Assembly Sequence Evaluation Research Articles

Large language models empower the reliability of disassembly in remanufacturing

Disassembly is a crucial stage in the product lifecycle, significantly contributing to remanufacturing, waste minimization, and resource recovery by facilitating the reuse and recycling of components and materials. However, engineers often face a plethora of choices during the disassembly process, each linked to uncertain outcomes, leading to substantial challenges. To address this, we introduce an assembly sequence evaluation model offering a quantitative analysis for diverse conditional decisions. Initially, an event graph is devised to outline the hierarchical structure, effectively shrinking the sequence space through the utilization of engineering semantics. Subsequently, a reinforcement learning model is established, with the reward function defined by large language models that use tailored prompts for their sequences. Deep Q-learning is then applied to train the reinforcement learning model, incorporating a selected set of ground truth sequences to highlight the correct one. Finally, a case study on a decelerator disassembly is presented to illustrate the effectiveness of the proposed method.

Optimization Method for Assembly Sequence Evaluation Based on Assembly Cost and Ontology of Aviation Reducers

Optimization Method for Assembly Sequence Evaluation Based on Assembly Cost and Ontology of Aviation Reducers

Mechanical Assembly Sequence Determination Using Artificial Neural Networks Based on Selected DFA Rating Factors

In this paper, an assembly sequence planning system, based on artificial neural networks, is developed. The problem of artificial neural network itself is largely related to symmetry at every stage of its creation. A new modeling scheme, known as artificial neural networks, takes into account selected DFA (Design for Assembly) rating factors, which allow the evaluation of assembly sequences, what are the input data to the network learning and then estimate the assembly time. The input to the assembly neural network procedure is the sequences for assembling the parts, extended by the assembly’s connection graph that represents the parts and relations between these parts. The operation of a neural network is to predict the assembly time based on the training dataset and indicate it as an output value. The network inputs are data based on selected DFA factors influencing the assembly time. The proposed neural network model outperforms the available assembly sequence planning model in predicting the optimum assembly time for the mechanical parts. In the neural networks, the BFGS (the Broyden–Fletcher–Goldfarb–Shanno algorithm), steepest descent and gradient scaling algorithms are used. The network efficiency was checked from a set of 20,000 test networks with randomly selected parameters: activation functions (linear, logistic, tanh, exponential and sine), the number of hidden neurons, percentage set of training and test dataset. The novelty of the article is therefore the use of parts of the DFA methodology and the neural network to estimate assembly time, under specific production conditions. This approach allows, according to the authors, to estimate which mechanical assembly sequence is the most advantageous, because the simulation results suggest that the neural predictor can be used as a predictor for an assembly sequence planning system.

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.

An augmented reality tool to validate the assembly sequence of a discrete product

The need for more flexible tools and reduction of time and cost has led to the implementation of augmented reality (AR) and virtual reality (VR) techniques in the product design and development process. Those techniques have already been used in the conceptual, manufacturing and assembly stages of product design instead of or in extension of the physical prototypes. Such virtual applications have demonstrated superior performance in assembly process design and evaluation of activities that present information about different assembly states in real time, thanks to their flexibility in manipulating and creating new working scenarios. Here, the development of an AR application, called PoliART, aimed at the visual evaluation of assembly sequences at early stages of design is presented. At an industrial level this allows collaborative work between designers and manufacturing engineers from the very beginning in order to consider assembly devices, times and resources, with a short implementation time and reduced costs.

Generation and evaluation of assembly sequences in computer–automated process planning

This paper discusses the sequencing of operations for the assembly of a product, and the effect of the methodology that is used on the efficiency and cost–effectiveness of the assembly system. Using two typical examples, the impact of the choice of a certain assembly sequence on the physical layout of the plant and on its performance are examined. The assembly sequence has a significant influence on the manufacturing cost of a product and this aspect has been discussed.

A Computer Aided Assembly Sequence Evaluation Method Based on the Fuzzy Theory for the Complex Products

Aim at the characteristic of airplane assembly sequence evaluation, which has too many rules, and the quantity process of qualitative rules is too complicated, this paper presents an evaluation mode based on the fuzzy theory and the multi-index progressive model. According to the evaluation rules space and the coefficient distributed method of rules compared, it divides the rules space into force associated decision-making gather and weak associated decision-making gather by defining the cut value. Then it adopts the maximal subjection fundamental to implement the function of assembly sequence evaluation. An instance about airplane sub-assembly is used to prove this method can solve a practical application availably.

Assembly Sequence Planning for Aircraft Component Based on Improved Clashes Matrix

The aircraft parts contain the information of a great number of curves and surfaces, which makes it a big challenge for the aircraft components’ Assembly Sequence Planning (ASP) processes. The traditional interference matrix has limitation of expressing the assembly direction and can easily lead to the failure of ASP optimization. Hence, an improved interference matrix is provided in this paper to deal with the aircraft ASP problem based on the Genetic Algorithms (GA). With the mainly consideration of assembly time according to the assembly sequence evaluation criteria, the objective function is established and evaluated. This paper presents an application of the method in the aircraft cabin door assembly process supported by an 863 program. Meanwhile, the verification of the approach is shown in the practical example on MATLAB platform.

Augmented assembly technologies based on 3D bare-hand interaction

Augmented reality has been applied to develop augmented assembly systems. However, most reported studies used pre-defined assembly information; AR is predominantly used to display information and interactions between users and the augmented environment are limited. This paper presents 3D bare-hand interaction in an augmented assembly environment to manipulate and assemble virtual components. A hybrid method based on constraint analysis is presented, which interprets users’ manual assembly intents robustly without the need for auxiliary CAD information. Algorithms for assembly constraint recognition, assembly tool operations and assembly sequence evaluation have been formulated. An augmented assembly system has been developed in this research.

Modeling of the Method of Product Assembly Sequence Evaluation

Influence factors of geometric feasible assembly sequence in actual application are studied. Based on the requirements of practical process rules and the integrity of index system, concrete calculation method and quantitative formulas of evaluation index system are proposed. According to the parallelism of actual assembly sequence, a unified algorithm—optimal order reference set of assembly performance membership degree calculation for assembly unit is put forwarded. Concrete application steps are verified through an example. Analysis of the example shows that the proposed method satisfies the requirements of practical process rules and the integrity of index system. The optimal selection of geometric feasible assembly sequence is realized. The new method has realistic significance to automatic assemble application.

Multi‐objective harmonious colony‐decision algorithm for more efficiently evaluating assembly sequences

PurposeComplex products, such as aircrafts and ships, are assembled from many parts and there are many available assembly sequences. Selecting the best from the available assembly sequences is challenging because of many factors, such as assembly performance, assemblability, assembly cost, assembly quality and assembly time. The purpose of this paper is to investigate a new and efficient algorithm aimed at this goal.Design/methodology/approachA new and efficient algorithm evaluating assembly sequences based on multi‐objective harmonious colony‐decision method is presented. This algorithm mainly includes three key steps: first, presenting the priority relationship between assembly sequences and the coefficient matrix for these objectives: assembly performance, assemblability, assembly cost, assembly quality, assembly time, and so on by several experts; second, calculating the maximum of harmonious values and harmonious priority value; third, if the maximum of harmonious priority value is not negative, the algorithm ends. Then the priority relationships of assembly sequences are obtained and the optimal assembly sequence can be selected.FindingsThis algorithm can efficiently support several experts to evaluate assembly sequences according to plenty of evaluation objectives and then to output a harmonious and recognized priority relationship of assembly sequences.Practical implicationsThe algorithm is applied successfully to evaluate assembly sequences and select the optimal assembly sequence for component of aircraft's wing with fixtures.Originality/valueThis algorithm provides a new method to synthetically evaluate assembly sequences for complex product according to multi‐objectives.

Case study: the impact of assembly reorientations on assembly time

Many prior-assembly planners have considered the number of assembly direction reorientations as an assembly sequence evaluation and selection criterion for assembly sequence planning. However, little study has been conducted to evaluate the effectiveness of the selected assembly sequences. This paper studies the impact of assembly direction reorientations on assembly time. Results of the case study show that, for both robot and human operator assembly processes, the number of reorientations in an assembly sequence has a significant impact on assembly time. The results support the study research hypothesis that using an assembly sequence which requires more assembly direction reorientations results in longer assembly time. The study conducted in this paper helps verify and quantify the importance and effectiveness of reducing the number of assembly direction reorientations in assembly sequence planning.

Planning for STEP-based electro-mechanical assemblies: an integrated approach

Assembly planning is a critical link between design and manufacturing stages. It is particularly critical for production automation and manufacturing integration due to the combinatorial complexity and the requirements of both flexibility and productivity. This paper presents an integrated approach and framework for planning STEP-based electro-mechanical assemblies (EMAs), including STEP-based assembly modelling, representation, planning and optimization of assembly sequences. Based on the author's previous work, the hierarchical relation model is used for a unified description of causal relations both on assembly level and on feature-based single part level. The generic product assembly model is organized in terms of STEP using mainly the entities of integrated resources and partially the self-defined entities necessary for assembly design and planning. With this representation, all feasible assembly sequences can be generated through reasoning and decomposing the levelled feasible subassemblies, and they are represented and visualized through a Petri net, that is, the assembly Petri net model is obtained. To obtain a good assembly sequence, the generated assembly sequences are evaluated and optimized by considering the assemblability and assembly sequence evaluation index. A prototype system is developed based on the proposed STEP-based framework for assembly planning. Finally, an industrial case study is provided.

A virtual prototyping approach to generation and evaluation of mechanical assembly sequences

It is desirable to integrate assembly planning with design early in the product development process. Usually, assembly plans can be generated by using either automated or interactive approaches. However, both kinds of approach have limitations. Interactive approaches require extensive user input in the tedious form of answering questions, whereas automated approaches can only be applied to products with simple component configuration and geometry. Based on virtual prototyping (VP), an intuitive and efficient approach is proposed for assembly planning by combining the potential strengths of the two kinds of approach. The main idea of this approach is to extract human expertise of mechanical assembly from human-computer interaction, and then use it for the automatic generation and interactive evaluation of assembly sequences. To incorporate expert knowledge, one or a few initial sequences are firstly created in a lesson by showing the way using the human-computer interface of virtual reality, where some intuitive and meaningful information such as motion path and assembly tool is captured automatically. Then, after representing assembly constraints as precedence expressions by reasoning about these initial sequences, all feasible sequences are generated efficiently based mainly on simple symbolic reasoning in place of costly geometric reasoning. In this way, the complexity of assembly sequence planning is reduced dramatically from seeking paths in a high-dimensional state space to a discrete graph-based searching problem. Finally, practical or good sequences are selected from the feasible sequences by exploiting the VP intuitive manipulation capability for judging the cost of a particular assembly operation. In the present authors’ experience, VP-based interaction has been proven to be a powerful and intuitive paradigm for both generation and evaluation of assembly plans. The feasibility of the proposed method is demonstrated through an implemented prototypical system called VPASPE, which has been tested with some practical assemblies.

A pragmatic approach to interactive assembly sequence evaluation

An assembly-oriented design system has been developed which includes several analysis tools to improve product assemblability during product development. One of these tools supports the parallel development of the product design and the assembly sequence, thus exploiting the benefits of concurrent consideration of product and process. However, this approach requires some method for evaluating the sequence against requirements. Previous work on assembly sequence evaluation has concentrated on identifying the best from a set of ranked alternatives. When a single sequence is constructed, as with this tool, another method is needed. This paper reports the development of this novel methodology for evaluating individual assembly sequences. A review of the relevant literature has found several measures for identifying good assembly sequences from a ranked list and the fundamental sequence attributes extrapolated and aggregated. This leads to the proposal of four new indices: insertion index, stability index, difficulty index and complexity index. A large number of assembly sequences have been analysed to define limiting values for the indices such that they can quantify the potential of an incomplete sequence resulting in a satisfactory solution. The application of these indices in concurrent design and assembly planning is illustrated through an industrial case study.

Multicriteria Evaluation of Assembly Sequences in Flexible Assembly Systems

Abstract This article presents a computerized multi-criteria evaluation system for assembly sequences, which evaluates all feasible assembly sequences based on ease of assembly criteria and finds the most efficient assembly sequence in the Flexible Assembly Systems (FAS) environment. The proposed system, called CASES (Computer-aided Assembly Sequence Evaluation System), automatically evaluates assembly sequences in two stages: (1) Evaluation based on assembly activity performance and (2) Evaluation based on the FAS performance. The first stage of CASES aims to reduce the solution space of all the feasible assembly sequences and finds only promising feasible sequences. The evaluation criteria are based on assembly activity performance such as equipment constraints, ease of part handling, ease of part joining, and the number of tool changes during an assembly process. With the reduced set of assembly sequences, the second stage finds the most efficient assembly sequence in the FAS environment. For evaluation based on the FAS performance, CASES solves a scheduling problem, MFASSP (Modified FAS Scheduling Problem), to find an assembly sequence which provides the minimal makespan. CASES is illustrated with a real assembly.

Evaluation of assembly sequences in an assembly-oriented design environment

To ensure that the best assembly sequence has been generated for a given design, it is necessary to use evaluation metrics. Most sequence generation systems fail to do this, offering only differing levels of validation criteria. Studies have also shown that sequence evaluation is hardly ever considered in industry. Design for assembly includes measures that evaluate assembly sequences, but design geometry is also included. However, it is found that geometry-independent measures are diffcult to define. After detailed consideration of these issues, this paper offers some suggestions for future approaches to sequence evaluation.

Interactive evaluation of assembly sequences using augmented reality

This paper describes an interactive tool for evaluating assembly sequences using the novel human-computer interface of augmented reality. The goal is to be able to consider various sequencing alternatives of the manufacturing design process by manipulating both virtual and real prototype components. The augmented reality-based assembly evaluation tool would allow a manufacturing engineer to interact with the assembly planner while manipulating the real and virtual prototype components in an assembly environment. Information from the assembly planner can be displayed directly superimposed on the real. A sensing technique is proposed that uses computer vision along with a system of markers for automatically monitoring the assembly state as the user manipulates the assembly components. An implemented system called AREAS is described. The advantage of using mixed prototyping and augmented reality as means of capturing human intuition in assembly planning is also discussed.

Generation of Optimized Assembly Sequences Using Genetic Algorithms

This paper describes a method based on genetic algorithms for the generation and the evaluation of assembly sequences. Genetic algorithms are here used to drastically reduce the high computational time, usually necessary to evaluate the best assembly sequences, owing to ‘combinatorial explosion’ phenomena. The generation of optimized sequences is performed using an appropriate fitness function which takes into account simultaneously the geometrical constraints, the minimization of gripper changes and object orientations, and the possibility of grouping similar assembly operations (screwing, pressing, etc.). The paper also presents the chromosome structure used in the system, the genetic operators and, finally, a meaningful example of application.

Evaluation of assembly sequences using generalized flexible assembly systems scheduling problem

The evaluation and selection of assembly sequences are performed in the planning stage of assembly processes. This paper deals with the effect of selecting particular assembly sequences to the performance of flexible assembly systems(FAS). The performance of FAS is evaluated using the generalized FAS scheduling problem(GFASSP). Compared to the conventional FASSP, GFASSP tries to select the most efficient assembly sequences for each product as well as minimize cycle time to complete the assemblies in FAS environment while considering the transfer time of subassemblies.