Trends and perspectives in dynamic environments point to an excuse for optimal operating levels in reconfigurable manufacturing activities. Central to the aim of meeting this require is the problem of appropriate approaches for manufacturing process planning optimization in reconfigurable manufacturing, i.e. (i) what making decisions models and (ii) what computational techniques, offer an optimal manufacturing process planning solution inside a multidimensional decision variables space? Conventional optimization techniques aren’t robust, hence they aren’t appropriate to handle multidimensional search spaces. However, process planning optimization for reconfigurable manufacturing isn’t amenable to classical modeling approaches because of the existence of complex system dynamics. Therefore, this research explores how you can model reconfigurable manufacturing activities within an optimization perspective and the way to develop and choose appropriate non-conventional optimization approaches for reconfigurable process planning.
Within this study, a brand new method of modeling Manufacturing Process Planning Optimization (MPPO) was created by extending the idea of manufacturing optimization via a decoupled optimization method. The distinctiveness of the approach is based on embedding a built-in scheduling function right into a partly integrated process planning function to be able to exploit the proper potentials of versatility and reconfigurability in manufacturing systems. Alternative MPPO models were built and variances connected using their utilization examined. Five (5) Alternative Formula Design Techniques (AADTs) were developed and investigated for appropriateness in supplying process planning solutions appropriate for reconfigurable manufacturing.
5 (5) AADTs incorporate a variant from the simulated annealing formula that implements heuristic understanding at critical decision points, two (2) cooperative search schemes with different “loose hybridization” from the Boltzmann Machine formula with (i) simulated annealing, and (ii) genetic formula search techniques, and 2 (2) modified genetic algorithms.
The comparative performances from the developed AADTs when tasked to resolve a clear case of a MPPO problem were examined and evaluated. Particularly, the relative performances from the novel variant of simulated annealing compared to: (a) (i) a simulated annealing search, and (ii) an inherited search within the Boltzmann Machine Architecture, and (b) (i) an altered genetic formula and (ii) an inherited formula having a customized threshold operator that implements a cutting-edge extension from the diversity control mechanism to gene and genome levels were went after within this thesis.
Results reveal that all five (5) AADTs can handle stable and asymptotic convergence to close optimal solutions instantly. Analysis signifies the performances from the implemented variant of simulated annealing are similar to individuals of other optimization techniques coded in this thesis. However, a computational study implies that as compared to the simulated annealing technique, significant enhancements in optimization control performance and excellence of computed solutions could be recognized through applying intelligent techniques.
As evidenced through the relative performances from the implemented cooperative schemes, an inherited search is preferable to a simulated annealing search within the Boltzmann Machine Architecture. Additionally, little performance gain could be recognized through parallelism within the Boltzmann Machine Architecture. However, the highest performance from the genetic formula that implements a long diversity control mechanism shows that more competent genetic algorithms could be designed through customized operators.
Therefore, this research has says extending manufacturing optimization concepts via a de-coupled optimization method is an efficient modeling approach that is capable of doing handling complex decision scenarios in reconfigurable manufacturing activities. The approach supplies a effective decision framework for process planning optimization activities of the multidimensional nature. This kind of approach could be implemented more proficiently through intelligent techniques. Hence intelligent techniques may be used in manufacturing process planning optimization strategies that try to improve operating levels in reconfigurable manufacturing using the resultant advantages of improved performance levels.
About The Writer
Presently, Farayi Musharavati is definitely an Assistant Professor at Qatar College, Department of Mechanical and Industrial Engineering within the Condition of Qatar. He acquired a Ph.D. in Manufacturing Systems Engineering from College Putra Malaysia. He holds a BTech. (Honors) Degree in Mechanical and Production Engineering, an MSc. in Alternative Energy, and the other MSc. in Manufacturing Systems and processes Management, all in the College of Zimbabwe. He’s had four (4) years experience being an Engineer, and three (3) years teaching experience at college level just before doctorate studies. His Primary research interests include: Reconfigurable Manufacturing Systems, Reconfigurable Process Planning and Scheduling, Manufacturing Optimization, Intelligent Manufacturing, Advanced Manufacturing Technologies and Systems, and Applying Artificial Intelligence in Manufacturing. Secondary research interests include: Alternative Energy Technologies and Systems, Gas and oil Processing, Applying Lean and 6 Sigma Techniques, Healthcare Systems and processes Management, and Industrial Ecosystem and Environments. His teaching interests include: Manufacturing Systems Design, Systems Engineering, Industrial Systems Design, Manufacturing Technologies, Systems Simulation, Manufacturing Processes, Facilities Planning and style, Maintenance Engineering and Management, Expert Systems, and Engineering Management.
Featured within this Category