0

ASME Conference Presenter Attendance Policy and Archival Proceedings

2017;():V003T00A001. doi:10.1115/MSEC2017-NS3.
FREE TO VIEW

This online compilation of papers from the ASME 2017 12th International Manufacturing Science and Engineering Conference (MSEC2017) represents the archival version of the Conference Proceedings. According to ASME’s conference presenter attendance policy, if a paper is not presented at the Conference by an author of the paper, the paper will not be published in the official archival Proceedings, which are registered with the Library of Congress and are submitted for abstracting and indexing. The paper also will not be published in The ASME Digital Collection and may not be cited as a published paper.

Commentary by Dr. Valentin Fuster

Manufacturing Equipment and Systems: Advances in Cyber Physical Systems, Stochastic Modeling, and Sensor Networks in Advanced Manufacturing

2017;():V003T04A001. doi:10.1115/MSEC2017-2888.

Researchers at the National Institute of Standards and Technology (NIST) have developed a set of draft standard test methods for measuring and promoting software agility in industrial robots. These test methods are being used as the basis for an upcoming competition called the Agile Robotics for Industrial Applications Competition (ARIAC). ARIAC is being used to promote and push forward the state of the art in software agility and enable technology transfer between academia and industry. This paper describes the background about the test methods, how they were developed, how they will be applied to the ARIAC Competition, and additional information about the ARIAC Competition.

Topics: Robotics
Commentary by Dr. Valentin Fuster
2017;():V003T04A002. doi:10.1115/MSEC2017-2955.

Robotic bin picking requires using a perception system to estimate the posture of parts in the bin. The selected singulation plan should be robust with respect to perception uncertainties. If the estimated posture is significantly different from the actual posture, then the singulation plan may fail during execution. In such cases, the singulation process will need to be repeated. We are interested in selecting singulation plans that minimize the expected task completion time. In order to estimate the expected task completion time for a proposed singulation plan, we need to estimate the probability of success and the plan execution time. Robotic bin picking needs to be done in real-time. Therefore candidate singulation plans need to be generated and evaluated in real-time. This paper presents an approach for utilizing computationally efficient simulations for on-line evaluation of singulation plans. Results from physical experiments match well with predictions obtained from simulations.

Commentary by Dr. Valentin Fuster
2017;():V003T04A003. doi:10.1115/MSEC2017-3001.

Currently, design and control of HVAC system in buildings rely heavily on simulation tools. However, the common tools available often fail to optimize occupants’ comfort directly, nor do they consider real-time variations in occupancy that affect comfort and energy performance.

To address these limits, this research designed an occupancy-based and thermal comfort-driven building automation simulation model. A single-space prototype lab room was co-simulated using EnergyPlus and MATLAB with the help of BCVTB and MLE+ as middleware. Various climate scenarios from four cities in the U.S. in different seasons were examined. Results suggest that overall, compared to a conventional temperature-driven control strategy baseline, the proposed system can minimize thermal comfort violation (in term of PMV model, |PMV|>0.5 is considered as a violation) to 7% and reduce occupants’ thermal discomfort by 62.5% on average. Meanwhile, energy consumption remains same or reduced (up to 2% reduction). Due to its simplicity, this strategy is relatively easy to implement in real-world building automation systems with appropriate sensor placement in modern buildings.

Commentary by Dr. Valentin Fuster

Manufacturing Equipment and Systems: Advances in Data Analytics and Engineering Modeling for Intelligent Manufacturing Systems

2017;():V003T04A004. doi:10.1115/MSEC2017-2695.

Fine-scale characterization and monitoring of spatiotemporal processes are crucial for high-performance quality control of manufacturing processes, such as ultrasonic metal welding and high-precision machining. However, it is generally expensive to acquire high-resolution spatiotemporal data in manufacturing due to the high cost of the 3D measurement system or the time-consuming measurement process. In this paper, we develop a novel dynamic sampling design algorithm to cost-effectively characterize spatiotemporal processes in manufacturing. A spatiotemporal state-space model and Kalman filter are used to predictively determine the measurement locations using a criterion considering both the prediction performance and the measurement cost. The determination of measurement locations is formulated as a binary integer programming problem, and genetic algorithm is applied for searching the optimal design. In addition, a new test statistic is proposed to monitor and update the surface progression rate. Both simulated and real-world spatiotemporal data are used to demonstrate the effectiveness of the proposed method.

Commentary by Dr. Valentin Fuster
2017;():V003T04A005. doi:10.1115/MSEC2017-2771.

It is attractive to reduce the total cost of a manufacture system with real-time control of the production. The total cost mainly consists of the production cost, the penalty of the permanent production loss, and the Work-In-Process (WIP) inventory level cost. However, it is difficult to derive an analytical model of manufacture system due to the complexity of starved and blocked phenomena, the random failure and maintenance processes. Therefore, finding a real-time control policy for the manufacture system without exact analytical model is dearly needed. In this paper, a novel reinforcement learning based control decision policy is proposed based on the action of switching the machines on or off at the start of each time slot. Firstly, a simulation model is developed with MTBF and MTTR evaluated from the history data to collect samples. Then, a reinforcement learning method, specifically, Least-Square-Policy-Iteration method, is applied to obtain a sub-optimal policy. The simulation results show that the proposed method performs well in reducing the total cost.

Commentary by Dr. Valentin Fuster
2017;():V003T04A006. doi:10.1115/MSEC2017-2782.

As robot systems become increasingly prevalent in manufacturing environments, the need for improved accuracy continues to grow. Recent accuracy improvements have greatly enhanced automotive and aerospace manufacturing capabilities, including high-precision assembly, two-sided drilling and fastening, material removal, automated fiber placement, and in-process inspection. The accuracy requirement of those applications is primarily a function of two main criteria: (1) The pose accuracy (position and orientation accuracy) of a robot system’s tool center position (TCP), and (2) the ability of a robot system’s TCP to remain in position or on-path when loads are applied. The degradation of a robot system’s tool center accuracy can lead to a decrease in manufacturing quality and production efficiency. Given the high output rate of production lines, it is critical to develop technologies to verify and validate robot systems’ health assessment techniques, particularly the accuracy degradation. In this paper, the National Institute of Standards and Technology’s (NIST) effort to develop the measurement science to support the monitoring, diagnostics, and prognostics (collectively known as prognostics and health management (PHM)) of robot accuracy degradation is presented. This discussion includes the modeling and algorithm development for the test method, the advanced sensor development to measure 7-D information (time, X, Y, Z, roll, pitch, and yaw), and algorithms to analyze the data.

Topics: Robots
Commentary by Dr. Valentin Fuster
2017;():V003T04A007. doi:10.1115/MSEC2017-2787.

Manufacturing work cell operations are typically complex, especially when considering machine tools or industrial robot systems. The execution of these manufacturing operations require the integration of layers of hardware and software. The integration of monitoring, diagnostic, and prognostic technologies (collectively known as prognostics and health management (PHM)) can aid manufacturers in maintaining the performance of machine tools and robot systems by providing intelligence to enhance maintenance and control strategies. PHM can improve asset availability, product quality, and overall productivity. It is unlikely that a manufacturer has the capability to implement PHM in every element of their system. This limitation makes it imperative that the manufacturer understand the complexity of their system. For example, a typical robot systems include a robot, end-effector(s), and any equipment, devices, or sensors required for the robot to perform its task. Each of these elements is bound, both physically and functionally, to one another and thereby holds a measure of influence. This paper focuses on research to decompose a work cell into a hierarchical structure to understand the physical and functional relationships among the system’s critical elements. These relationships will be leveraged to identify areas of risk, which would drive a manufacturer to implement PHM within specific areas.

Topics: Manufacturing
Commentary by Dr. Valentin Fuster
2017;():V003T04A008. doi:10.1115/MSEC2017-2830.

Non-combustible aluminum composite panel is a new type of green building and decoration material with high security. However, during its manufacturing process, the incongruity of temperature cyclings between a series of air circulating tempering furnaces on the production line may cause a serious negative impact on the stability of product quality. In this paper, a model of the temperature control system of a tempering furnace was built at first by applying parameter identification technique to the off-line data of the furnace. Then, an approach based on online parameter identification and model predictive control was proposed to solve the dilemma that the specific temperature range of one single tempering furnace and the temperature cyclings coordination of multiple tempering furnaces can not be attained at the same time when using PID or On-Off control method. A method was presented to optimize the phase difference between the temperature cyclings of differents furnaces’ to lower the fluctuation of product quality. Finally, experiments are used to demonstrate the descent in fluctuation using the methods proposed in this paper.

Commentary by Dr. Valentin Fuster
2017;():V003T04A009. doi:10.1115/MSEC2017-2840.

It has been long overdue to revamp existing statistical process control (SPC) approaches in manufacturing since Water Shewhart first proposed the use of control charts in 1924. The combination development of big data, cloud computing, and manufacturing reshoring back to Untied States has opened up the opportunities to rethink implementation strategies of SPC for manufacturing. This paper first reviews the history of SPC development in traditional manufacturing environments and then contrasts it with the opportunities presented in big data era. Five SPC implementation approaches are proposed based on the opportunities identified.

Commentary by Dr. Valentin Fuster
2017;():V003T04A010. doi:10.1115/MSEC2017-2856.

A virtual prototype of the moving beam balancing system of a heavy-duty hydraulic press working under die forging function is built with Adams, AMESim and Simulink, and the balancing control process is analyzed using this prototype. The moving beam of the heavy-duty hydraulic press may tilt due to the eccentric load during the die forging processing, and thus affect the forging quality and the safety of the press. So it is necessary to research the beam balancing control process. Compared to the traditional methods based on simplified mathematical models, virtual prototype technology can obtain a co-simulation model, avoid tedious formula derivation and solving work, and save test time and cost. Based on the analysis of the working principle of balancing system, this paper establishes a dynamical model of the moving beam, a hydraulic circuit model of the single balancing system and a controller model using Adams, AMESim and Simulink, respectively. Then a virtual prototype is built using the three models via co-simulation interface files. The eccentric load signal is constructed in AMESim according to the variation of eccentric load during die forging process. By adjusting the controller parameters, the rapid balancing of the moving beam under eccentric load conditions is realized, and high precision of dynamic balancing and steady equilibrium is obtained. The simulation results show that the single balancing unit can achieve effective balancing of the moving beam, and the co-simulation analysis method based on the virtual prototype built with Adams, AMESim and Simulink is feasible in the research of the synchronous rectification of the moving beam. This work is a useful exploration in the research of synchronous rectification of moving beams.

Commentary by Dr. Valentin Fuster
2017;():V003T04A011. doi:10.1115/MSEC2017-2926.

In order to break the bottleneck of low efficiency, bad quality following drilling alloy Ti6Al4V, the effect of cutting parameters on thrust force, drilling vibration, burr height and surface roughness was studied based on response surface method. The optimized parameters were obtained. Results showed that feed rate had significant effect on thrust force and little on drilling vibration, while cutting speed had significant effect on vibration and little on thrust force. It is also observed that surface roughness decreased with cutting speed increasing, as well as increased with feed rate increasing. In addition, microstructure on the drilled hole surface showed mobility along feeding direction. Grain refinement on the drilling hole surface became serious with the increase of cutting speed and feed rate.

Commentary by Dr. Valentin Fuster
2017;():V003T04A012. doi:10.1115/MSEC2017-2937.

Various techniques are used to diagnose problems throughout all levels of the organization within the manufacturing industry. Often times, this root cause analysis is ad-hoc with no standard representation for artifacts or terminology (i.e., no standard representation for terms used in techniques such as fishbone diagrams, 5 why’s, etc.). Once a problem is diagnosed and alleviated, the results are discarded or stored locally as paper/digital text documents. When the same or similar problem reoccurs with different employees or in a different factory, the whole process has to be repeated without taking advantage of knowledge gained from previous problem(s) and corresponding solution(s). When discussing the diagnosis, personnel may miscommunicate over terms used in the root cause analysis leading to wasted time and errors. This paper presents a framework for a knowledge-based manufacturing diagnosis system that aims to alleviate these miscommunications. By learning from diagnosis methods used in manufacturing and in the medical community, this paper proposes a framework which integrates and formalizes root cause analysis by categorizing faults and failures that span multiple organizational levels. The proposed framework aims to enable manufacturing operations by leveraging machine learning and semantic technologies for the manufacturing system diagnosis. A use case for the manufacture of a bottle opener demonstrates the framework.

Topics: Manufacturing
Commentary by Dr. Valentin Fuster
2017;():V003T04A013. doi:10.1115/MSEC2017-2940.

Mixed-model assembly systems (MMASs) have been well recognized for their ability to handle product variants, and thus are particularly useful to meet the requirement brought by mass personalization. However, operational decision-making in MMASs is challenging due to the system complexity. Production selection and maintenance are two important operational decisions. In this paper, we investigate the joint production and maintenance policies in MMASs that consist of both common and variant operation stations. Markov Decision Process (MDP) is used to formulate the problem and numerical examples are presented to illustrate the structure of the policy in an MMAS that produces two types of product variants.

Commentary by Dr. Valentin Fuster
2017;():V003T04A014. doi:10.1115/MSEC2017-3065.

In condition-based maintenance, preventive replacement threshold and inspection scheme play important roles in maintenance performance. Major research considers cost as the main objective for measuring maintenance performance; here the average cost per unit time is used as the only objective in a single-unit system. The intention of this study was to investigate that how the average cost per unit time varies through changing replacement threshold and inspection scheme, the aim is to simultaneously determine an optimal replacement threshold and inspection scheme. The heterogeneity of hazard rate over stages of the equipment entails less frequent inspections when the equipment is at a healthy condition and more frequent inspections when the equipment is at a progressively deteriorated condition. Therefore the proposed condition-based inspection in the present work is non-periodic. A method based on dynamic programming is developed in order to implement a condition-based inspection scheme; furthermore the threshold for preventive replacement and the inspection scheme are simultaneously determined and then the optimal average maintenance cost is obtained.

Commentary by Dr. Valentin Fuster

Manufacturing Equipment and Systems: Advances in Data Management for the Digital Thread in Manufacturing

2017;():V003T04A015. doi:10.1115/MSEC2017-2790.

Manufacturing taxonomies and accompanying metadata of manufacturing processes have been catalogued in both reference books and databases on-line. However, such information remains in a form that is uninformative to the various stages of the product life cycle, including the design phase and manufacturing-related activities. This challenge lies in the varying nature in how the data is captured and represented. In this paper, we explore measures for comparing manufacturing data with the goal of developing a capability-based similarity metric for manufacturing processes. To judge the effectiveness of these metrics, we apply permutations of them to 26 manufacturing process models, such as blow molding, die casting, and milling, that were created based on the ASTM E3012-16 standard. Furthermore, we provide directions towards the development of an aggregate similarity metric considering multiple capability features. In the future, this work will contribute to a broad vision of a manufacturing process model repository by helping ease decision-making for engineering design and planning.

Topics: Manufacturing
Commentary by Dr. Valentin Fuster
2017;():V003T04A016. doi:10.1115/MSEC2017-2825.

The application of computer technology to engineering and manufacturing domains has drastically transformed the way products and systems are designed and produced. However, a major drawback of CAD/CAM/CAE systems is the steep learning curve required to understand and master their extensive and increasingly complex set of functionalities. In this paper, we present a new approach to deliver CAD training materials that is inspired by Model-Based Definition (MDB) strategies, where annotated 3D models become the center of the training process. In our system, textual 3D annotations are connected to a Product Lifecycle Management (PLM) system to provide access to interactive video tutorials which are linked to specific features of a CAD model. As a proof of concerto to validate this approach, a plugin for a commercial CAD package was developed that enhances the functionality of standard 3D annotation mechanisms and enables users to interact with the technical training materials directly within the CAD interface. New data structures were implemented to support the connection and integration with PLM systems. A group of tutorials are described to illustrate the system architecture and implementation details.

Commentary by Dr. Valentin Fuster
2017;():V003T04A017. doi:10.1115/MSEC2017-2979.

The Model-Based Enterprise (MBE) paradigm is being adopted by manufacturing companies in a variety of industries. Companies benefit from enhanced visualization, documentation, and communication capabilities when 3D annotated product definitions, or Model-Based Definitions (MBD) replace two-dimensional drawings throughout an enterprise. It is critical that product information, much of which is defined implicitly in drawings, is not lost in this transition. This presents a challenge to authors and translators of 3D models used through the product lifecycle. They must understand the semantics of the product information typically presented by a drawing then explicitly include this information, in a computer-interpretable form, in the MBD.

The research study described in this paper seeks to discover what is the minimum set of required information to carry out all the tasks in a given workflow of a model-based enterprise. A survey was conducted across various industry sectors to identify the foundational elements of this Minimum Information Model (MIM) in selected workflows. This study identified the information used within the specific workflows, the capabilities of 3D CAD models to carry this information, and the implications for doing so.

Commentary by Dr. Valentin Fuster

Manufacturing Equipment and Systems: Advances in Multi-Axis and Multi-Tasking Machine Tools

2017;():V003T04A018. doi:10.1115/MSEC2017-2665.

Although machine control data can be obtained by means of converting cutter location (CL) data comprised of the tool tip coordinate and the tool axis orientation vector in the workpiece coordinate frame with postprocessor, it’s uncertain whether they can be used for 5-axis machining. Owing to the fact that most postprocessors focus on the method to derive solutions for the equations of NC data by the form-shaping function matrix and the inverse kinematics model without taking the manufacturing scene into consideration, this study has presented a new post-processing system to generate and optimize NC data more effectively by correcting and selecting optimum solution intelligently for machining based on the solid model of machine tool in simulation environment. In general, the post-processing system consists of user interface layer, data access layer and data processing layer to give expression to the characteristics of universality, practicality and adaptability. User interface layer is mainly about loading the machine model and setting the relevant parameters. Data access layer includes model library of generalized five-axis machine tool configurations, rules library of cutter location data and NC data. Data processing layer is the major research in the paper, which illustrates how to correct the inverse solutions set and select the optimization solution for actual machining. The visual interface for post-processing system written by C++ was successfully applied in the experiment on a five-axis machine tool with a C-axis behind a B-axis rotary table, which demonstrated the effectiveness of the proposed post-processing methodology in the field of manufacturing.

Commentary by Dr. Valentin Fuster
2017;():V003T04A019. doi:10.1115/MSEC2017-2689.

The quality of light reflectance model mainly depends upon the correctness of cloud of points generated by the high-resolution charge coupled-device (CCD) camera. We have performed an experiment to capture the dimension of an object using Optigo 200 robot which provides an innovative way for solving dimensional control related challenges. The integrated system collects highly accurate and dense cloud of points for measuring an object at different design of experiment (DOE) levels. It also performs immediate analysis of collected data and calculates the deviations from the given specifications. Thus we tried to visualize the object using its key parameters. In this paper, we describe the functional relationship of real-world surfaces with the dependence of light exposure and camera direction. Seven step wedge (as a measurement object) is used in our case study to carry out the stereovision based experiment. Finally, a comparative analysis shows model accuracy over conventional measurement techniques.

Topics: Gages , Reflectance , Modeling
Commentary by Dr. Valentin Fuster
2017;():V003T04A020. doi:10.1115/MSEC2017-2711.

