The aim of the work was to develop a prioritizing and scheduling method to be followed in small and medium-sized companies operating under conditions of non-rhythmic and nonrepeatable production. A system in which make to stock, make to order and engineer to order (MTS, MTO and ETO) tasks are carried out concurrently, referred to as a non-homogenous system, has been considered. Particular types of tasks have different priority indicators. Processes involved in the implementation of these tasks are dependent processes, which compete for access to resources. The work is based on the assumption that the developed procedure should be a universal tool that can be easily used by planners. It should also eliminate the intuitive manner of prioritizing tasks while providing a fast and easy to calculate way of obtaining an answer, i.e. a ready plan or schedule. As orders enter the system on an ongoing basis, the created plan and schedule should enable fast analysis of the result and make it possible to implement subsequent orders appearing in the system. The investigations were based on data from the non-homogenous production system functioning at the Experimental Plant of the Łukasiewicz Research Network – Institute of Ceramics and Building Materials, Refractory Materials Division – ICIMB. The developed procedure includes the following steps: 1 – Initial estimation of resource availability, 2 – MTS tasks planning, 3 – Production system capacity analysis, 4 – ETO tasks planning, 5 – MTO orders planning, 6 – Evaluation of the obtained schedule. The scheduling procedure is supported by KbRS (Knowledge-based Rescheduling System), which has been modified in functional terms for the needs of this work assumption.

Modeling and simulation are key performance analysis and control techniques to optimize decision-making as well as design and operate complex production systems. They are also indicated as one of the technological pillars of modern industry and IT solutions supporting the implementation of the roadmap toward Industry 4.0 in the areas of digital transformation and automation. In the context of the required rapid transformation of today’s enterprises, it becomes extremely important to look for solutions that allow the use of the existing infrastructure, information, and energy, so as to minimize the negative impact of new technologies and the transformation process itself on the environment. The article presents an approach to modeling large and complex production systems with the use of distributed Petri net models allowing the use of the possessed IT infrastructure as consistent with the idea of sustainable development in the activities of enterprises. This eliminates two major problems that render traditional models unusable. The first is related to the difficulties in analyzing and verifying models of enormous size and infinite space of states. The second is related to the required computing power, if such analyzes are to be performed on one computing unit, which would force the producers to replace the IT infrastructure. For this purpose, modular Petri nets are introduced. Other benefits of modularization, such as smaller components that can be independently analyzed, are also presented in the paper. The proposed modular Petri net has been implemented in the proprietary GPenSIM software. The paper is complemented by a practical example of industrial modeling of production systems with automated guided vehicles (AGVs) using the Modular Model with Intelligent Petri Modules.