Academia.eduAcademia.edu

Figure 1 – uploaded by IEEE/CAA J. Autom. Sinica

See full PDFdownloadDownload figure
Fig. 1. | Closed-loop supply chain (CLSC).

Figure 1 | Closed-loop supply chain (CLSC).

Related Figures (15)

COMPARISON OF EXISTING SURVEY PAPERS
COMPARISON OF EXISTING SURVEY PAPERS
Fig. 2. The connection structure graph of an assembly. ig.3. The undirected graph of a product in Fig. 2.
Fig. 2. The connection structure graph of an assembly. ig.3. The undirected graph of a product in Fig. 2.
AND/OR graph is an important representation method in DSP. It can express the disassembly information of components. It not only expresses a DSP problem among components, but also effectively expresses the logical com- bination relationship among the components. The disadv- antage is that as the number of components increases, it is easy to produce a phenomenon of combination explosion. G = V, E) represents a disassembly AND/OR graph. The set of nodes is represented by V= {v,, vy, ..., Vy}. Each element in V has one-to-one correspondence to a component or subassembly of a product. The number of the subassemblies is represented by N in V. Components or subassemblies in a product are denoted by the nodes, while hyper-arcs represent disassembly tasks. A more complicated subassembly can be formed by two or more components or subassemblies to join together. The set of directed edges is represented by E = {e;}. A directed edge is represented by e;, representing a disassembly operation from node v; to node v,. Notice that an AND relationship is that the two lines connect with an arc, and two lines must exist at the same time; an OR relationship is that a node has a variety of decomposition methods, and each decomposition method can independently exist. According to the above description of the diagram, a ballpoint link structure is constructed as shown in Figs.5 and 6, respectively. ig. 5. Schematic diagram of a ballpoint.
AND/OR graph is an important representation method in DSP. It can express the disassembly information of components. It not only expresses a DSP problem among components, but also effectively expresses the logical com- bination relationship among the components. The disadv- antage is that as the number of components increases, it is easy to produce a phenomenon of combination explosion. G = V, E) represents a disassembly AND/OR graph. The set of nodes is represented by V= {v,, vy, ..., Vy}. Each element in V has one-to-one correspondence to a component or subassembly of a product. The number of the subassemblies is represented by N in V. Components or subassemblies in a product are denoted by the nodes, while hyper-arcs represent disassembly tasks. A more complicated subassembly can be formed by two or more components or subassemblies to join together. The set of directed edges is represented by E = {e;}. A directed edge is represented by e;, representing a disassembly operation from node v; to node v,. Notice that an AND relationship is that the two lines connect with an arc, and two lines must exist at the same time; an OR relationship is that a node has a variety of decomposition methods, and each decomposition method can independently exist. According to the above description of the diagram, a ballpoint link structure is constructed as shown in Figs.5 and 6, respectively. ig. 5. Schematic diagram of a ballpoint.
‘ig.4. The directed graph of a product in Fig. 2. construct a directed graph as shown in Fig. 4, where the nodes and edges mainly record the geometrics, constraints, and joint information of product components. Yet it is difficult to generate such a graph from product design information automatically. It can only be used for a product with a simple structure and a few components. Otherwise, it becomes very complicated.
‘ig.4. The directed graph of a product in Fig. 2. construct a directed graph as shown in Fig. 4, where the nodes and edges mainly record the geometrics, constraints, and joint information of product components. Yet it is difficult to generate such a graph from product design information automatically. It can only be used for a product with a simple structure and a few components. Otherwise, it becomes very complicated.
Fig. 6. The AND/OR diagram of a product.
Fig. 6. The AND/OR diagram of a product.
Therefore, PNs are extensively used in _ flexible manufacturing systems. A DPN of a product is established based on the concept of PN. A DPN of a ballpoint is shown in 2. FF
Therefore, PNs are extensively used in _ flexible manufacturing systems. A DPN of a product is established based on the concept of PN. A DPN of a ballpoint is shown in 2. FF
A constraint condition should be satisfied such that one subassembly cannot be obtained by multiple disassembly operations at the same time. For convenience, a disassembly matrix D = [d;] is used to define the relationship between subassemblies and operations. In d;, j and i denote disassembly operation j and subassembly i, respectively. Thus, the connection matrix of a product as shown in Fig. 5 can be expressed as:
A constraint condition should be satisfied such that one subassembly cannot be obtained by multiple disassembly operations at the same time. For convenience, a disassembly matrix D = [d;] is used to define the relationship between subassemblies and operations. In d;, j and i denote disassembly operation j and subassembly i, respectively. Thus, the connection matrix of a product as shown in Fig. 5 can be expressed as:
In order to present the geometrical and motion constraints between fasteners and components, Smith and Chen [49] adopted different matrix representations for fasteners, components, and constraints in computer recognizable formats for DSP. Smith and Chen [50] used four matrices: disassembly matrix and motion constraint matrix for components, disassembly matrix and motion constraint matrix for fasteners. A close relationship among product elements was included in an adjacent matrix. ElSayed et al. [51] developedmatrixrepresentationstostoreproducts’structureinform- ation, components’ constraint relations as well as connections. Products’ geometrical constraints and connections are represented by matrices in DSP. Jiang and Wang [52] proposed a constraint matrix to store data of component constraints of disassembly operations in the format of a matrix.