Energy consumption of numerical control (NC) machine tools is one of the key issues in modern industrial field. This study focuses on reducing the energy consumed by a five-axis machining center by changing only the workpiece-setting position. Previous studies show that the movements along each axis in five-axis machining centers depend on the workpiece-setting position, regardless of whether the same operation is performed. In addition, the energy consumptions required for the movements are different along each axis.

From these considerations, an optimum workpiece-setting position that can minimize the energy consumed during these motions is assumed to exist. To verify this assumption, in this study, the energy consumed by the feed drive systems of an actual five-axis machining center is first measured and then estimated using the proposed model in this study. The model for estimating the energy consumption comprises the friction, motor, and amplifier losses along each axis. The total energy consumption can be estimated by adding the energy consumptions along each axis.

The effect of the workpiece setting-position on the energy consumption is investigated by employing the cone-frustum cutting motion with simultaneous five-axis motions. The energy consumption that depends on the workpiece-setting position is first measured and then estimated. The results confirm that the proposed model can estimate the energy consumption accurately. Moreover, the energy consumption is confirmed to depend on the workpiece-setting position; the minimum energy consumption is found to be 20% lower than the maximum one.

Commentary by Dr. Valentin Fuster
2017;():V003T04A021. doi:10.1115/MSEC2017-2777.

A new cutting force simulator has been developed to predict cutting force in ball end milling. This new simulator discretely calculates uncut chip thickness based on a fully voxel representation of the cutting edge and instantaneous workpiece shape.

Previously, a workpiece voxel model was used to calculate uncut chip thickness under a complex change of workpiece shape. Using a workpiece voxel model, uncut chip thickness is detected by extracting the voxels removed per cutting edge tooth for the amount of material fed into the cutting edge. However, it is difficult to define the complicated shape of a cutting edge using the workpiece voxel model; the shape of the cutting edge must be defined by a mathematical expression. It is also difficult to model the voxels removed by the cutting edge when the tool posture is non-uniformly changed.

We therefore propose a new method to detect uncut chip thickness, one in which both the cutting edge and the instantaneous workpiece shape are fully represented by a voxel model. Our proposed method precisely detects uncut chip thickness at minute tool rotational angles, making it possible to detect the uncut chip thickness between the complex surface shape of the workpiece and the particular shape of the cutting edge.

To validate the effectiveness of our proposed method, experimental 5-axis milling tests using a ball end mill were conducted. Estimated milling forces for several tool postures were found to be in good agreement with the measured milling forces. Results from the experimental 5-axis milling validate the effectiveness of our proposed method.

Topics: Cutting , Milling , Shapes
Commentary by Dr. Valentin Fuster
2017;():V003T04A022. doi:10.1115/MSEC2017-2958.

In driven rotary cutting of stainless steel, adhesions sometimes occur on the tool, causing increased wear. The type of coolant supplying methods and tool rotation speed affects largely on the adhesion because it depends on the temperature and lubricating performance. Results showed that in a circumferential velocity ratio of 1.0, which means tangential component of tool peripheral speed is equal to work surface speed, there is no adhesion on the tool after cutting. In a circumferential velocity ratio of 2.0, adhesion occurred with overcooling of the flood coolant, and wear increased by adhesions to the rotating tool. It was found that the thermal cracks on the cutting edge was one of the factors of increased wear and chipping. Adhesives on the tool edge also accelerated the chipping.

Commentary by Dr. Valentin Fuster
2017;():V003T04A023. doi:10.1115/MSEC2017-3030.

Drillings and millings of cylinders are performed in operations on turning centers. Then, the drill holes are machined on the cylinder surfaces with offsets of the cutting positions from the workpiece axis. The paper presents an analytical model to predict the cutting force in drilling of cylinders. Three dimensional chip flow is interpreted as a piling of the orthogonal cuttings in the planes containing the cutting velocities and the chip flow velocities. In drilling, the chip flow models are made on the chisel and the lips. The chip flow directions are determined to minimize the cutting energies on the chisel and the lips. The workpiece model on the cylinder is defined by the cutting point with respect to the cylinder shape. The cutting force in drilling of a cylinder is compared with that in drilling of a plate. Then, the effect of the offset cutting points on the cutting force is discussed in terms of the resultant cutting force, which induces the machining error and the tool breakage.

Commentary by Dr. Valentin Fuster

Manufacturing Equipment and Systems: Cloud Manufacturing

2017;():V003T04A024. doi:10.1115/MSEC2017-2639.

Product configurator, as an effective tool in mapping customer requirements with company’s existing product attributes, enables customers’ satisfaction and companies’ competitiveness in a cost-efficient way. However, with the tendency towards mass personalization, customers are not only just selecting from each company’s own options in a ‘configure-to-order’ model, but also more actively involved in the product development process to create their own individualized products in an ‘engineer-to-order’ model. Besides, the existing configurators generally apply the same matching procedures to all the customers in the same sequential way, which is tedious and time consuming, especially for the complicated product configuration. Aiming to solve these problems, this paper proposes a personalized product configuration process to determine design attributes in a cloud-based environment, which is based on two assumptions: 1) products need to be adaptable enough for configuration; 2) customers prefer to develop new designs from the existing products in a tangible or visualized way other than design from scratch. The proposed process is capable of handling personalized requirements by adding new modules or upgrading design attributes in the existing product family. An illustrative example shows its advantages in customer-centric product development process.

Commentary by Dr. Valentin Fuster
2017;():V003T04A025. doi:10.1115/MSEC2017-2656.

Nowadays, the key goal in manufacturing is being very efficient within changing markets and under turbulent conditions. Therefore, production plants with their machines logistics and all the other involved components have to be adaptable to changing conditions.

For this reason, reconfigurable manufacturing systems are needed, which allow a fast adaption to new requirements of the product to be manufactured. Today, reconfiguration in manufacturing is mostly limited due to missing reconfigurability of the control software in combination with the underlying hardware. The coupling is that strong that in manufacturing control software is always bound to special hardware. Until now, flexibility is only possible by changing application or part programs that are interpreted by a fixed control kernel. The adaption of any core functionality is impossible, and any other changes require high manual effort for redesigning software systems and parametrizing their functionalities. For better adaptability in manufacturing this coupling has to be dissolved.

Other disciplines and industries have similar requirements like the information and communication technology (ICT). In the area of ICT, there are more and more concepts of Software Defined Anything (SDX) like Software Defined Networking (SDN) or Software Defined Radio (SDR). Flexible, adaptive and really reconfigurable manufacturing should be improved by a new concept of Software Defined Manufacturing (SDM). SDM allows freely defined functionalities within the physical limitations of the mechanical and electrical components of a machine. But current manufacturing equipment with its control architecture does not offer the technical basis for such a concept.

Existing concepts of cloud-based control architectures show indeed a virtualization of the control algorithms. Due to the fact that the software is running remotely, the software is decoupled from its hardware. However, the local control algorithms with hard real-time requirements still have a very strong coupling with the hardware. The local control software could not be defined freely according to the requirements of the product to be manufactured. In this paper, a new control architecture for manufacturing that combines cloud-based control as a service (CaaS) and Software Defined Manufacturing is presented. As a result, an architecture of an operating system for manufacturing equipment is shown, which is freely programmable. This paper deals with Software Defined Manufacturing for local control software, communication and cloud-based control systems. SDM allows defining the behavior of the entire manufacturing process based on design description of a product to be manufactured. In addition, methods are described, which allow the automatic configuration and optimization of such an architecture by using simulation technics and collected process data.

Commentary by Dr. Valentin Fuster
2017;():V003T04A026. doi:10.1115/MSEC2017-2703.

Service-oriented robotic manufacturing system is an integrated system, in which the industrial robots (IRs) operate within a service-oriented manufacturing model, and can be virtualized and servicelized as services, so as to provide on-demand, agile, configurable and sustainable manufacturing capability services to users in workshop environment. Manufacturing capability of such systems can be divided into three layers, including manufacturing cell layer, production process layer and workshop layer. However, most of existing works carried out the optimization on each layer individually. Manufacturing cells are the component parts of a production process, and there are close relationships between them and can effect the operation and performance for each other, therefore it is essential to jointly consider the manufacturing capability service optimization on both layers. In this context, a cross-layer optimization model is proposed to conquer the existing limitation and provide a comprehensive performance assurance to service-oriented robotic manufacturing systems. The proposed model has different decision-making mechanisms on each layer and the communications and interaction between the two layers can further coordinate the optimizations. A case study based on robotic assembly is implemented to demonstrate the availability and effectiveness of the proposed model.

Commentary by Dr. Valentin Fuster
2017;():V003T04A027. doi:10.1115/MSEC2017-2704.

With the development of information and computer network technology, cloud manufacturing has been developing rapidly, industrial robots (IRs) as a vital symbol and an advanced technology of manufacturing industry, in scheduling service, the constantly changing information data will result in the corresponding vary of the manufacturing capability. Under a fixed constraint of some capability service request, this will decrease the number of the optimal solutions and provide the inaccurate service to users. So it is important to make the manufacturing capability stable and obtain more optimal solutions to satisfy the constraint, thus the dynamic assessment of manufacturing capability based on information feedback is investigated in this paper. A set of indicators is established considering the IRs’ manufacturing capability and a new dynamic assessment model is proposed to achieve the actual data and the expected data information feedback, using the “normal distribution” model, which can correct the assessment weight. By the way, a case study is simulated in the MATLAB, which shows the reliability and reasonability of this method in evaluate the manufacturing capability in IR.

Commentary by Dr. Valentin Fuster
2017;():V003T04A028. doi:10.1115/MSEC2017-2705.

Cloud manufacturing (CMfg) aims to realize the full-scale sharing, free circulation and transaction, and on-demand use of various manufacturing resources and capabilities in the form of manufacturing services. During the whole product life-cycle, the number of manufacturing services is huge, and services are highly dynamic and changeful. Without the effective operation and technical support of manufacturing service management, the implementation and aim of CMfg could not be achieved. In this paper, a multi-layer model of manufacturing service is proposed for a job shop in cloud manufacturing, in order to solve the description problem of different manufacturing services from different level view, e.g. machine level, process level and shop level. Consequently, a hypergraph-based network model of manufacturing service is developed, so as to facilitate the management of different services during the whole production process in job shop. A case study and some applications of the proposed model for supporting the manufacturing services management to practical manufacturing system are studied, to demonstrate the feasibility and efficiency of such model.

Commentary by Dr. Valentin Fuster
2017;():V003T04A029. doi:10.1115/MSEC2017-2708.

Additive manufacturing (AM) has experienced a phenomenal expansion in recent years and new technologies and materials rapidly emerge in the market. Design for Additive Manufacturing (DfAM) becomes more and more important to take full advantage of the capabilities provided by AM. However, most people still have limited knowledge to make informed decisions in the design stage. Therefore, an interactive DfAM system in the cloud platform is proposed to enable people sharing the knowledge in this field and guide the designers to utilize AM efficiently. There are two major modules in the system, decision support module and knowledge management module. A case study is presented to illustrate how this system can help the designers understand the capabilities of AM processes and make rational decisions.

Commentary by Dr. Valentin Fuster
2017;():V003T04A030. doi:10.1115/MSEC2017-2719.

Cloud manufacturing is a new service-oriented networked manufacturing mode based on the concept of “Manufacture as a Service” and achieves the sharing of manufacturing resources and manufacturing capacity. Multi-tenancy technology can improve utilization efficiency of manufacturing resources and ensure information security among tenants, enabling users to share the cloud manufacturing resources better. To execute this new mode, isolation access and on-demand services are indispensable. However, the traditional access control model cannot satisfy the demands of multi-tenant environment on cloud manufacturing platform. To solve the demands in such an environment, a model named Multi-Tenant Access Control Model for Cloud Manufacturing (CM-MTAC) is proposed. Based on cloud manufacturing architecture, we build a hierarchical cloud manufacturing access control architecture combining multi-tenancy. Considering the demands under this condition, the elements of cloud manufacturing access control model and the relationships between them are redefined by extending the ABAC model. Then multi-tenancy authorization framework is proposed and XACML language is used to describe the policy to provide our model with on-demand service, isolation access and inter-tenant collaboration. Finally, we develop this model into the cloud manufacturing monitoring platform. Results show that our model, compared with traditional models, has a better performance of on-demand service, isolation access and inter-tenant cooperation under the environment of cloud manufacturing.

Topics: Manufacturing
Commentary by Dr. Valentin Fuster
2017;():V003T04A031. doi:10.1115/MSEC2017-2720.

Cloud Manufacturing is a new model to increase the manufacturing and business benefits by sharing manufacturing resources. These resources can bring users convenience, but also may be maliciously analyzed by the attacker which may result in personal or corporate privacy disclosure. In this paper, we discuss the privacy disclosure problem in cloud manufacturing, and propose a method for releasing order data securely with the complex relationship between enterprises and other vendors. With regards to the risk of privacy leakage in the process of data analysis or data mining, we improve the traditional method of anonymous releasing for original order data, and introduce the thought of safe k-anonymization to achieve the process. To meet the needs of protecting sensitive information in data, we analyze the users’ different demands for order data in the cloud manufacturing, use the sampling function to satisfy (β, ε, δ) - DPS to increase the uncertainty of the differential privacy, improve the k-anonymization method, apply the anonymous method with generalization, concealment, and reduce data associations to different attributes. The improved method not only preserves the statistical characteristics of the data, but also protects the privacy information in the order data in the cloud manufacturing environment.

Topics: Manufacturing
Commentary by Dr. Valentin Fuster
2017;():V003T04A032. doi:10.1115/MSEC2017-2747.

By enabling consumer products to be made on-demand and eliminating waste from overproduction and transport, online 3D printing service is more and more popular with unprofessional customers. As a growing number of 3D printers are becoming accessible on various online 3D printing service platforms, there raises the concern over online 3D printing service evaluation and selection for novices as well as users with 3D printing experience. In this paper, we analyze this problem using information transformation techniques and multinomial distribution probabilistic model. Evaluation factors, the major attributes that significantly affect the performance of an online 3D printing service, are described with standard description form. Meanwhile, historical service data is introduced to identify and update these evaluation factor values. Based on these parameters, evaluation and comparison can be implemented upon online 3D printing services using the probabilistic model. An example is presented to illustrate the assessment process based on the proposed evaluation model. The presented objective probabilistic evaluation method can serve as the basis of online 3D printing service evaluation and selection on an online 3D printing service platform. Although the focus of the work was on 3D printing service, the idea can be applied to other online rapid prototyping sharing systems.

Commentary by Dr. Valentin Fuster
2017;():V003T04A033. doi:10.1115/MSEC2017-2752.

To evaluate the effects of customers’ participation levels in various business activities on pricing in service-oriented manufacturing, the indices of pricing are proposed through extracting the influential factors in the four stages (i.e., design, manufacturing, production and services) from the whole value chain to comprehensively reflect customers’ demands. A new pricing model based on these indices is formulated by Support Vector Machine (SVM). It can predict a more accurate product price regarding the products’ similarity by the values of the influential factors that are determined in terms of business activities participated by customers. Finally, a case study from a molding company in China is conducted to verify the effectiveness of this pricing methodology. The results indicate that the model by SVM fares better in comparison with that by Back Propagation Neural Networks in small scale samples, especially in the performances of generalization and robustness. The outcomes also testify that this price prediction methodology can increase the accuracy of a product’s price as well as the customer’s satisfaction.

Commentary by Dr. Valentin Fuster
2017;():V003T04A034. doi:10.1115/MSEC2017-2807.

In a service-oriented networked manufacturing (SONM) environment, geographically distributed manufacturing resources are encapsulated as various manufacturing services. These manufacturing services release via the Internet and can provide services on the demand of manufacturing tasks. Usually one manufacturing task needs several different services belonged to different organizers to work together. Hence, effective cooperation among services is the foundation of the efficient operation of SONM. In this paper, a bipartite network model is presented to describe the relationship of two different kinds of nodes in SONM, and also is projected as a weighed network for further exploring the behaviors of service nodes. Furthermore, an agent-based model is built for modeling the interactive behaviors of service nodes in a cooperative network and an agent-based simulating system is developed with Repast. The simulation results show that the emergence of cooperative behaviors among service nodes is related to both the cost of cooperation and initial trust of services in the SONM environment.

Commentary by Dr. Valentin Fuster
2017;():V003T04A035. doi:10.1115/MSEC2017-2815.

An industrial park is a cluster of enterprises who locate in the same location to share common infrastructures with governmental or privately finance support, which plays an important role in promoting regional economic and industrial development. However, demand volatilities push manufacturing enterprises in a dilemma that excessive resource configuration leads to waste of resource and high production cost when customer demands is low, or lack of resource and capacity for satisfying customer demands when customer demands is high. Therefore, how to temporarily configure dynamic resource for satisfying customer demand with high quality and quick delivery at low cost is a burning issue. To solve above-named issue, combining advantages of both Cloud Manufacturing (CMfg) and product service system (PSS), this paper design CMfg-enabled production logistics service system (CMPLSS) to handle demand volatilities together by sharing resource as service in industrial park. The use of CMPLSS leads to benefits for enterprises in industrial park.

Commentary by Dr. Valentin Fuster
2017;():V003T04A036. doi:10.1115/MSEC2017-2817.

Cloud manufacturing is a novel service-oriented networked manufacturing paradigm for the manufacturing industry. Through aggregating distributed manufacturing resources from different enterprises and transforming them into services, cloud manufacturing is able to provide on-demand manufacturing services to customers. Scheduling, including resource scheduling and task scheduling, is a critical instrument for achieving on-demand service provisioning, and also an important research issue in cloud manufacturing. In the process of service scheduling in cloud manufacturing, the manufacturing services are firstly matched according to the service demander’s functional requirements and service availability to form the candidate service sets. And then the optimized service scheduling scheme is generated according to the service demander’s non-functional requirements. The individual requirements of service demanders are analyzed from aspects of functional and non-functional requirements in this paper. On this basis, the scheduling process for individual requirements in cloud manufacturing system is studied and a cloud manufacturing service scheduling method is proposed. This work can provide support and foundation for the related research of task planning and scheduling in cloud manufacturing system. Finally, a case study is given to verify the proposed service scheduling method.

Topics: Manufacturing
Commentary by Dr. Valentin Fuster
2017;():V003T04A037. doi:10.1115/MSEC2017-2827.

Disruptions happen in the actual manufacturing system under environment of the Internet of Things and they make the system difficult to manage. However, the convenient access to information of orders, equipment and participants make disruption recovery easier. In this paper we build a disruption recovery scheduling integer programming model considering the objective of minimizing total weighted completion-time (as the original objective) and the objectives of maximizing total consumer satisfaction degree & minimizing total deviation degree (as the revising objective). A PVPS (PSO & VNS Parallel Search) algorithm is proposed. The experiments results prove all above are effective.

Commentary by Dr. Valentin Fuster
2017;():V003T04A038. doi:10.1115/MSEC2017-2839.

Cloud manufacturing (CMfg) mode provides an effective means to intensely utilize distributed resources and manufacturing capability for personalized production. Increasing personalized customization implies more and more heterogeneous tasks and hence more sorts of requirements for services. As the granularity of tasks vary with changing users and products, the solution (or scheme) of task scheduling should be different. In order to efficiently provide the most suitable solution for each kind of tasks, different scheduling ways should be adopted under different circumstances. In this paper, we study scheduling issues for heterogeneous tasks with variable granularity and present two kinds of optimal scheduling mode based on user-oriented comprehensive evaluation. Then different encoding schemes relied on the genetic algorithm are proposed according to different scheduling strategies.

Commentary by Dr. Valentin Fuster
2017;():V003T04A039. doi:10.1115/MSEC2017-2889.

To facilitate the vision of service-oriented manufacturing, Cloud Manufacturing (CMfg) will need to support business process model life-cycle management. In this paper, we propose the Business Process Cataloging and Classification System (BPCCS) to support that role. BPCCS can facilitate adaptation of business process models. We validate life-cycle management requirements for such a system and propose capabilities the system must have to address these requirements. We analyze related work in academia and industry as a basis for synthesizing a meta-model and a conceptual architecture for the system. We conclude that contextual information for business process models is a critical part of such a system and where the CMfg community can contribute new and valuable results.

Commentary by Dr. Valentin Fuster
2017;():V003T04A040. doi:10.1115/MSEC2017-2896.

Smart factory and smart production are the common aims for many countries’ manufacturing development strategies, which have attracted attentions from both academia and industry. Smart production line, which is a basic unit for implementing smart factory and smart production, is emphasized in this paper. The evolution process of production line is summarized first. Then the common factors for smart production line are investigated. Accordingly, a data-driven method for realizing smart production line is proposed. At the same time, the corresponding applications and attainable goals are given. Finally, a case for data-driven energy emulation and analysis in production line is illustrated.

Topics: Assembly lines
Commentary by Dr. Valentin Fuster
2017;():V003T04A041. doi:10.1115/MSEC2017-2904.

Cloud manufacturing simulation platform is used to simulate the collaboration and evolution, which among the resources, services, tasks, participants in cloud manufacturing environment. As an important part of the platform of simulation, simulation data generation method can effectively support the simulation accuracy. Data in cloud manufacturing environment are not completely random, and are closely related to the actual environment and resource characteristics. The workload of traditional random generate method or artificial method is very heavy and cannot completely rebuild the simulation environment. In this paper, Using clustering method to extract characteristics from an actual environment, and then extend the characteristics to generate new simulation data. To build a similar environment to the real environment used in the simulation. The result is shown that compared with the method to generate random data. This method can generate the reference data similar environment, the simulation can reflect the real effect in the process.

Commentary by Dr. Valentin Fuster
2017;():V003T04A042. doi:10.1115/MSEC2017-2952.

Cloud manufacturing is a new manufacturing paradigm in which manufacturing resources and capabilities offered by different enterprises are provided as cloud-based services over the Internet. In addition to the cloud manufacturing platform, enterprises as resource providers are also essential part of a cloud manufacturing system as customer orders that are submitted to the cloud platform will ultimately need to be dispatched to enterprises’ production sites for execution. So far, however, issues concerning enterprises in cloud manufacturing has attracted little attention of researchers, and the most frequently mentioned issue in existing research that are concerned with enterprises is how to connect enterprises’ resources to the cloud infrastructure. This, to a large extent, hinders the development and implementation of cloud manufacturing as the lack of research on enterprises fails to uncover the requirements for enterprises in cloud manufacturing (i.e. Cloud Manufacturing Enterprises, CMEs), and thus enterprises have no reference for evaluating the changes that need to be made to adopt this new manufacturing paradigm. This paper focuses on this important issue and conducts a preliminary exploration on CMEs. We first analyze the requirements for CMEs, and then discuss some critical issues with CMEs in detail, including enterprise information systems, enterprise architecture, and enterprise modeling.

Topics: Manufacturing
Commentary by Dr. Valentin Fuster
2017;():V003T04A043. doi:10.1115/MSEC2017-2987.

Rapid responsiveness to diverse customer needs is considered a competitive advantage in manufacturing business. To shrink the inquiry-to-order process, manufacturing firms will benefit a lot from building a product configuration system (PCS) which is the enabler of mass customisation (MC). PCS has matured in consumer businesses for decades but in capital goods industries, typically operating in engineer-to-order (ETO) manner, things differ a lot. It is for the reason that conventional PCS is incapable of extending customisation from order-delivery processes to the design/engineering phase. Cloud manufacturing, which is an emerging service-oriented manufacturing paradigms enabled by cyber-physical system, the Internet of Things and the Internet of Service, is promising to break the bottleneck of “ETO PCS” by the provision of technical infrastructure for product, service and data customisation. With the introducing of manufacturing-as-a-service (MaaS) concept, a product family is extended to a product-service family (PSF) in this paper for implementing in-depth product configuration process with scalable customisation depth (i.e., the degree of customisation freedom). Additionally, an approach of service delegation in product configuration process is proposed to support customer-centric product customisation. At last, the methodology proposed in this paper is validated by a case study in which the product configuration process of a complex ETO product is performed.

Topics: Manufacturing
Commentary by Dr. Valentin Fuster
2017;():V003T04A044. doi:10.1115/MSEC2017-3003.

Personal thermal comfort is a crucial yet often over-simplified factor in building climate control. Traditional comfort models lack the adaptability to fit individuals’ demand. Recent advances of machine learning and ubiquitous sensor networks enable the data-driven approach of thermal comfort. In this paper, we built a platform that can simulate occupants with different thermal sensations and used it to examine the performance of support vector machine (SVM) and compared with several other popular machine learning algorithms on thermal comfort prediction. We also proposed a hybrid SVM-LDA thermal comfort classifier that can improve the efficiency of model training.

Commentary by Dr. Valentin Fuster
2017;():V003T04A045. doi:10.1115/MSEC2017-3006.

Today, in the era of modern Intelligent Production Environments (IPE) and Industry 4.0, the manufacturing of a product takes place in various partial steps and these mostly in different locations, potentially distributed all over the world. The producing companies must assert in the global market and always find new ways to cut costs by saving tax, changing to the best providers, and by using the most efficient and fastest production processes. Furthermore, they must be inevitably based on a cloud-based repository and distributed architectures to make data and information accessible everywhere as well as development processes and knowledge available for a worldwide cooperation. A so called Collaborative Adaptive (Production) Process Planning (CAPP) can be supported by semantic approaches for knowledge representation and management as well as knowledge sharing, access, and re-use in a flexible and efficient way. In this way, to support CAPP scenarios, semantic representations of such knowledge integrated into a machine-readable process formalization is a key enabling factor for sharing in cloud-based knowledge repositories. This is especially required for, e.g., Small and Medium Enterprises (SMEs). When SMEs work together on a production planning for a joint product, they exchange component production and manufacturing change information between different planning subsystems. These exchanges are mostly based on the already well-established Standard for the Exchange of Product model data (STEP), not least to obtain a computer-interpretable representation. Moreover, so-called Function Block (FB) Domain Models could support these planning process. FBs serve as a high-level planning-process knowledge-resource template and to the representation of knowledge. Furthermore, methodologies are required, which based on process-oriented semantic knowledge-representation, such as Process-oriented Knowledge-based Innovation Management (German: Wissens-basiertes Prozess-orientiertes Innovations Management, WPIM). WPIM is already a web- and cloud-based tool suites and can represent such planning processes and their knowledge resources and can therefore be used to support the integration and the management of distributed CAPP knowledge in Manufacturing Change Management (MCM), as well as its access and re-use. That is also valid for Assembly-, Logistics- and Layout Planning (ALLP). On the one hand, a collaborative planning in a machine-readable and integrated representation will be possible as well as an optimization for mass production. On the other hand, within a cloud-based semantic knowledge repository, that knowledge can be shared with all partners and contributors. To combine all these functionalities, in 2016 we have already introduced a method, called Knowledge-based Production Planning (KPP). We outlined the theoretical advantages of integrating CAPP with Collaborative Manufacturing Change Management (CMCM) in the last year at MSEC16. In this Paper, we will demonstrate our first implementations of the KPP application with an integrated visual direct manipulative process editor as well as a first prototype of our mediator architecture with a semantic integration including a query library based on the KPP ontology.

Commentary by Dr. Valentin Fuster
2017;():V003T04A046. doi:10.1115/MSEC2017-3012.

The amount of data that can be gathered from a machining process is often misunderstood, and even if these data are collected, they are frequently underutilized. Intelligent uses of data collected from a manufacturing operation can lead to increased productivity and lower costs. While some large-scale manufacturers have developed custom solutions for data collection from their machine tools, small- and medium-size enterprises need efficient and easily deployable methods for data collection and analysis. This paper presents three broad solutions to data collection from machine tools, all of which rely on the open-source and royalty-free MTConnect protocol: the first is a machine monitoring dashboard based on Microsoft Excel; the second is an open source solution using Python and MTConnect; and the third is a cloud-based system using Google Sheets. Time studies are performed on these systems to determine their capability to gather near real-time data from a machining process.

Commentary by Dr. Valentin Fuster
2017;():V003T04A047. doi:10.1115/MSEC2017-3038.

In recent years, Cloud manufacturing has become a new research trend in manufacturing systems leading to the next generation of production paradigm. However, the interoperability issue still requires more research due to the heterogeneous environment caused by multiple Cloud services and applications developed in different platforms and languages. Therefore, this research aims to combat the interoperability issue in Cloud Manufacturing System. During implementation, the industrial users, especially Small- and Medium-sized Enterprises (SMEs), are normally short of budget for hardware and software investment due to financial stresses, but they are facing multiple challenges required by customers at the same time including security requirements, safety regulations. Therefore in this research work, the proposed Cloud manufacturing system is specifically tailored for SMEs.

Topics: Manufacturing
Commentary by Dr. Valentin Fuster

Manufacturing Equipment and Systems: High Performance Computing and Artificial Intelligence for CyberManufacturing

2017;():V003T04A048. doi:10.1115/MSEC2017-2679.

Manufacturers have faced an increasing need for the development of predictive models that help predict mechanical failures and remaining useful life of a manufacturing system or its system components. Model-based or physics-based prognostics develops mathematical models based on physical laws or probability distributions, while an in-depth physical understanding of system behaviors is required. In practice, however, some of the distributional assumptions do not hold true. To overcome the limitations of model-based prognostics, data-driven methods have been increasingly applied to machinery prognostics and maintenance management, transforming legacy manufacturing systems into smart manufacturing systems with artificial intelligence. While earlier work demonstrated the effectiveness of data-driven approaches, most of these methods applied to prognostics and health management (PHM) in manufacturing are based on artificial neural networks (ANNs) and support vector regression (SVR). With the rapid advancement in artificial intelligence, various machine learning algorithms have been developed and widely applied in many engineering fields. The objective of this research is to explore the ability of random forests (RFs) to predict tool wear in milling operations. The performance of ANNs, SVR, and RFs are compared using an experimental dataset. The experimental results have shown that RFs can generate more accurate predictions than ANNs and SVR in this experiment.

Topics: Wear
Commentary by Dr. Valentin Fuster
2017;():V003T04A049. doi:10.1115/MSEC2017-2783.

Ontologies serve robotics in many ways, particularly in describing and driving autonomous functions. These functions are built around robot tasks. In this paper, we introduce the IEEE Robot Task Representation Study Group, including its work plan, initial development efforts, and proposed use cases. This effort aims to develop a standard that provides a comprehensive ontology encompassing robot task structures and reasoning across robotic domains, addressing both the relationships between tasks and platforms and the relationships between tasks and users. Its goal is to develop a knowledge representation that addresses task structure, with decomposition into subclasses, categories, and/or relations. It includes attributes, both common across tasks and specific to particular tasks and task types.

Topics: Robots , Ontologies
Commentary by Dr. Valentin Fuster
2017;():V003T04A050. doi:10.1115/MSEC2017-3069.

In a fully automated manufacturing system, tool condition monitoring system is essential to detect the failure in advance and minimize the manufacturing loses with the increase in productivity. To look for a reliable, simple and cheap solution, this paper proposes a new tool wear monitoring model to detect the tool wear progression and early detection of tool failure in end milling using audible sound signals. In this study, cutting tools are classified into six classes based on different flank wear ranges. A series of end milling experiments are operated with a broad range of cutting conditions for each class to collect sound signals. A machine learning algorithm that incorporates support vector machine (SVM) approach coupled with the application of time and frequency domain analysis is developed to correlate observed sound signals’ signatures to tool wear conditions. The performance evaluation results of the proposed algorithm have shown accurate predictions in detecting tool wear conditions from the sound signals. In addition, the proposed machine learning approach has shown the fastest response rate, which provides the good solution for on-line cutting tool monitoring.

Commentary by Dr. Valentin Fuster

Manufacturing Equipment and Systems: Innovations in Equipment Design, Tooling, and Control/Automation to Enhance Manufacturing Processes

2017;():V003T04A051. doi:10.1115/MSEC2017-2611.

With global megatrends like automation and digitization changing societies, economies, and ultimately businesses, shift is underway, disrupting current business plans and entire industries. Business actors have accordingly developed an instinctive fear of economic decline and realized the necessity of taking adequate measures to keep up with the times. Increasingly, organizations find themselves in an evolve-or-die race with their success depending on their capability of recognizing the requirements for serving a specific market and adopting those requirements accurately into their own structure. In the transportation and logistics sector, emerging technological and information challenges are reflected in fierce competition from within and outside. Especially, processes and supporting information systems are put to the test when technological innovation start to spread among an increasing number of actors and promise higher performance or lower cost. As to warehousing, technological innovation continuously finds its way into the premises of the heterogeneous warehouse operators, leading to modifications and process improvements. Such innovation can be at the side of the hardware equipment or in the form of new software solutions. Particularly, the fourth industrial revolution is globally underway. Same applies to Future Internet technologies, a European term for innovative software technologies and the research upon them. On the one hand, new hardware solutions using robotics, cyber-physical systems and sensors, and advanced materials are constantly put to widespread use. On the other one, software solutions based on intensified digitization including new and more heterogeneous sources of information, higher volumes of data, and increasing processing speed are also becoming an integral part of popular information systems for warehouses, particularly for warehouse management systems.

With a rapidly and dynamically changing environment and new legal and business requirements towards processes in the warehouses and supporting information systems, new performance levels in terms of quality and cost of service are to be obtained. For this purpose, new expectations of the functionality of warehouse management systems need to be derived. While introducing wholly new solutions is one option, retrofitting and adapting existing systems to the new requirements is another one. The warehouse management systems will need to deal with more types of data from new and heterogeneous data sources. Also, it will need to connect to innovative machines and represent their respective operating principles. In both scenarios, systems need to satisfy the demand for new features in order to remain capable of processing information and acting and, thereby, to optimize logistics processes in real time.

By taking a closer look at an industrial use case of a warehouse management system, opportunities of incorporating such new requirements are presented as the system adapts to new data types, increased processing speed, and new machines and equipment used in the warehouse. Eventually, the present paper proves the adaptability of existing warehouse management systems to the requirements of the new digital world, and viable methods to adopt the necessary renovation processes.

Topics: Warehouses
Commentary by Dr. Valentin Fuster
2017;():V003T04A052. doi:10.1115/MSEC2017-2699.

To improve the compressive capacity and straightening efficiency of straightening process, the stroke-deflection model (SDM) using dual indenter and dual clamp system (DIDCS) for linear guide rails is developed in this paper. The DIDCS is actually simplified as a symmetrically supported beam with two symmetrical concentrated forces on the top surface of workpiece, so the straightening process is regarded as pure bending process. To explore the deflection variation during the whole process with DIDCS, the curvature-deflection model (CDM) considering the span control of dual indenter and dual clamp is firstly analyzed based on elastic-plastic deformation theory and small deformation principle. The geometrical features and material properties of linear guide rails, which are the main factors influencing bending characteristics, are then mathematically modeled for the further analysis of stress and strain distributions in straightening process. Besides, to obtain the actual bending moment model (BMM) of different model parameters, the distribution regulations of elastic and plastic regions are analyzed followed by pure bending assumptions. The bending rebound model (BRM) is established with bending moment, geometrical features and material properties, and the SDM is finally calculated by initial deflection, rebound deflection and span parameters of the DIDCS. On basis of the DIDCS, the straightening process is simulated with the established finite element analysis model (FEM) to demonstrate the longitudinal stress distribution and the reflection of different straightening stages. The proposed SDM is also experimentally validated on the ROSE-JZ50 straightening machine with different materials.

Commentary by Dr. Valentin Fuster
2017;():V003T04A053. doi:10.1115/MSEC2017-2702.

Given the chord length of equal angle isn’t equal on elliptical section of piston skirt with middle-bulged varying ellipse (PSMVE), a larger theoretical processing error will inevitably be introduced with the interpolation algorithm of equal angle for PSMVE. To improve the manufacturing precision of PSMVE, a novel interpolation algorithm of equal-length-chord and spiral-line (IAES) and a method of tool radius compensation are presented based on symbolic computation. Firstly, a three-dimensional model of PSMVE is generated through the symbolic computation method, meanwhile, the coordinate values of arbitrary cutter-contact point can be expressed accurately. Secondly, the turning spiral trajectory is generated via updated cutter-contact point which can be searched from the obtained cutter-contact point with equal length chord. Besides, this paper proposes a method of tool radius compensation and obtains the cutter location points through appropriate transformation of coordinates. Last, some simulation, which mainly includes the establishment of 3D model, the generation of spiral trajectory with equal-length-chord, the transformation between cutter-contact point and cutter-location points, is carried out. In addition, this paper takes CNC turning center (System: SINUMERIK 802C) as an example to complete the processing of PSMVE. Experiment results verify that the machining method is appropriate for PSMVE.

Commentary by Dr. Valentin Fuster
2017;():V003T04A054. doi:10.1115/MSEC2017-2749.

In this study, we propose a new process planning system for machining operations, one which considers user strategies and intentions for such operations.

In previous process planning systems, the machining sequence is calculated geometrically, based on the Total Removal Volume (TRV) and the machining primitive region split from TRV. However, it remains difficult to determine the best machining sequence from among the large number of machining sequences calculated. Also, previous process planning systems do not consider user strategies and intentions in determining the appropriate machining sequence.

Our new approach stores geometrical properties of the machining primitives when the user selects a machining sequence. Using these stored geometrical properties, the appropriate machining sequence can be automatically selected. User strategies and intentions are thus considered in determining a machining sequence based on learned geometrical properties.

A case study was conducted to show the effectiveness of our proposed process planning approach. In the case study, user-specific machining sequences were automatically determined for various users, based on the relation among the geometrical properties of the machining primitives and the individual user’s strategies and intentions.

Commentary by Dr. Valentin Fuster
2017;():V003T04A055. doi:10.1115/MSEC2017-2928.

