Manufacturing Process Sequence Research Articles

Deep learning and sequence mining for manufacturing process and sequence selection

ABSTRACT Automatic determination of manufacturing process sequences for the physical production of given part designs is key to facilitate on-demand cyber manufacturing. In this work, we propose an integrated framework that (i) identifies manufacturing features from 3D part designs using a Graph Neural Network (GNN), (ii) identifies the manufacturing processes necessary to produce all features in the part using a Convolutional Neural Network (CNN) that considers shape, material properties, and quality information, and (iii) outputs an ordered manufacturing sequence that can produce the designed part with the help of sequence mining. Using these methods, the knowledge required to enable automated manufacturing process selection is easily scalable and updatable without requiring manual population of ad-hoc or rule-based descriptions. We present exemplar implementations of the proposed framework by suggesting manufacturing sequences for discrete parts with multiple features. The suggested manufacturing sequences demonstrate the potential of the proposed framework for use in future on-demand cyber manufacturing applications.

Potential of prediction in manufacturing process and inspection sequences for scrap reduction

With increasing quality requirements, especially for safety-critical products such as in medical technology, manufacturing companies face the challenge, that the quality inspection of component characteristics may become a crucial factor for economic production. For increasing the ratio of correctly identified scrap without increasing the amount of physical inspection, prediction models pose a valuable tool. Prediction models in manufacturing build process models through process data collected during the manufacturing process. The prediction models can for example be used to predict the product quality from the process data, which enables the substitution of physical quality inspections through virtual inspections. Before selecting suitable prediction models, checkpoints in the manufacturing process and inspection sequence (MPIS) for integrating the prediction models profitably have to be identified. Therefore, the paper aims to provide a methodology to generate integrated quality strategies for MPIS that combine physical inspection with the prediction of quality characteristics. Furthermore, a methodology for the economic evaluation of the quality strategies as decision support is provided. Since the number of prediction models available for predicting quality based on process data is manifold, the explicit recommendation for the implementation of a certain prediction model is not the focus of this paper. The methodology presented rather focuses on the identification of suitable checkpoints in the MPIS for prediction model integration. For generating quality strategies suitable checkpoints in the manufacturing process sequence have to be identified. Therefore, three aspects are considered, the influence of the manufacturing process on the change in a characteristic and the overall importance of a characteristic for the component function as well as the ability of the characteristic and process for being predicted. A prediction recommendation is calculated that enables the derivation of quality strategies as a combination of physical and virtual inspection for different characteristics in different checkpoints. For the economic evaluation of the derived quality strategies two cases for supplementing physical inspection with prediction models to either compensate for the alpha- or beta-error rate are presented. The most reasonable quality strategy is chosen according to its economic potential. For validation, the applicability of the methodology for integrating prediction models in manufacturing process and inspection sequences to a use case of medical technology is demonstrated successfully. It is shown that the methodology supports users in identifying possible quality strategies and their economic evaluation for manufacturing process and inspection sequences as a basis for the decision regarding quality strategies to be implemented.

Process planning for hybrid manufacturing using additive friction stir deposition

Additive friction stir deposition (AFSD) provides a solid-state approach to metal deposition that does not rely on local melting and solidification, but rather on kinetic energy and plastic flow. In this study, AFSD is combined with structured light scanning, turning, and milling to produce metal components while considering the unique requirements imposed by the hybrid manufacturing process sequences. Two demonstrations are presented which include: 1) a cylindrical build plate selection to enable coordinate system transfer between deposition and turning of a hollow cone; and 2) intermittent deposition-machining operations with structured light scanning to fabricate a two-sided hexagon-cylinder geometry.

Design of demand-oriented flexible manufacturing process sequences under consideration of future and uncertain component variants

Manufacturing companies must increasingly produce different component variants. Thus, flexibility is becoming a success factor for manufacturing and must be considered in the planning of manufacturing process sequences (MPSs) to be successful. Since maximizing manufacturing flexibility results in additional costs, a demand-oriented consideration of flexibility is necessary to maximize the economic efficiency. For the planning of demand-oriented flexible MPSs, the integration of potential, uncertain component variants into the manufacturing planning is necessary, since these define the flexibility demand. The paper presents three challenges in the design of demand-oriented flexible MPSs. Furthermore, a concept for overcoming these challenges is outlined.