In order to present the geometrical and motion constraints between fasteners and components, Smith and Chen [49] adopted different matrix representations for fasteners, components, and constraints in computer recognizable formats for DSP. Smith and Chen [50] used four matrices: disassembly matrix and motion constraint matrix for components, disassembly matrix and motion constraint matrix for fasteners. A close relationship among product elements was included in an adjacent matrix. ElSayed et al. [51] developedmatrixrepresentationstostoreproducts’structureinform- ation, components’ constraint relations as well as connections. Products’ geometrical constraints and connections are represented by matrices in DSP. Jiang and Wang [52] proposed a constraint matrix to store data of component constraints of disassembly operations in the format of a matrix.
a used product and then proposed a new integer programming model for minimizing the total disassembly cost. Zhu et al. [61] introduced an efficient machine-readable disassembly information modeling method and linear program to find the optimal disassembly sequence. The information model adds dynamic capabilities to change the overall disassembly operation, which has been applied in the dismantling of mechanical products. COMPARISON OF MODELING METHOD FOR PRODUCTS TO BE DISASSEMBLED techniques to other formulations. Generally, DSP aims to find the optimal disassembly sequences given some objective functions and a domain. Different domains and objective functions are often involved.
a used product and then proposed a new integer programming model for minimizing the total disassembly cost. Zhu et al. [61] introduced an efficient machine-readable disassembly information modeling method and linear program to find the optimal disassembly sequence. The information model adds dynamic capabilities to change the overall disassembly operation, which has been applied in the dismantling of mechanical products. COMPARISON OF MODELING METHOD FOR PRODUCTS TO BE DISASSEMBLED techniques to other formulations. Generally, DSP aims to find the optimal disassembly sequences given some objective functions and a domain. Different domains and objective functions are often involved.
Fig.9. Uncertainty analysis.
Fig.9. Uncertainty analysis.
network. COMPARISON OF INTELLIGENT OPTIMIZATION ALGORITHM product disassembly. Tian et al. [112], [113] considered an energy management method for DSP integrating stochastic simulation and the maximum entropy principle. Behdad and Thurston [2] proposed the uncertainty sources that were different depend ing on the purpose of disassembly. Uncertainty in the incoming feedstock design, material, age, and quantity crea operations when recycling is per es enorm formed. ous impediments to cost-effective component reusing, remanufacturing or Uncertainty lies in dimensional instability and the possibility of causing damage to valuable components when disassembly is for product maintenance. Tian et al. [114], 115] ana yzed the main uncertainty factors for a disassembly process and perform the evaluation of their impact on the process. A approach was proposed to determine the optimal disassembly chance-constrained programming
network. COMPARISON OF INTELLIGENT OPTIMIZATION ALGORITHM product disassembly. Tian et al. [112], [113] considered an energy management method for DSP integrating stochastic simulation and the maximum entropy principle. Behdad and Thurston [2] proposed the uncertainty sources that were different depend ing on the purpose of disassembly. Uncertainty in the incoming feedstock design, material, age, and quantity crea operations when recycling is per es enorm formed. ous impediments to cost-effective component reusing, remanufacturing or Uncertainty lies in dimensional instability and the possibility of causing damage to valuable components when disassembly is for product maintenance. Tian et al. [114], 115] ana yzed the main uncertainty factors for a disassembly process and perform the evaluation of their impact on the process. A approach was proposed to determine the optimal disassembly chance-constrained programming
Fig. 10. | Percentages of publications based on the application of DSP. should be developed for the analysis of disassembly systems in terms of organizational dynamics and _ technological advances. Advanced DSP techniques are expected to become highly cost-effective in realizing product reuse and recovery to handle hig the use of h penalty cost and strict regulations associated with virgin material resources and waste management. It is increasingly important to design efficient disassembly lines, and solve planning and scheduling problems to automate and optimize a product recovery process [132]. Furthermore, we should inv estigate the issues arising from a disassembly line’s design and operations. We have to address the challenges in workload machine | balancing among its workstations and introduce eaming and AI approaches to provide fast and desired solutions to nonlinear and multi-objective disassembly line balancing, planning and scheduling problems. Besides, such problems subject to multi-resources constraints and reverse logistics require much research.
Fig. 10. | Percentages of publications based on the application of DSP. should be developed for the analysis of disassembly systems in terms of organizational dynamics and _ technological advances. Advanced DSP techniques are expected to become highly cost-effective in realizing product reuse and recovery to handle hig the use of h penalty cost and strict regulations associated with virgin material resources and waste management. It is increasingly important to design efficient disassembly lines, and solve planning and scheduling problems to automate and optimize a product recovery process [132]. Furthermore, we should inv estigate the issues arising from a disassembly line’s design and operations. We have to address the challenges in workload machine | balancing among its workstations and introduce eaming and AI approaches to provide fast and desired solutions to nonlinear and multi-objective disassembly line balancing, planning and scheduling problems. Besides, such problems subject to multi-resources constraints and reverse logistics require much research.
Related topics:
Artificial IntelligenceIntelligent Optimizationmodeling methodsSingle-Objective Optimization
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved