Configuration space–based discrete event system specification formalism for a smart factory with real-time flexibility

Abstract

A manufacturing system consists of various manufacturing devices, and each device has a set of tasks which are triggered by specific commands. Traditionally, simulation has been considered as an essential technology for the evaluation and analysis of manufacturing systems. Although discrete event system specification formalism has been a popular modeling tool for manufacturing systems, it has limitations in describing situations such as sudden cancelation of tasks. Proposed in this article is an extended discrete event system specification formalism for the effective description of a smart factory which requires the intelligence to handle turbulences in real-time production. The extended discrete event system specification formalism incorporates the configuration space concept, which is well-known in classical mechanics. While the conventional discrete event system specification formalism uses only the logical states set to represent the device states, the proposed formalism employs the combination of two sets: a logical states set (sequential states set) and a physical states set (configuration space of the device). As a result, the extended formalism enables the effective description of nondeterministic tasks which may occur frequently in a smart factory.

Modeling and Simulation of Manufacturing Processes and Systems: Overview of Tools, Challenges, and Future Opportunities

“…Different techniques for process improvement are exploited. Linear programming, Discrete-Event Simulation (DES), System Dynamic (SD), Markov Chain Analysis, and Monte Carlo Simulation are considered the most popular conventional simulation tools [71][72][73][74]. Various modeling formalisms have been established to describe the manufacturing processes and systems for varied fields and engineering applications (e.g.…”

A Robust Discrete Event Method for the Design of Cyber-Physical Systems

“…In this control system, different parts of a real system were modeled using DEVS; a model controller was used to validate the design and was incorporated into the actual control system. In [117] RT DEVS was proposed. RT-DEVS is a major milestone because other approaches that extended the DEVS formalism have some similarities with RT-DEVS.…”

“…Thirdly, in DEVS, state transition events are defined to occur at instances of time however in real-time simulation state changes and operations may occur during a time interval (or a time window). Fourthly, it is difficult to validate simulated models in the real world[116][117].…”