Digitalized manufacturing process sequences – foundations and analysis of the economic and ecological potential

As a consequence of the increasing digitalization of production, the availability of data from individual manufacturing processes and components is rising continuously. Digitalization affects various levels of production, such as the design of manufacturing processes, technology planning, quality monitoring, or logistics control. The contribution of this paper is to present the foundations for digitalized manufacturing process sequences and to develop data-based models and methods for the design of economically and ecologically optimized process sequences. For this purpose, a procedure for cross-process life cycle assessment was developed and carried out in order to evaluate the various ecological effects resulting from manufacturing and, in particular, different process designs. This ecological evaluation was combined with the evaluation of economic efficiency in order to provide an integrated evaluation as the basis for the design of manufacturing process sequences. In addition, the effects of the individual manufacturing process designs or process parameters on component quality were analyzed and an approach was derived for determining cross-process dependencies. The basis for these developments is a process sequence for the manufacture of pinion shafts used in electric car gearboxes. This digitalized process sequence consists of the processes soft cutting, gear hobbing, case-hardening, cylindrical grinding, and gear grinding, and over 100 components were manufactured with various process designs. This enabled an extensive amount of process and quality data to be collected throughout the process sequence. The application of the developed methods to the aforementioned process sequence underlines the high potential for the optimization of the target parameters quality, economy and ecology through cross-process analyses of digitalized manufacturing process sequences. Finally, an outlook on further developments regarding the digitalization of manufacturing process sequences is given.

A data-driven framework for learning the capability of manufacturing process sequences

Automatically acquiring knowledge of manufacturing process capabilities described in terms of the part shapes they can produce, materials they can process, and part qualities they can generate is necessary to enable on-demand cyber manufacturing. This paper aims to present a novel data-driven framework to (i) identify the sequences of processes used to manufacture a discrete part, and (ii) describe their capabilities via quantifiable descriptors of shape, material properties, and part quality. Specifically, given existing manufacturing data consisting of different parts and their corresponding manufacturing methods, the proposed framework utilizes a sequence mining algorithm to identify frequently occurring sequence patterns of different manufacturing processes. In addition, the manufacturing capability of each sequence pattern is described quantitatively in terms of the achievable shapes, material properties, and part quality metrics. Such manufacturing process capability descriptions can be queried to obtain suggestions of feasible process sequences with the capability to manufacture a new part design. An exemplar implementation of the proposed framework with a curated manufacturing dataset is given to illustrate how the framework can enable manufacturing process selection and planning. The high Confidence Rates achieved by the proposed framework show its predictive strength for use in Computer Aided Process Planning (CAPP).

Deterministic and cost-efficient change propagation analysis method for manufacturing process sequences

Manufacturing companies are increasingly confronted with the challenge of adapting components that are already established on the market. As a result, the existing manufacturing process sequences of the corresponding components must also be adapted. Due to the complex technical interdependencies of manufacturing process sequences, adaptations potentially cause change propagation. The change propagation must be considered to ensure the component’s functionality and to enable the selection of the most economical adaptation for the established manufacturing process sequence to implement the required component change. As a contribution to considering change propagation in manufacturing process sequences, a methodology for the deterministic and cost-efficient change propagation analysis is introduced. A case study from medical technology is presented to supplement the explanation of the methodology’s steps and to validate its applicability.

Methodology for the integrative adaption of manufacturing process and inspection sequences to component changes of safety–critical medical products

Ever-shorter product lifecycles and more frequently changing customer demands challenge manufacturing companies to change their products and the components these products are composed of in ever-shorter periods. As a result, the existing manufacturing process and inspection sequences (MPISs) of the corresponding components must be adapted. Particularly in the production of safety–critical components, such as in the medical industries, component changes represent a major challenge, as very high quality requirements are placed on the components and a costly re-certification of the adapted processes is necessary. Due to the high quality requirements, not only the manufacturing process but also the inspection processes must be adapted for the realization of a component change. Furthermore, a high degree of planning reliability is necessary when deriving adaptations of MPISs for component changes to keep the re-certification effort and adaptation costs for safety–critical components as low as possible. Therefore, a methodology is introduced for the integrative derivation of adaptation options of MPISs due to component changes that supports users in identifying suitable adaptations in a systematic and time-efficient way. The methodology is successfully applied to a use case from the medical industry and supports users in identifying adaptations in established MPISs to implement a component change.

Experimental study of a virtual allowables approach for the design of composite aircraft structures

Statistically derived allowables are representative for the behavior of a composite material system in a given structural context where data variability needs to be contemplated. In particular, they define the strength of the material as characterized by various coupon tests according to ASTM standard procedure subjected to the actual manufacturing process and layup stacking sequences. The current paper presents a detailed comparison between virtual allowables predictions obtained using MSC Digimat and the experimentally determined values of a certified material backed up by a 3x2x3 (batch/panel/specimen) stochastic approach. Tests generate the required data for building material model ab initio of carbon fibers reinforced epoxide resin in form of 2x2 twill woven fabric used for qualifying an innovative composite wing for the Next Generation Tilt Rotor (NGCTR). Numerical models of in-plane tension, compression and shear test methods have been assembled along with an associated material model allowing numerical predictions to be validated with the coupon level experimental results as final purpose. Standard statistical model has been adopted in order to include material and manufacturing process variabilities to define B-basis allowables. The calibrated statistical FEM method performed for different layup configurations have been used to fully characterize the mechanical behavior of the analyzed CFRP material and to predict performances for thicker laminates.

Design of economically-optimized manufacturing process sequences using cross-process models

Manufacturing companies have to design their manufacturing process sequences so that the product requirements are fulfilled cost-efficiently. It is important that process sequences are designed holistically and that the technological and economic relationships between individual processes are taken into account. It is particularly important to consider these relationships in serial production, because technology experts often design manufacturing processes one process at a time. While the different processes are linked at higher planning levels, the relationships between the manufacturing processes are insufficiently considered. At the design phase, the economic potential of cross-process optimization could be tapped into but remains unused. An approach for the design of economically-optimized manufacturing process sequences is presented in this paper. It describes how technological relationships in a process sequence can be modeled using cross-process models. The approach includes a newly developed model which can be used to economically evaluate process sequences. Within this model, process parameter-dependent variables of the individual manufacturing processes and temporal as well as technological relationships between the processes are considered in order to optimize manufacturing process sequences. A case study shows the potential of cross-process optimization.

Integrated Modelling and Simulation of the Additive Manufacturing and Heat Treatment Process Chain

Additive manufacturing of metal parts is widely used for different alloys and different types of application. During building of the parts, local heating of the powder generates a small melt pool with high thermal gradients followed by rapid solidification. Already solidified material is re-melted and affected by heating when layers are added. The thermal history of the moving heat source and the layer building of the part have a high influence on the formed microstructure, the risk of porosities and evolution of cracks. High stresses and permanent deformations develop during the building process, which lead to large deformations when e.g. the base plate is removed. Heat treatment of the parts is often used to reduce the stress level and to minimize the deformations. This paper presents an integrated modelling and simulation approach, where results from the additive manufacturing process are used as initial conditions for the subsequent heat treatment process. An integrated simulation approach has been developed and implemented as a dedicated solution to simulate the sequence of the additive manufacturing process and subsequent heat treatment steps. To get reasonable calculation times multiscale methods have been tested to perform virtual experiments, where the influence of different scanning strategies have been evaluated to optimize temperature distributions and the influence on mechanical performance. A unified creep model is used to describe the mechanical behavior of the material to ensure a consistent description through the large temperature interval and the different levels of time and strain rates.

Adaptive design of manufacturing process sequences in case of short-term disruptions in the production process

Manufacturing companies are more frequently confronted with short-term disruptions due to increasing manufacturing complexity. These disruptions lead to additional costs due to e.g. downtimes. Technological reaction potentials resulting from the adaptation of process parameters are currently neglected when reacting to disruptions. Moreover, technological interrelations between the manufacturing processes are not sufficiently examined. Thus, cost saving potentials are not fully utilized. Therefore, an adaptive reaction is necessary to exploit these potentials. In this paper, three challenges in the adaptive design of manufacturing process sequences in case of short-term disruptions are presented. Furthermore, a concept for overcoming these challenges is outlined.

Product Quality Detection through Manufacturing Process Based on Sequential Patterns Considering Deep Semantic Learning and Process Rules

Companies accumulate a large amount of production process data during product manufacturing. Sequence data from the mining production process can enable a company to evaluate the manufacturing process, to find the key factors affecting product quality, and to improve product quality. However, the production process mainly exists in the form of text. To solve this problem, we propose a novel frequent pattern mining algorithm (EABMC) based on the text context semantics and rules of the manufacturing process to remove redundant sequences and to obtain good mining results. In this algorithm, first, we use embeddings from language models (ELMo ) to improve the process of text similarity matching and to classify similar semantic processes into one class. Then, the manufacturing process unit (MPU) is proposed by extracting the characteristics of manufacturing process data according to the constraints of the manufacturing process and other conditions. The above two steps cause the complex manufacturing process sequence to merge and simplify. Once again, a frequent pattern mining algorithm (CloFAST) is used to explore the important manufacturing process relationships behind a large amount of manufacturing data. In addition, taking the data from a production enterprise in Guizhou Province as an example, the validity of the method is verified. Compared with other methods, this method is shown to have greater mining efficiency and better results and can find out the key factors that affect product quality, especially for text data.

A methodology for the ecological and economic assessment of manufacturing process sequences

Abstract Technology planning methodologies enable the design of alternative manufacturing process sequences that are capable of manufacturing a product, which fulfills its designated functions. Today, capable process sequences are assessed according to economic criteria to identify the preferable sequence. However, changing customer demands towards sustainable products and legal regulations force companies to consider ecologic criteria in process sequence assessment as well. Therefore, a new methodology to enable an integrated ecological-economic assessment of manufacturing processes is introduced. It is based on monetary valuation of environmental impacts. Subsequently, a procedure to identify levers to reduce environmental impacts of process sequences is outlined.

Identification of Unintended Effects Caused by Adaptations of Manufacturing Process Sequences for Safety-Critical Components

Changes of components are inevitable for companies to maintain competitiveness. To implement component changes, the manufacturing has to be adapted as well. For safety-critical components, unintended effects caused by adaptations pose a challenge due to the required reliability to ensure the component’s functionality. Therefore, a methodology to identify unintended effects of manufacturing adaptations is introduced and applied to a case study. Interdependencies within manufacturing are modeled through a multiple domain matrix and the model is extended by unintended effects within single manufacturing processes. Graph theory is further used to analyze the propagation of the instigating change within the manufacturing system.

Experimental evaluation of orientation and temperature dependent material stress-strain curves of Zr2.5%Nb Indian pressure tube material and development of a suitable anisotropic material model

Because of their hexagonal close-packed (HCP) structure and easy prismatic slip, the tensile properties of Zirconium based alloys show strong texture dependence and hence, anisotropy. The type of texture, microstructure and hence, the mechanical properties depend upon the sequence of manufacturing process followed in fabrication of components, such as pressure tubes (PT) in pressurized heavy water type reactors. Due to small wall thickness (i.e., 3.75 mm) of the pressure tubes, designing and machining of specimens in all the three directions, especially in the radial direction, is a challenge and this explains the reason for scarcity of work in this area in research literature. In this work, different types of specimens have been machined from transverse, longitudinal and radial orientations of the Zr2.5%Nb Indian PT following a novel design in order to study the anisotropy in the plastic deformation and hardening properties. In addition, shear specimens have been extracted from longitudinal-transverse plane to study the plastic deformation in shear. Tests on tensile and shear specimens have been carried out at room temperature (25), 100, 200 and 300 °C respectively. Using the experimental data, the parameters of Hill’s anisotropic yield function have been evaluated. In addition to evaluating the parameters at initial yield point, the nature of variation of these parameters with equivalent plastic strain, has also been established. This new material model where the anisotropic parameters evolve with plastic strain, shall be useful for design (by analysis) and integrity assessment of the pressure tubes for different types of postulated loading conditions.

(Invited) Recent Advances in SiC CMP

The need for high-quality single crystal SiC substrates continues to grow as the demand for SiC-based power and RF devices increases. The physical properties of SiC result in better thermal conductivity, higher breakdown voltage, faster switching performance and lower resistivity when compared to similar devices based on silicon. Even with the performance advantages, the adoption rate of SiC is slowed by high cost and inherent difficulty of processing the material. Conventional processing of SiC substrates involves double-sided batch lapping followed by batch diamond polishing then final polishing with a non-diamond slurry to achieve an epi-ready surface. This process flow involves multiple process tools and is fraught with issues including the need for wafer sorting to group wafers by thickness, manual loading and unloading of wafers, low yields, and unacceptable risk of batch loss (an entire batch of wafers may be scrapped if even one wafer cracks or breaks during process). This paper presents an improved process for single-wafer CMP during SiC substrate manufacturing or backside thinning. In contrast to batch polishing, key improvements include individual wafer processing, fully-automated handling, and elimination of the need for wafer sorting and grouping by thickness. Recent development efforts have achieved removal rates for the 4H Si face up to 2.5 µm/hour for the first step CMP. Removal rates for the C-face can be as much as 3-4x higher consistent with other findings in the industry. Fig 1 shows surface finish after the polishing sequence is complete which is consistently at or below 1 Å Ra as measured by AFM. The combination of an optimized grind followed by an optimized CMP process provides a viable path to both improving wafer yield and reducing costs for SiC wafer manufacturing or backside thinning process sequences. Figure 1

Towards smart manufacturing process selection in Cyber-Physical Systems

Cyber-Physical Systems (CPS) are becoming increasingly important in manufacturing due to the digitization of the industry driven by advances in technology and connectivity. However, the variation of process selection criteria, both geometric and non-geometric, across the main groups of manufacturing processes, i.e., additive, subtractive, and deformation, makes it difficult to implement smart manufacturing process sequence selection which is an essential step toward achieving an autonomous CPS. This paper presents a conceptual solution to process selection called Constraint Satisfaction Problem for Manufacturing (CSP4M) which leverages current engineering design methods and architectures.

Integrative Technology and Inspection Planning—A Case Study in Medical Industry

In most trendsetting industries like the aerospace, automotive and medical industry functionally critical parts are of highest importance. Due to strict legal requirements regarding the securing of the functionality of high-risk parts, both production costs and quality costs contribute significantly to the manufacturing costs. Thus, both types of costs have to be taken into consideration during the stage of technology planning. Due to the high variety of potential interactions between individual component properties as well as between component properties and manufacturing processes, the analysis of the influence of the manufacturing history on an efficient design of inspection processes and inspection strategies is extremely complex. Furthermore, the effects of inspection strategies and quality costs on the planning of manufacturing process sequences cannot be modeled to date. As a consequence, manufacturing and inspection processes are designed separately and thus a high cost reduction potential remains untapped. In this paper, a new approach for an integrative technology and inspection planning is presented and applied to a case study in medical industry. At first, existing approaches with regard to technology and inspection planning are reviewed. After a definition of relevant terms, the case study is introduced. Following, an approach for an integrative technology and inspection planning is presented and applied to the case study. In the presented approach, the complex causalities between technology planning, manufacturing history, and inspection planning are considered to enable a cost-effective production process and inspection sequence design.

Environmental impact assessment of an innovative strategy based on an additive and subtractive manufacturing combination

Recently, the combination of additive and subtractive manufacturing technologies is gaining significant attention from both the industrial and academic sectors. Due to consolidated advantages of individual techniques, this combination provides the capability of producing innovative products, and opens new perspectives for developing new remanufacturing/manufacturing strategies. In this paper, an innovative strategy is proposed. The strategy aims to manufacture a new part directly from an EoL part (or existing part), using a sequence of additive and subtractive manufacturing processes. The existing part is transformed into the new part without involving the material recycling phase. This paper particularly focuses on the environmental assessment of the proposed strategy. For this purpose, a methodology based on the life cycle assessment method is developed. The environmental trade-offs between the proposed strategy and the conventional strategy are also discussed through the case study. The conventional strategy uses conventional processes (such as material recycling, casting and machining) to manufacture the new part, whereas the proposed strategy is based on combining electron beam melting (EBM) and CNC machining processes. The results show that the proposed strategy becomes more environmentally friendly when the material volume of existing part reused increases more than 60%. The proposed methodology can help designers and manufacturers to select the most suitable strategy to manufacture new parts from existing parts with minimum environmental impacts.