A fuzzy based algorithm to solve the machine-loading problems of a FMS and its neuro fuzzy petri net model

Abstract

This paper aims to develop a fuzzy-based solution approach to address a machine-loading problem of a flexible manufacturing system (FMS). The proposed solution methodology effectively deals with all the three main constituents of a machine loading problem, viz. job sequence determination, operation machine allocation and the reallocation of jobs. The main objectives of the FMS loading problem considered here are minimisation of system imbalance and maximisation of throughput; the constraints to be satisfied are the available machining time and tool slots. An analytical argument has been provided to support the membership function related to the operation machine allocation vector. Computational results revealed the superiority of the proposed algorithm over other heuristics when it is tested on a standard data set adopted from literature. A new class of petri net model called the ''Extended neuro fuzzy petri net'' is constructed to capture clearly the various details of the machine loading problem which can be further extended to learn from experience and perform inferences so that truly intelligent system characteristics can be realised.

A hybrid genetic algorithm for a loading problem in flexible manufacturing systems

“…In assigning operations to machines, Nagarjuna et al (2006) give preference to the machines with more optional time requirements, more remaining processing times, and less essential time requirements. Further works on this problem have considered the twin objectives of maximising throughput and minimising system unbalance (Vidyarthi and Tiwari 2001, Kumar et al 2004, Srinivas et al 2004, Swarnkar and Tiwari 2004, Kumar et al 2006, Nagarjuna et al 2006, Prakash et al 2008, Yogeswaran et al 2009.…”

A compact timed state space approach for the analysis of manufacturing systems: key algorithmic improvements

“…Due its nature scientific community has paid for several years a lot of attention on solving these types of problems, which are considered the toughest to be solved (Lee and DiCesare, 1994;Dauzére-Peres and Laserre, 1994;Chan et al, 2010;Renna 2010). The PN modelling formalism possesses the necessary characteristics to model the complexity of these problems; therefore it has been widely used as a modelling formalism to cope with the allocation problem of these types of systems (Van der Aalst, 1995;Adballah et al, 1998;Kumar et al, 2004). In the 3x3 job-shop three jobs must go through processes in three different machines.…”

Modeling production configuration using nested colored object-oriented Petri-nets with changeable structures

“…In the EEHLCDPN, control places together with other relevant net elements are introduced to deal with the selection of disassembly operations. Kumar et al (2004) develop a formalism of extended neuro fuzzy Petri nets (ENFPNs) to solve the machine-loading problems in flexible manufacturing systems. Such major concerns in machine-loading problems as determination of optimal job sequences, job reallocation and operation-machine allocation are judged based on multiple alternatives, for which fuzzy AND/OR rules are introduced on the basis of neuron networks in ENFPNs.…”