Ameneh Deljoo and Marijn Janssen
tions of summing and averaging”. In PPSN agents are dynamic and can move in the network freely, therefore predict and analysis this kind of nonlinear system is difficult.
Aggregation( Property): This property captures the hierarchical organization that is so striking in CAS. Bodies are made of cells. Economies are made of individuals. In CAS, often the formations of more complex metaagents are built of simpler agents( Holland 1992),“ Aggregation... is indeed a basic characteristic of all CAS, and the emergent phenomena that result are the most enigmatic aspect of CAS.”( p. 19). As we shown in figure 2 we divided PPSN in 4 levels( technical level, agent level, department level and organizational level).
Diversity( Property): Populations in CAS are diverse and tend to operate far from equilibrium.( Holland 1992):“ It should be evident then that we will not find CAS settling to a few highly adapted types that exploit all opportunities. Perpetual novelty is the hallmark of CAS”( p. 22)
Flows of Resources( Property): Another characteristic of CAS is that resources continually flow through the system— the familiar“ open‐system” or“ matter‐energy throughput” of CAS( e. g. flow of information). In the technical layer of the figure 2 we have the flow of information and data between passive agents.
Evolving( Property): Gradual changing in a system during the time without changing the fundamental and infrastructure of system is called“ Evolving”. For this concept, systems( organization) are maturated and expand their domain while these developments are occurred during the life of system( organization). System will grow gradually( i. e. develop the scope, more employee, more facilities, more sub‐domain, etc.) without changing in the mission of this system.( Stebbings and Braganza 2009). In PPSN the structure of network is changed because of the new agent enter in the market and the old one is disappear.
In Figure 2 we illustrated a PPSN as agents. The top level in figure shows the interacting organizations. Using hierarchical decompositions we can decompose the organization level. Organizations are purposeful entities( March and Smith 1995) that consist of departments populated by self‐interested individuals. These individuals use all kind of software and technologies to support their activities. Whereas individuals can be modeled as active agents and can change the structure, the supporting software is passive in the sense that they provide services to the active agents. Passive agents, transfer message and service between active agents or can support active agents by providing support. Passive agents have no action and just play intermediaries roles in the organizational network. Examples of elements of PPSN that can be modeled with passive agents are computers and connections. Active agents are used to conceptualizing and modeling the organizations( Janssen 2001). In the technology level, each modification or changes in the software or technology such as: define new rules or policy, adding new systems, change in the database, have an effect on the user’ s level; so, active agents and in the top level each department must to adapt themselves to these changes as we shown it by red flash in Figure 1. By using CAS lens, each level in the organizational level learns how to adapt and deal with challenges.
Changes in the passive agent and effect on the user level and active agents
Figure 2: PPSN conceptualized using agents
598