Skip to main content

2023 | Book

Flexible Automation and Intelligent Manufacturing: The Human-Data-Technology Nexus

Proceedings of FAIM 2022, June 19–23, 2022, Detroit, Michigan, USA, Volume 2


About this book

This book gathers the second volume of the proceedings of the 31st edition of the International Conference on Flexible Automation and Intelligent Manufacturing, FAIM 2022, held on June 19 – 23, 2022, in Detroit, Michigan, USA. Covering four thematic areas including Manufacturing Processes, Machine Tools, Manufacturing Systems, and Enabling Technologies, it highlights advances in micro- and nanoscales processes, additive manufacturing, artificial intelligence and robotic applications, human-robot collaboration, as well as quality control, supply chain, industrial monitoring and management strategies. It also discusses important issues related to sustainability, waste management and remanufacturing. All in all, this book provides academicians, engineers and professionals with extensive information on both scientific and industrial advances in the converging fields of manufacturing, production, and automation.

Table of Contents


Manufacturing Processes

Electric Pulse Aided Draw-Bending of Ti-6Al-4V

Electric pulse aided deformation is gaining importance because of its potential to deform difficult-to-form materials (high strength steels, magnesium, and titanium alloys) at very low temperatures compared to hot/superplastic forming due to the electro-plastic effect. In the present work, electrically assisted draw-bending experiments on Ti-6Al-4V alloy are carried out to study the effect of electric pulse parameters (current density, energy density, frequency) on deformation energy and retained height of components. A custom-designed experimental setup is fabricated where there is a provision to pass the current only in the deformation zone to minimize the overall joule heating. Results indicate that deformation energy decreased, and retained height increased with increased energy density due to increased temperature rise. It is observed that force-drop and retained height increased with an increase in current density at constant energy density (temperature rise nearly the same), which is attributed to the electro-plastic effect. No significant change in the hardness values is observed with and without electric aid.

A. Subrahmanyam, S. Saurabh, K. Praveen, G. Ramu, N. Venkata Reddy
Investigation of Multi-material Composite Parts Manufactured by Multi-extrusion FDM Printer

FDM 3D printers with the ability of multi-material fabrication open great opportunities for designers and engineers to produce composite products that have properties tailored to customers’ requirements. In addition, a multi- nozzle-type printer can facilitate the manufacturing process by producing support structures for overhanging parts. In any case, printing objects which combine properties of several materials is often a difficult task. For example, if two materials have noticeable differences in physical or thermal properties, they can be rendered incompatible. Having considered a rich variety of available 3D printable polymers, this paper aims to analyze the mechanical and physical properties of 3D printed multi-material parts made of PLA and wax-based filaments. The wax being soft and flexible can be used as support structures or an infill pattern for a rapid casting application. The paper contains the experiment results of tensile testing that indicate the presence of strong interlayer bonding between two materials. In addition, it has been found that depending on the pattern of the composite and the thickness of the wax part, the mechanical properties such as plasticity and yield stress can change to some extent.

Muslim Mukhtarkanov, Essam Shehab, M. Hassan Tanveer, Sherif Araby, Md.Hazrat Ali
Enhancing Mechanical Property of Multi-material Printed Object Through Machine-Learning

Machine learning is gaining more popularity in the FDM process in the way of performance enhancement. The multi-functionality of multi-material printing and its rising employment makes the Machine-Learning (ML) tool more attractive as the diversity of process parameters involves many fabrication combinations. This paper describes the implementation of ML techniques in the production of multi-material objects to achieve a high mechanical outcome. A nozzle temperature of PLA and ABS extruders was chosen as an input feature for ML, whereas UTS was the target. 125 samples with additional 6 pieces for deviation cases printed for 25 temperature combinations. The decision Tree model exhibited improper prediction values. Although the next Random Forest model had a fairly good R2-0.78, the 3D graph of UTS had a coarse curve. The highest R2-0.81 belonged to the 5th degree Polynomial Regression model. According to this model, to acquire the highest UTS value-41.171MPa, extruding temperatures should be 216 C, and 246 C for PLA and ABS respectively.

