolution-generated communications affect the topologies of both the business network and the underlying communication network and also system performance. We provide the details of the discrete event simulation and discusses the experimental results. The approach in this paper can be applied to model and analyze other complex real-world applications as our previous work [1]. Index Terms—Evolutionary communications; Agent-based modeling; Business ecosystem I. I NTRODUCTION Business ecosystem is a relatively new concept introduced by Moore[2], who describes the increasingly interrelated nature of enterprises as competing and evolving organisms in a business environment. The significance of regarding business networks as ecosystems is that it highlights the importance of understanding the implication of a company’s decision on its surrounding environment. The Digital Business Ecosys- tem (DBE) emphasizes the ’digital’ aspect of the business cooperation[3]. It intends to provide an open source dis- tributed environment, where software components, services, applications and also business models are regarded as ’digital species’ that can interact with each other, reproduce and evolve according to laws of market selection. In this paper, our primary concern is with the support for open communities of Small and medium enterprises (SMEs) so to aid them avoiding the ever increasing gap of technology between SMEs and large size enterprises. The DBE provides a platform for SMEs to publish their services as service provider clients of the system and also obtain their required services as service consumer clients of the system. In the ecosystem, SMEs manage and communicate with other SMEs. Services and products provided by SMEs are stored in their habitats. Through selection process, genetic algorithms and exchanges of services between the habitats, individuals of the SME network can benefit by combining their services with those of others, building new products and services so that they can expand their own business easily. For large-scale complex evolving systems, agent-based modeling and simulation provides a way to model and better understand such systems[4]. The construction of a model is not an easy task. It needs to reveal the most important properties of the system and reflect real-world scenarios but still keeping the simulation possible. In this paper, we present a simplified model of the SME business network as an ecosystem, where numerous SMEs communicate with others to produce service chains and some services of high fittness values are exchanged between SMEs. Communications between different SMEs are supported and executed by a distributed lookup servers system called FADA, which will be described in detail in the following sections. A lot of research work have presented a number of ap- proaches of information and communication infrastructures to support business processes between different business enter- prises. Some work provide basic information infrastructures [5] and some approaches are proposed to improve these basic infrastructures [6], and some work design middleware platform that supports the runtime creation and management of services [7]. Nevertheless, these infrastructures emphasize primarily on how to support and improve the integration of cooperative business processes among multiple enterprises, but do not take the effects of infrastructure itself on the business process into account. Despite business process and information infrastruc- ture interact in practice, efforts on business process engineering (BPR) and information infrastructure are rarely conducted together. Some practitioners have started to pay attention to the correlation between business processes and information infrastructures[8]. However, the study is inadequate and further analysis on this relationship in more detail is needed. In this paper, we construct a model of a real-world DBE system, consisting of a business network with evolution dy- namics generated by the evolutionary environment and a P2P communication network to support the business network. Business processes induce communications between SMEs, which might affect the topology of the business network and be affected by the connectivity of the SMEs. On the other hand, communications between SMEs may in influence the topology of the underlying P2P communication infrastructure. This is a challenging problem. 978–1–4244–1674–5/08/$25.00 c  2008 IEEE CIS 2008 Authorized licensed use limited to: Heriot-Watt University. Downloaded on April 29, 2009 at 05:37 from IEEE Xplore. Restrictions apply. 618