This paper presents design of a novel boring bar with a ratio of cylinder length to diameter (L/D) of 10 to suppress chatter vibration regardless of low stiffness. It is essentially difficult to decrease the compliance of the long slender structures. However, nominal compliance of the displacement along the depth of cut direction against the resultant cutting force may be regulated by giving the designated anisotropy upon the boring bar dynamics. The past research has clarified the feasibility through turning experiments by using the developed boring bar with L/D of 4. In the present study, much slenderer ones with L/D of 10 are designed, which are significantly flexible but attractive to manufacturers. Finite element analysis (FEA) is utilized to estimate dynamics of the anisotropic boring bar. Through analytical investigations, two kinds of boring bars were designed, where the compliance ratio of 1.53 or 1.88 was accomplished. Influence of several conditions on the chatter stability was investigated. Analytical investigations revealed that the chatter stability is significantly improved at a designated depth of cut by utilizing the proposed designs regardless of feed rate. In particular, the compliance ratio of 1.55 showed wider stable zone to attain chatter free boring, while chatter avoidance is impossible by use of the conventional isotropic boring bar under the same conditions.

Commentary by Dr. Valentin Fuster
2017;():V003T04A056. doi:10.1115/MSEC2017-3054.

Parallel turning attracts attention as one of the important technologies for the multi-tasking machine tools. This is because there is a potential to enhance the stability limits compared to turning operation using single tool when cutting conditions are properly selected. Although stability prediction models for parallel turning have been developed in recent years, in-process monitoring technique of chatter is almost out of focus.

In this study, to suppress chatter vibration, unequal pitch turning method was proposed. In this method, the upper tool was controlled based on optimum pitch angle calculated from spindle speed and chatter frequency. Chatter frequency was identified from estimated cutting force by disturbance observer. From the result of parallel turning test, it is clear that chatter vibration can be suppressed by controlling the upper tool based on optimum pitch angle.

Topics: Turning , Chatter , Cutting
Commentary by Dr. Valentin Fuster
2017;():V003T04A057. doi:10.1115/MSEC2017-3104.

Dramatic advancements and adoption of computing capabilities, communication technologies, and advanced, pervasive sensing have impacted every aspect of modern manufacturing. Furthermore, as society explores the 4th Industrial Revolution characterized by access to and leveraging of knowledge in the manufacturing enterprise, the very character of manufacturing is rapidly evolving, with new, more complex processes and radically new products appearing in both the industries and academe. As for traditional manufacturing processes, they are also undergoing transformations in the sense that they face ever-increasing requirements in terms of quality, reliability and productivity, needs that are being addressed in the knowledge domain. Finally, across all manufacturing we see the need to understand and control interactions between various stages of any given process, as well as interactions between multiple products produced in a manufacturing system. All these factors have motivated tremendous advancements in methodologies and applications of control theory in all aspects of manufacturing: at process and equipment level, manufacturing systems level and operations level. Motivated by these factors, the purpose of this paper is to give a high-level overview of latest progress in process and operations control in modern manufacturing. Such a review of relevant work at various scales of manufacturing is aimed not only to offer interested readers information about state-of-the art in control methods and applications in manufacturing, but also to give researchers and practitioners a vision about where the direction of future research may be, especially in light of opportunities that lay as one concurrently looks at the process, system and operation levels of manufacturing.

Topics: Manufacturing
Commentary by Dr. Valentin Fuster

Manufacturing Equipment and Systems: Intelligent Maintenance Decision Making of Manufacturing Systems

2017;():V003T04A058. doi:10.1115/MSEC2017-2643.

This paper provides a case study of diagnosing helicopter swashplate ball bearing faults using vibration signals. We develop and apply feature extraction and selection techniques in the time, frequency, and joint time-frequency domains to differentiate six types of swashplate bearing conditions: low-time, to-be-overhauled, corroded, cage-popping, spalled, and case-overlapping. With proper selection of the features, it is shown that even the simple k-nearest neighbor (k-NN) algorithm is able to correctly identify these six types of conditions on the tested data. The developed method is useful for helicopter swashplate condition monitoring and maintenance scheduling. It is also helpful for testing the manufactured swashplate ball bearings for quality control purposes.

Commentary by Dr. Valentin Fuster
2017;():V003T04A059. doi:10.1115/MSEC2017-2736.

Online condition monitoring systems play an important role in preventing catastrophic failure, reducing maintenance costs, and improving the system reliability. In this paper, wind turbine gearbox mechanical fault detection system is developed. An adaptive filtering technique is applied to separate the impulsive components from the periodic components of the vibration signals. Then different features of the periodic components and impulsive components are extracted. An extreme learning machine based classifier is designed and trained by using the features extracted from simulated vibration data of wind turbine gearbox. Simulated vibration signals of wind turbines gearbox are used to demonstrate the effectiveness of the presented methodology.

Commentary by Dr. Valentin Fuster
2017;():V003T04A060. doi:10.1115/MSEC2017-2765.

In this paper, a segmental hidden Markov model (SHMM) with continuous observations, is developed to tackle the problem of remaining useful life (RUL) estimation. The proposed approach has the advantage of predicting the RUL and detecting the degradation states simultaneously. As the observation space is discretized into N segments corresponding to N hidden states, the explicit relationship between actual degradation paths and the hidden states can be depicted. The continuous observations are fitted by Gaussian, Gamma and Lognormal distribution, respectively. To select a more suitable distribution, model validation metrics are employed for evaluating the goodness-of-fit of the available models to the observed data. The unknown parameters of the SHMM can be estimated by the maximum likelihood method with the complete data. Then a recursive method is used for RUL estimation. Finally, an illustrate case is analyzed to demonstrate the accuracy and efficiency of the proposed method. The result also suggests that SHMM with observation probability distribution which is closer to the real data behavior may be more suitable for the prediction of RUL.

Commentary by Dr. Valentin Fuster
2017;():V003T04A061. doi:10.1115/MSEC2017-2880.

Artificial intelligence techniques can play a significant role in solving problems encountered in the domain of Total Productive Maintenance (TPM). This paper considers a new reinforcement learning algorithm called iSMART, which can solve semi-Markov decision processes underlying control problems related to TPM. The algorithm uses a constant exploration rate, unlike its precursor R-SMART, which required exploration decay. Numerical experiments conducted here show encouraging behavior with the new algorithm.

Commentary by Dr. Valentin Fuster
2017;():V003T04A062. doi:10.1115/MSEC2017-2936.

As the U.S. manufacturing sector becomes more and more competitive, manufactures are seeking to adopt more cost and resource efficient operation practices. Several research studies that optimize production, maintenance, and energy have been conducted. However, research integrating them all in one model is less developed. In this paper, we present a framework on joint maintenance and energy planning while considering production throughput target and buffer constraints. The problem considers a Time-of-Use (TOU) demand response program such that the cost of production, energy, and maintenance is reduced. A sensitivity analysis considering the effect of production system parameters, such as machine rated power, machine production rate, number of maintenance crew resources, etc., on the cost per part is conducted. In addition, the sensitivity due to varying the individual unit costs incurred from production throughput, power demand, and maintenance is investigated. This study can guide manufacturers and researchers in determining the value of using such joint methods. Furthermore, the sensitivity analysis considering joint energy and maintenance can aid manufacturers and researchers in data acquisition so that the most sensitive parameters are given priority, help identify which controllable parameters have the largest impact on the system performance; and determine which parameters are most impactful.

Commentary by Dr. Valentin Fuster

Manufacturing Equipment and Systems: Monitoring, Sensing, and Control for Intelligent Machining and Inspection

2017;():V003T04A063. doi:10.1115/MSEC2017-2641.

Fiber Bragg grating (FBG) sensors have been widely used in monitoring of the mechanic equipment. However, for measuring high-speed dynamic signal of a large mechanical equipment, the demodulation rate of the interrogator should be very high, while the number of sensors could be tens or hundreds, thus, a large amount of sensing data could be generated. Nonetheless, a network throughput of the interrogator based on the software stack is relatively low and a large amount of data cannot be transmitted simultaneously, which becomes the bottleneck of the sensing system. In order to promote the network throughput, a hardware TCP/IP stack based on the field programmable gate array (FPGA) is proposed. In contrast to the existing hardware stacks, this stack is designed with a new module structure that is divided according to functions instead of protocol types. It can realize both UDP and TCP transmissions with less logic elements than similar designs. Unlike ASIC TCP/IP stack, the entire system can be realized on a single FPGA chip and upgraded without changing of the original hardware circuit. The proposed design has two key features. Firstly, the hardware stack can be connected directly to the data acquisition logic part without software operations thus the data throughput from the signal acquisition to the network transmission can maintain a relatively high speed. Therefore, the system can demodulate data from hundreds of sensors at high speed and transmit them in real time. Secondly, the module structure is clear and independent of specific FPGA platform. Consequently, it can be transplanted or upgraded easily in order to meet different practical demands. The proposed design embodies the characteristics and advantages of the system on a programmable chip (SOPC). In order to validate the proposed design, all logic modules were simulated and the design was tested on the circuit board. Performance test results have shown that UDP and TCP throughputs of the proposed hardware stack are up to 80Mbps in the case of 100Mbps Ethernet controller chip, which is about eight times higher than throughput of software design. Finally the design was verified by monitoring of the oil pipeline platform. The obtained results have shown that proposed design can detect the vibration frequencies of the oil pipeline that are around 600Hz and it can sample 288 FBG sensors and transmit sensor data correctly. Thus the proposed design is suitable for a large sensing system intended for the dynamic monitoring of the mechanical equipment.

Topics: Hardware , Design
Commentary by Dr. Valentin Fuster
2017;():V003T04A064. doi:10.1115/MSEC2017-2756.

A new computer numerically controlled lathe containing a carbon fiber-reinforced plastic pipe frame was developed in this study. The pipe frame structure is lightweight and allows the machine size and stiffness to be adjusted by varying the parameters of the frame structure. Furthermore, using carbon fiber-reinforced plastic instead of steel improves the heat deformation properties of the frame.

