Optimization of Process Design Time in a Distributed Multi-Factory Environment Using Genetic Algorithms to Organize the Process and Support the Development of Technical Designs for Part Production Based on Information Available in the Production Database

Abstract

Today, with the increase in the power of computers in storing and processing data, as well as the advances made in the field of information technologies, especially artificial intelligence, the facilities and capabilities of CAD systems have increased significantly. In the fundamental approach, the process design is created based on the information available in the production database and process mining. In this approach, the process design system operates in the form of knowledge-based systems and artificial intelligence, and in some cases as a DSS system, in order to automate the design, and by receiving information about the details of the desired part, the types of production operations available and their capabilities in terms of accuracy and tolerance, experience related to previous parts, etc., it designs the appropriate process for the part. These systems have been created for the production process of parts that were previously carried out by specialists in manufacturing methods. These systems are very important in terms of integration. The outputs of a process design system include: selecting the appropriate operations and determining the sequence of these operations on the part, selecting the necessary machines to perform the operations, determining the tools and fixtures, as well as the operating instructions for adjusting the machine, the path of the tools, the operation parameters such as speed, duration, load, etc. Of course, it should be noted that since the planning and design of the parts manufacturing process is very dependent on the experience and judgment of the planners, automating all of the aforementioned activities is a very difficult task and most of the existing process design systems are not capable of performing all of the above activities, but in most cases they can only provide decision support services. In a distributed industrial environment, different factories with different machines and tools in different geographical locations are often combined to achieve the highest production efficiency. Process designs may differ due to different resource constraints. Therefore, obtaining an optimal or near-optimal process design seems important. In other words, it is necessary and essential to determine which factory and with which machines and tools each product should be produced. For this purpose, it is necessary to choose a design from different designs that minimizes the cost of producing products while being possible. In this research, a genetic algorithm is introduced that can quickly search for the optimal process design for a single manufacturing system as well as a distributed manufacturing system according to predetermined criteria such as minimizing the process time. This research presents an emerging genetic algorithm for process design with the help of data obtained from process mining that has a successful application in a traditional and distributed multi-factory environment. The application of genetic algorithm for a distributed multi-factory environment in distributed manufacturing systems was developed based on the geographical distribution of manufacturing machines and tools. It can produce an optimal or near-optimal process design compared to other main approaches for single-product manufacturing. The most suitable manufacturing plant can be established when the distributed manufacturing problems are considered. Furthermore, this approach is able to perform multiple optimization objectives based on the lowest production cost or the shortest production time. Based on the selected objectives, near-optimal solutions can be obtained by using genetic algorithm. Through experiments, it has been shown that the developed techniques are better or comparable to other systems in a distributed multi-factory environment.