Md.Hazrat Ali, Nurbol Sabyrov, M. Hassan Tanveer, Syuhei Kurokawa, Essam Shehab
Study of the Influence of Laser Welding Parameters on the Weld Quality and Microstructure of S355JR Structural Steel

Welding is employed greatly in the construction industry, with processes such as SMAW being the most used for welding structural steels. However, processes such as laser welding exhibit quite some advantages when compared to the conventionally used welding processes, offering the possibility to obtain welded joints in a faster manner, without the need for filler metals, even causing an improvement in mechanical properties, in some cases. There are, however, some limitations with the laser welding of structural steels, especially lack of penetration, which can cause welding defects, such as porosities and voids. In the present work, S355JR steel profiles with 8 mm thickness are produced by laser welding. This weld was analyzed, identifying possible defects, assessing weld quality and grain size, by performing SEM and EBSD analyses. The welding process was then adapted and used to produce thick S355JR steels, with 25 mm thickness. As for the thinner profiles, the weld quality was assessed and characterized, identifying possible defects and the feasibility of employing the laser welding process on the joining of thick structural steel plates.

V. F. C. Sousa, F. J. G. Silva, R. D. S. G. Campilho, A. G. Pinto, J. S. Fecheira
Optimizing the Ag Filler Metal Content on Brazing of Cu-Stainless Steel Pipes Joints for Carbon Dioxide Refrigeration Plants

A carbon dioxide refrigeration plant is characterized by reaching maximum temperatures in the order of 150 ℃, and maximum pressures in the order of 130 bar, which means that the materials used in the pipes, as well as the connections, must present high resistance, so as not to cause disturbances in the refrigerant fluid. To make these connections, brazing and TIG welding are the processes normally used in copper-copper, copper-carbon steel, and copper-stainless steel connections, optimizing their parameters, in order to create defect-free joints. In this work copper-stainless steel joints were produced by brazing, using brazing alloys with differing silver contents, 56% and 34% respectively. The microstructure of the produced joints was analyzed and characterized, evaluating the influence of silver content on joints’ quality. It was concluded that the joints obtained by using a brazing alloy with lower content of silver presented less defects and an overall better quality.

F. J. G. Silva, V. F. C. Sousa, R. D. S. G. Campilho, A. G. Pinto, J. Fecheira
Reflow Thermal Recipe Segment Optimization Model Based on Artificial Neural Network Approach

The temperature settings for the reflow oven chamber (i.e., recipe) are critical to the quality of the Printed Circuit Board (PCB) in the surface mount technology because solder joints are formed on the boards with the placed components during the reflow process. Inappropriate profiles cause various defects such as cracks, bridging, delamination, etc. Solder pastes manufacturers have generally provided the ideal thermal profile (i.e., target profile), and PCB manufacturers have attempted to meet the given profile by fine-tuning the oven’s recipe. The conventional method tunes the recipe to gather thermal data with a thermal measurement device and adjust the profile relying on the trial-and-error method. This method took a lot of time and effort, and it cannot guarantee consistent product quality because it’s so dependent on the engineers. We proposed (1) a stage-based (ramp, soak, and reflow) input data segmentation method for data preprocessing, (2) a model for predicting the zone temperature in the soldering reflow process (SRP) using a state-of-the-art machine learning, (3) an algorithm for generating the optimal recipe to reduce the gap between the actual processing profile and the target profile. Our method uses artificial intelligence, specifically a backpropagation neural network, to enable non-contact prediction using thermal data from a single experiment (BPNN). In the fully equipped in-house laboratory, the validity of the approach was tested. As a result, within 10 min of starting the experiment, the generated optimal recipe shows 99% fitness to the targeted profile.

Zhenxuan Zhang, Yuanyuan Li, Sang Won Yoon, Daehan Won
Experimental Verification of Knowledge-Based Welding Distortion Estimation Method

Automation of welding robot programming is a method to improve the productivity of multi-robotic welding production. However, robotic welding off-line programming and simulation software packages often neglect the effect of the welding distortions to a workpiece, which is one of the main problems during jigless assembly in multi-robot welding. Therefore, knowledge-based information of the amount of welding distortion is often required for the off-line programming of a jigless multi-robot welding to be successful. This paper assesses a knowledge-based welding distortion estimation method through practical verification experiments. The measured angular distortions, from preliminary welding experiments, were used as a basis for setting the preset angle and modifying the position of the handling robot in the verification welding experiment. The results indicate that the consistency in achieving the required preset angle during part positioning and the consistency of the amount of angular welding distortion, is sufficient to beneficially exploit the welding distortion in obtaining perpendicular fillet weld joint geometry, which fulfils the tolerances set in standards EN ISO 5817 and EN ISO 13920 respectively.

Hannu Lund, Sakari Penttilä, Tuomas Skriko
Use of Fused Filament Fabrication and Stereolithography Methods for the Additive Manufacturing of Horn Antennas

Additive Manufacturing (AM) is a rapidly developing set of methods with lots of benefits. This technology promises fast, cheap and lightweight fabrication for the antennas and waveguides. Different AM technologies such as Fused Filament Fabrication (FFF), Stereolithography (SLA), Material Jetting and Powder Bed Fusion are used to fabricate antennas operating at various frequency bands and provide desirable results. To test the feasibility of the AM technologies for antenna manufacturing, a specific antenna has been modeled and fabricated via FFF and SLA type of 3D printers. The fabricated antennas are coated with copper spray and aluminum and the radiation parameters are measured. The accordance of the measurement results with the simulations and CNC machined pair have been inspected.

Burak Caliskan, Kenan Capraz, Ulas Yaman
A Scale-Free Classification Model for Defect Diagnosis in the Pick and Place Machine

This study aims to develop a scale-free classification-based defect diagnosis model for the pick-and-place (P&P) machine in Surface Mount Technology (SMT) assembly line. SMT is a manufacturing process used to assemble Printed Circuit Boards (PCBs). The P&P process is the primary procedure that follows the application of solder paste or adhesive to the board. Generally, the industry uses an automatic optical inspection (AOI) machine to detect assembly defects just after the P&P process. However, inspection data from the AOI machine can only identify assembly defects; it cannot reveal the underlying causes of assembly failure. By conducting experiments with initial machine defects, it is possible to identify patterns associated with various root causes. Using the AOI and machine performance data, it is possible to trace the root causes of assembly defects using various machine learning methods. As the number of components used in the SMT assembly line increases, processing design of experiments (DOE), collecting sufficient data, and developing a defect diagnosis model for each type of component becomes time-consuming. The proposed model is trained on a single component type and then applied to other component types. Using the proposed model, when a new component is applied, the identification accuracies are more than 75.00% for most of the root causes without conducting DOE. It can significantly reduce the time required to process experiments, collect data, and adjust models for new types of components.

Yuqiao Cen, Jingxi He, Zhenxuan Zhang, Daehan Won
Effect of Post Weld Heat Treatment on Donor Material Assisted Friction Stir Welding of AA6061-T6 Alloy on Microstructure and Mechanical Properties

In this study, the influence of post weld heat treatment (PWHT) on the microstructure and mechanical properties of a copper (Cu) donor material assisted friction stir welding (FSW) of AA6061-T6 aluminum alloy has been investigated. Cu assisted FSW joints of AA6061-T6 alloy were prepared at an optimized constant tool rotational rate of 1400 rpm and welding speed at 1 mm/s. The Cu donor material of 20% thickness with respect to the workpiece thickness was selected to assist the FSW joining at the plunge stage. FSWed AA6061-T6 samples were prepared using solid solution treatment at 540 °C for 1 hour followed by quenching in water at room temperature. It was then artificially aged at 180 °C for 6 hours and 24 hours followed by air cooling. The microstructure and tensile fractured surfaces were analyzed using scanning electron microscope and optical microscope. The microstructure depicts an additional grain refinement in the stir zone (SZ) due to the occurrence of recovery and recrystallization with increasing aging time. Vickers micro hardness indicates a softening effect due to the dissolution of hardening precipitates. Hardness recovery is most likely attributed to the uniform distribution of fine hardening precipitates at all aging time levels. The maximum hardness was 92.5 HV at the SZ and the tensile properties were significantly improved by 20 % after solution heat treated at 540 °C for 1 hour followed by artificial aging at 180 °C for 24 hours.

