The paper considers conceptual and methodological aspects in creating computer simulation agent-based models which provide decision-making on the formation of an effective territorial network of forest roads. The key aim of the research is to interpret and test a service-oriented approach to building such models. The approach is based on the principle of interaction between the following objects distributed in space: service providers and consumers. We consider issues related to constructing and operating a network of forest roads used for forest harvesting processes. Complexity in solving these problems is due to a large number of interacting elements and estimated parameters that differ in their impact on the system depending on a particular territory. Thus, such problems are solved using simulation modeling methods, in particular, the agent-based approach, which is one of the most promising ones among them. At the same time, it is determined that they are constructed without using a formalized methodological framework, which does not allow us to interpret the results obtained here as a unified concept of formation of agent-based models in the subject area. To fill this niche, we propose our own concept based on the service-oriented approach. We present main fragments of formal description of the approach and describe some aspects of its applied interpretation using a prototype of game agent-based model in AnyLogic environment. We use general scientific methods: formalization, abstraction, generalization; methods of system analysis and agent-based modeling. As a result of the work we developed and tested our own concept of agent-based modeling of effective territorial transport infrastructure of the harvesting system. The novelty of the concept lies in the use of service-oriented approach. In addition, the concept can be used as a universal basis for building the models in question. This will require the development of methodological aspects of its application, a full-scale interpretation of the possible uses, as well as practical testing on real systems
Keywords
agent-based modeling, harvesting system infrastructure, transport accessibility of forest resources, efficient road network, service-oriented approach