Minimum Initial Marking Estimation in Labeled Petri Nets for Required Resource Minimization of Automated Manufacturing Systems

With an increase in the scale and complexity of practical industrial applications, the performance of automated manufacturing systems is usually more likely to be affected by resource limitation. In order to make good use of resources for automated manufacturing systems by using the formalism of Petri nets, one of the mainly studied problems is how to estimate the minimum initial markings (MIMs) in a known labeled Petri net model based on the observation of its label sequence. Unfortunately, this problem's computational complexity is NP-hard in theory. Previous research either suffers from excessively high computation costs or attempts to alleviate the computational burden at the expense of obtaining only a subset of MIMs. In this paper, an approach using an improved genetic algorithm with the token number prediction is proposed to compute the MIM estimation. Illustrative examples and comparative studies are provided to show the effectiveness and advantages of our proposed approach. © 2025 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.