Connector-link-part-based disassembly sequence planning

Under the trend of concurrent engineering, the correspondence between functions and physical structures in product design is gaining importance. Between the functions and parts, connectors are the basic unit for engineers to consider. Moreover, the relationship between connector-liaison-part will help accomplish the integration of information. Such efforts will help the development of the Knowledge Intensive CAD (KICAD) system. Therefore, we proposed a Connector-liaison-part-based disassembly sequence planning (DSP) in this study. First, the authors construct a release diagram through an interference relationship to express the priority of disassembly between parts. The release diagram will allow designers to review the rationality of product disassembly planning. Then, the cost calculation method and disassembly time matrix are established. Last, the greedy algorithm is used to find an appropriate disassembly sequence and seek suggestions for design improvement. Through the reference information, the function and corresponding modules are improved, from which the disassembly value of a product can be reviewed from a functional perspective. In this study, a fixed support holder is used as an example to validate the proposed method. The discussion of the connector-liaison-part will help the integration of the DSP and the functional connector approach.

Coherent Dialog Generation with Query Graph

Learning to generate coherent and informative dialogs is an enduring challenge for open-domain conversation generation. Previous work leverage knowledge graph or documents to facilitate informative dialog generation, with little attention on dialog coherence. In this article, to enhance multi-turn open-domain dialog coherence, we propose to leverage a new knowledge source, web search session data, to facilitate hierarchical knowledge sequence planning, which determines a sketch of a multi-turn dialog. Specifically, we formulate knowledge sequence planning or dialog policy learning as a graph grounded Reinforcement Learning (RL) problem. To this end, we first build a two-level query graph with queries as utterance-level vertices and their topics (entities in queries) as topic-level vertices. We then present a two-level dialog policy model that plans a high-level topic sequence and a low-level query sequence over the query graph to guide a knowledge aware response generator. In particular, to foster forward-looking knowledge planning decisions for better dialog coherence, we devise a heterogeneous graph neural network to incorporate neighbouring vertex information, or possible future RL action information, into each vertex (as an RL action) representation. Experiment results on two benchmark dialog datasets demonstrate that our framework can outperform strong baselines in terms of dialog coherence, informativeness, and engagingness.

Assembly Sequence Planning Using Artificial Neural Networks for Mechanical Parts Based on Selected Criteria

The proposed model of the neural network describes the task of planning the assembly sequence on the basis of predicting the optimal assembly time of mechanical parts. In the proposed neural approach, the k-means clustering algorithm is used. In order to find the most effective network, 10,000 network models were made using various training methods, including the steepest descent method, the conjugate gradients method, and Broyden–Fletcher–Goldfarb–Shanno algorithm. Changes to network parameters also included the following activation functions: linear, logistic, tanh, exponential, and sine. The simulation results suggest that the neural predictor would be used as a predictor for the assembly sequence planning system. This paper discusses a new modeling scheme known as artificial neural networks, taking into account selected criteria for the evaluation of assembly sequences based on data that can be automatically downloaded from CAx systems.

Landing gear disassembly sequence planning using multi-level constraint matrix ant colony algorithm

The life cycle of most complex engineering systems is greatly a function of maintenance. Generally, most maintenance operation usually requires the removal of failed part. Disassembly sequence planning is an optimization program that seeks to identify the optimal sequence for the removal of the failed part. Most studies in this area usually, use single constraint matrix while implementing varied complex algorithm to identify the optimal sequence that saves time associated with carrying out maintenance operation. The used of single constraint matrix typically has the drawback of computer higher storage requirement as well as time consumption. To address this problem, this study proposes Multi-Level Constraint Matrix Ant Colony Algorithm (MLCMACA). MLCMACA efficiency was validated using complex aircraft landing gear systems in comparison with genetic algorithms. The result shows MLCMACA superior performance from the perspective of reduced search time and faster tracking of optimal disassembly sequence. Hence is recommended for handling of disassembly sequence planning problems.

Objective Disassembly Sequence Planning of Used Mobile Phones Based on Improved Hybrid Graph Disassembly Model

To solve the problems of environmental pollution and waste of resources caused by used mobile phones, the study of objective disassembly sequence planning is carried out for used mobile phones. In view of the connection of mobile phone parts with multiple parts and the need to disassemble components, the concepts of containment, exclusion, and components are integrated into the hybrid graph. An improved hybrid graph is proposed and the improved hybrid graph disassembly model suitable for mobile phone disassembly is established. The ant colony algorithm is used to search for the optimal disassembly sequence, with the objective of minimum disassembly time. Finally, the improved hybrid graph disassembly model is applied to obtain the disassembly solution of HUAWEI Honor 6. The experimental results demonstrate that the disassembly sequence generated by the improved hybrid graph disassembly model can describe the actual disassembly process of disassembling components with less disassembly time.