Srinivasa Bhukya, Zhenhua Wu, Abdelmageed Elmustafa
Understanding the Microwire Casting Process Through Convolutional Neural Networks

Even though the Taylor-Ulitovsky process for producing microwire has existed and been widely used in the past century, there are various challenges facing the microwire manufacturing process, such as inconsistent wire diameter, constant breaks of microwire during fabrication, and the difficulty of producing wires with a smaller diameter. These challenges can make the microwire fabrication process inefficient, and this research aims to understand how thermal images from the fabrication process under various parameter settings can be used to assess and classify the quality of the microwire. Thermal videos and other process variables were collected from a microwire manufacturing lab, and the thermal image datasets from the video were trained using a pretrained Convolutional Neural Networks (CNN) in order to better understand how changing certain parameters for the microwire manufacturing process can affect the microwire quality. The features extracted from the thermal images using the proposed CNN-based machine learning algorithm is capable of classifying the microwire fabrication process into four stages, i.e., initialization stage, stable stage 1, stable stage 2, and ending stage. The stage classification accuracy reveals high repeatability and performance from the proposed CNN model. The results are promising since manufacturing process parameter settings can be adjusted and optimized by referring to thermal image characteristics, and therefore the CNN model can improve microwire quality and predict failure of the microwire.

Charles Z. Li, Yuri A. Gulak, Jingzhou (Frank) Zhao
Ontological Knowledge Graph Framework for 4D Printed Product Design: Elongated Homogenous Rod Case

This article presents an ontological knowledge graph for the 4D printed product design information model. Information models and repositories are essential to systematize/integrate 4D printed (smart) product design; however existing research works are limited to fully support the 4D product design modeling due to the interdisciplinary nature. This research examines whether the 4D printed product design data and transformation can be represented as a knowledge graph, stored in a graph database and the envisioned 4D printing design repository. For this article, we studied different types of 4D printing designs and abstracted basic elements from the perspective of engineering mechanics. An ontological knowledge graph formalism is developed based on Gruber and Olsen’s EngMath ontology. The presented ontological knowledge graph framework includes 4D primitive shapes, basic mechanical equations, an ontological formalism that contains relationships and equation parameters, and a graph database formatted output. We discuss in this paper how this formalism can be used to design a rod in a model visualization software to support 4D printed product design. Finally, we conclude with discussing the elongated homogenous rod case study in detail.

Shengyu Liu, Kyoung-Yun Kim
Development, Microstructural and Mechanical Analysis of Cu-Zn Alloy Produced by Sand Casting Process

Alpha-brass (Cu-Zn Alloy) boasts of excellent mechanical properties such as high strength, ductility, and excellent corrosion resistance. In this study, the process design, development and mechanical analysis of Cu-Zn alloy produced by sand casting process were carried out. The process parameter optimization was carried out using the Response Surface Methodology (RSM) with the process conditions in the following range: temperature (300–500 ℃) and zinc content (5–25%) having the hardness as the response of the designed experiment. The raw materials were scraps of copper wire and zinc battery casing and 13 different compositions of the alloy were prepared having the total mass for each weight percentage weighing 1.5 kg. The results obtained indicated that developed brass possesses good hardness and that the hardness increases with an increases in the zinc content but decreases with an increase in the temperature. The optimum process parameters which produced the highest value of hardness (85.6 BHN) are temperature (300 ℃) and zinc content (20%). It is envisaged that the findings of this work will assist the brass material developers and the end users in the development of products with excellent mechanical properties.

Ilesanmi Daniyan, Adefemi Adeodu, Khumbulani Mpofu, Ikenna Damian Uchegbu

Machine Tools

Voronoi Tessellation Application for Controlling Frequency Domain of a Titanium Plate

Aim of all designers is to optimize the product principally in term of mass. The classic manufacturing processes constraint the designer to use a limited number of parameters for obtaining the best results. New manufacturing processes like Additive Manufacturing, open the way to a new optimization strategies, one of the most important is the topology optimization. The objective function is to reduce the mass keeping other functionalities of the product intact. The starting geometry of each topology optimization can be the geometry used for the classic manufacturing method or it can be the lattice structure or a geometry with a tessellation applied by means Voronoi technique. Aim of this paper is to investigate the potential of Voronoi tessellation in the field of structural engineering. A titanium plate with Voronoi tessellation is modelled varying the number of seeds and keeping the total mass unaltered. Thanks to a finite element simulation, for each condition a modal analysis has been performed and the natural frequencies have been extracted. The paper discusses about the influence of the number of seeds to the natural frequencies of plate. This could be a new way and a starting point for topology optimization oriented to the management of natural frequency domain exploiting the Voronoi parameters.