When employing a pipe frame structure for machine tools, structural vibration can be problematic. In particular, the relative vibration between the tool and the workpiece must be suppressed to improve the accuracy of the machined surface. To achieve this vibration suppression, an actuator driven by piezoelectric elements was installed in the frame of the developed CNC lathe structure to counteract the vibration at specific modes by applying active vibration control. As a result of using the proposed vibration control method, the vibration amplitude was reduced by up to 88.6% compared with that without control. Additionally, the circularity of workpiece was improved by 27%.

Commentary by Dr. Valentin Fuster
2017;():V003T04A065. doi:10.1115/MSEC2017-2773.

In production and manufacturing engineering, some redesign, distorted or worn parts cannot be milled according to their original CAD geometries due to the existed shape deviations. Generating CAM data in traditional manner for each individual part is time-consuming and labor intensive. This paper proposes a quick and convenient adaptive approach for machining distorted hollow blade, in which a template CAM data is spatially deformed according to some measured points from actual shape. The actual shape of the part was firstly inspected by on-machine measurement method. The measured points data was matched to the original nominal CAD geometry with ICP algorithm afterwards, by which the point-pairs between the measurement points and their corresponding points were established. Based on the distance deviations between these point-pairs, global and local modifying methods of template CAM data were developed using spatial deformation. By embedding the template CAM data in the calculated deformation volume, a new CAM data was achieved. Finally, a series of measurement and machining tests were performed, which validates the feasibility of proposed adaptive machining approach in this paper.

Commentary by Dr. Valentin Fuster
2017;():V003T04A066. doi:10.1115/MSEC2017-2789.

This paper describes a method using electrical characteristics of the torch, flame, and work piece to replace active sensing elements most commonly used for mechanized oxyfuel cutting applications; height, fuel/oxygen ratio, work temperature, and preheat flow rate. Calibrations are given for the torch under test for standoff accurate to ±1/32 in (0.8 mm) and F/O ratio accurate to ±.008. Methods are proposed for balancing flow across multi-torch systems, and detecting the work kindling temperature. Additional work is needed if calibrated flow and work temperatures are to be measured electrically.

Topics: Sensors , Cutting
Commentary by Dr. Valentin Fuster
2017;():V003T04A067. doi:10.1115/MSEC2017-2858.

Augmented reality is currently riding a wave of success in the consumer sector. In manufacturing, in spite of the global hype surrounding Industry 4.0 and the ever growing demand for more personalized and more complex products, there are currently only a handful of viable augmented reality applications that have actually made their way onto the production line. But why is this, and what applications genuinely bring added value? These are the questions that will be considered in this paper, presenting four examples of augmented reality concepts that have made their way from research into manufacturing. The augmented reality examples range from process support on a machine and support for a process chain, to use in education and training and even marketing applications.

Commentary by Dr. Valentin Fuster
2017;():V003T04A068. doi:10.1115/MSEC2017-2934.

The objective of this work is to fabricate instrumented cutting tools with embedded thermocouples to accurately measure the tool-chip interface temperature in interrupted and continuous turning. Thin-film thermocouples were sputtered directly onto the flat rake face of a commercially available tungsten carbide cutting insert using micro machined stencils and coated the measurement junction with a protective layer to obtain temperature data 1.3 μm below the tool-chip interface. Oblique interrupted cutting tests on AISI 12L14 steel were performed to observe the influence of varying cutting speeds and cooling intervals on tool chip interface temperature. An additional cutting experiment was conducted to monitor the interface temperature change between interrupted and continuous cuts.

Commentary by Dr. Valentin Fuster
2017;():V003T04A069. doi:10.1115/MSEC2017-3022.

Packing processing parameters, including packing pressure and packing time, have significant impact on the internal molecular orientations, mechanical properties and optical performance of injection molded polymeric products. One of the limitations of cold-runner injection molding machines is the lack of real-time control of packing processing parameters during an injection molding cycle. As a result, a new melt modulation device has been developed and experimentally validated to control melt flow and manipulate processing parameters during cold-runner manufacturing. The use of the integrated melt modulation device has shown enhancement of physical properties and optical performance of injection molded polymeric products. Numerical simulations and experimental results of common thermoplastic optical polymers, such as PMMA, PC, and GPPS have been conducted and briefly demonstrated herein.

Commentary by Dr. Valentin Fuster
2017;():V003T04A070. doi:10.1115/MSEC2017-3027.

The key performance indicators in Chemical Mechanical Planarization (CMP) processes are usually assessed by measuring the material removal rate (MRR) and Within-Wafer-Nonuniformity (WIWNU), which are vitally dependent on the processing variables including down pressure, wafer rotation, polishing pad rotation, polishing table rotation, slurry flow, and the condition of the polishing pad etc. MRR is critical to the WIWNU also since MRR can infer the end-point in the polishing process. In this study, empirical approaches were conducted to model the MRR with the production CMP settings. With the collected data from real semiconductor manufacturing processes, correlation and principle component analysis (PCA) were conducted to select the features mostly related to the CMP processes, then neural network (NN) and adaptive neuro fuzzy inference system (ANFIS) based models were proposed to understand processing variables in CMP process and estimate the MRR. The NN and ANFIS models were compared on the performance metrics of 1) mean square error (MSE), and determination coefficient (R2) based on bootstrap. The bootstrap based evaluation shows that NN achieved a MSE of 9.68e03 with the R2 value of 0.81 in the training stage and MSE of 9.59e3 with the R2 value of 0.81 in the validation stage; ANFIS achieved a MSE of 126.24 with the R2 value of 0.9102 in the training stage and MSE of 6.17e4 with the R2 value of 0.3133 in the validation stage. The empirical models are promising to be integrated with the data-driven based control of CMP processes.

Commentary by Dr. Valentin Fuster
2017;():V003T04A071. doi:10.1115/MSEC2017-3028.

In the measurement of machine tool and robot geometric errors, one of the most extensively used instruments is the Laser Tracker (LT). Errors in the LT measurements will decrease the effectiveness of the error modeling and compensation methods that utilize these measurements. When the LT’s Absolute Distance Meter (ADM) is used without frequent referencing to a home position, large and long-term shifts occur. The ADM shift directly introduces errors in the radial component of every measurement in spherical coordinates, which will result in measurement errors in the Cartesian coordinates. Although the ADM shift is addressed in newer LT designs using internal referencing hardware, this paper presents a pragmatic and efficient software solution to ADM shift for LTs in which the internal referencing hardware is not embedded. The LT was measured for 22 hr in a temperature-constant room to examine the ADM shift effects on measurements. An ADM shift model was then proposed by assuming that the ADM shift equally affects radial components of all measurements wherever the target is, as long as it is within the measurement range. Another experiment was then performed to test the validity of the proposed model. After the model was identified and errors were corrected, the maximum temporal variation in the radial distance measurement is reduced by 80–86%.

Topics: Lasers , Modeling
Commentary by Dr. Valentin Fuster

Manufacturing Equipment and Systems: Nanomanufacturing of Multi-Functional Systems

2017;():V003T04A072. doi:10.1115/MSEC2017-2680.

Maskless nanolithography is an agile and cost effective approach if their throughputs can be scaled for mass production purposes. Using plasmonic nanolithography approach, direct pattern writing was successfully demonstrated with 22 nm half-pitch at high speed. Plasmonic nanolithography uses an array of plasmonic lenses to directly pattern features on a rotating substrate. Taking the advantage of air bearing surface techniques, the system can expose the wafer pixel by pixel with a speed of ∼10 m/s, much faster than any conventional scanning based lithography system. It is a low-cost, high-throughput maskless approach for the next generation lithography and also for the emerging nanotechnology applications, such as nanoscale metrology and imaging. A critical part of the PNL is to use plasmonic lens to deliver highly concentrated optical power at nanoscale. We have demonstrated such nanoscale process and achieved 22 nm resolution. Here, we report our recent efforts of designing new plasmonic nanofocusing structures that is capable of achieving optical confinement below 20 nm which can potentially support direct patterning at sub-10nm resolution.

Commentary by Dr. Valentin Fuster
2017;():V003T04A073. doi:10.1115/MSEC2017-2818.

Laser processing (sintering, melting, crystallization and ablation) of nanoscale materials has been extensively employed for electronics manufacturing including both integrated circuit and emerging printable electronics. Many applications in semiconductor devices require annealing step to fabricate high quality crystalline domains on substrates that may not intrinsically promote the growth of high crystalline films. The recent emergence of FinFETs (Fin-shaped Field Effect Transistor) and 3D Integrated Circuits (3D-IC) has inspired the study of crystallization of amorphous materials in nano/micro confined domains. Using Molecular Dynamics (MD) simulation, we study the characteristics of unseeded crystallization within nano/microscale confining domains. Firstly, it is demonstrated that unseeded crystallization can yield single crystal domains facilitated by the confinement effects. A phenomenological model has been developed and tailored by MD simulations, which was applied to quantitatively evaluate the effects of domain size and processing laser pulse width on single crystal formation. Secondly, to predict crystallization behaviors on confining walls, a thermodynamics integration scheme will be used to calculate interfacial energies of Si-SiO2 interfaces.

Commentary by Dr. Valentin Fuster
2017;():V003T04A074. doi:10.1115/MSEC2017-2833.

Supercapacitors fill the gap between the batteries and capacitors in the Nyquist plots, and being considered as the candidates for next generation energy storage due to the high-power density, long-term cyclability and moderate energy density. In order to fulfill the requirement for practical applications, it is necessary to further develop the current supercapacitors and enhance the energy density. Hence in this paper, we discuss the work we have done for developing high-performance supercapacitors, including synthesis of large surface area and high conductive carbonaceous materials and highly electroactive pseudocapacitive materials. Our works may pave the way for synthesis of high-performance supercapacitor electrode materials.

Commentary by Dr. Valentin Fuster

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In