Modeling, analysis, synthesis, and performance evaluation of multioperational production systems with hybrid timed Petri nets

• Published in: IEEE transactions on automation science and engineering Vol. 3; no. 1; pp. 29 - 46
• Main Authors: Tsinarakis, G.J., Tsourveloudis, N.C., Valavanis, K.P.
• Format: Journal Article
• Language: English
• Published: Piscataway, NJ IEEE 01-01-2006; Institute of Electrical and Electronics Engineers; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Applied sciences; Associate members; Automatic control; Buffers; Complexity analysis; Computer science; control theory; systems; Computer simulation; Control theory. Systems; Derivation; discrete and continuous components; Exact sciences and technology; Hybrid control systems; hybrid timed Petri nets (HTPNs); Invariants; Mathematical model; Mathematical models; Modules; multioperational production systems; Object oriented modeling; P-invariants; Performance analysis; Petri nets; Power system modeling; Production systems; Student members; Synthesis; Systems analysis; Topology; Performance evaluation; Invariant; Complex system; multioperational production systems; In-process inventory; hybrid timed Petri nets (HTPNs); Complexity; Hybrid system; Work capacity; P-invariants; System synthesis; Time interval; Invarying system; System analysis; Petri net; Discrete event system; Production system; discrete and continuous components; Systems theory; Buffer system; Topology; modules; Complexity analysis; Analysis method; Capacity constraint
• ISSN: 1545-5955; 1558-3783
• DOI: 10.1109/TASE.2005.857655

Hybrid timed Petri nets (HTPNs) are derived to study random topology and complexity multioperational production systems where parts of one type follow the same route to produce a final product. Each production system is first decomposed into a fundamental multiproductive machine, multiassembly and multidisassembly modules, followed by derivation of their corresponding HTPN models. The overall system HTPN model is obtained via individual module synthesis, satisfying system constraints. Individual module and overall HTPN models nodes (places and transitions) are calculated. Individual module and overall system model invariants are derived mathematically. Performance and possible tradeoffs due to varying operational constraints (buffer capacity, work in process, machine utilization, backlog, etc.) are investigated through extensive simulations. Results show the applicability of the proposed methodology and justify its modeling power and generality.