Roadworks are highly mechanized, and their efficiency is conditioned by selection of machines and synchronizing their work. The work of machine sets can be modeled using queuing theory, which enables thorough analysis of the system operation and finding optimal composition of the machine fleet. For simple queuing models, there exist analytic formulas to calculate basic properties of the system (such as a machine’s downtime probability, average queue length, average waiting time, servers utilization levels). However, more complex, real-life systems (queuing networks) can be analyzed more efficiently by means of simulations. The paper presents a model that describes road resurfacing works. Using simulation technique, the authors conducted experiments to evaluate the economic effect and productivity resulting from various configurations of the system and, for a number of optimization criteria (lowest cost, highest output, etc.), to find most suitable machine sets.