Michele Calì, Filippo Cucinotta, Marcello Raffaele, Fabio Salmeri, Felice Sfravara
Use of Auxetic Infill Structures for the Compensation of Shrinkage in Fused Filament Fabrication Process

The polymer parts fabricated by Fused Filament Fabrication (FFF) process deteriorated noticeably in terms of dimensional accuracy due to the thermal processes during the fabrication and cooling phases. Commonly, researchers tend to concentrate on modifying the CAD model or fabricating the parts larger than the nominal then post-processing the artifacts to compensate for shrinkage. This paper investigates the effects of auxetic infill structures on shrinkage compensation. Parts with different internal geometries and infill structures were designed and fabricated via FFF using polylactic acid material. Critical dimensions of these parts were measured on a Coordinate Measuring Machine to observe the effects of different infill patterns on the dimensional accuracy of the parts. Measurement results showed that using an auxetic infill pattern with a symmetrical layout improves the accuracy of the parts significantly.

Rahman Uncu, Mehmet Canberk Bacikoglu, Burak Caliskan, Ulas Yaman
Graph-Based Analysis of Tool Life Parameters of the Turning Process in Small-Lot Production

Estimating tool life in small-lot production uses the same tool to optimize production plans of different products with various geometry and material. Process planning in small-lot production influences production efficiency, energy consumption, time, and cost. This paper studies the parameters that affect tool life in the turning process for small lot production. The paper expresses some parameters that affect tool life and other factors which impact these parameters. The results of this research are represented as a complex web of interdependencies and a relationship matrix. The relationship matrix illustrates the direct and indirect interdependencies of the parameters which influence tool life in the turning process. Moreover, the graph-based analysis demonstrates the weight of the parameters in estimating tool life in small-lot production. Finally, the results show which parameters have the most effect on tool life to consider in conducting a production plan to manufacture different products with different geometry and material using the same tool.

Sara Moghadaszadeh Bazaz, Juho Ratava, Mika Lohtander, Juha Varis
Milling a Titanium Alloy Using Different Machining Parameters: A Comparative Study on Tool Wear, Tool Life and Performance

The machining process is still quite relevant in the manufacturing industry, being employed in the production of high-precision and quality parts for a wide variety of industries, for instance, the aerospace and aircraft industries. The components produced for these industries are usually comprised of high-performance alloys, such as titanium alloys. However, there are some problems associated with the machining of titanium alloys. Therefore, new strategies, tools and sets of parameters are being used for the machining of these alloys, trying to correct problems associated with production quality and tool-life. In the present work, milling tests are conducted on a Ti6Al4V titanium alloy, using different machining strategies, mainly focused on the variation of the cut’s radial depth. The employed tools consist of 16 mm diameter mills, coated with TiAlN. After the machining, the tools are subjected to optical and scanning electron microscopy, in order to quantify the tool flank wear and the developed tool wear mechanisms for each of the used cutting conditions. The machined surface quality and machining time for each of the conditions are also analyzed. This work aims to expand the knowledge on the machining of these alloys, offering viable solutions that can be adopted in the future, when it is intended to optimize tool life and performance for the machining of titanium alloys.

V. F. C. Sousa, F. J. G. Silva, A. G. Pinto, R. D. S. G. Campilho, M. L. S. Barbosa, R. D. F. S. Costa
Proposal for a Double-Casing Prototype of a Pedometer for Dairy Cows, Made to Be Interchangeable, Through Numerical Investigation and 3D Modeling of Geometry

In barns for dairy cows, pedometers are used for monitoring cow behavioral activities. In detail, the increase in the motor activity recorded by pedometers is widespread used as an indicator of oestrus onset. Generally, all the cows are equipped with one pedometer attached to one cow leg for a working life of nearly 5 years. After that period, pedometers are disposed of. This practice could be a relevant issue regarding the environmental sustainability. The aim of this paper was to develop a prototype of a pedometer in compliance with the guidelines of the Green Deal, with particular attention to both the materials used in the production process and its management within dairy farms. An accurate study of the pedometer geometry was performed through both a Finite Element Analysis (FEA) and 3D Fusion Deposition Modeling (FDP) techniques. The material proposed to build the prototype, in addition to guarantying adequate chemical and mechanical resistance, is biocompatible and recyclable. The design of the prototype is characterized by the adoption of a double casing: the external one, with a protective function, is fixed to one cow foreleg; the inner one, used for housing the measurement devices, is removable and interchangeable. This solution will provide the pedometers with a longer lifetime than the existing commercial ones and, therefore, could contribute to limit the environmental burden deriving from pedometers dismissing.

Marco Bonfanti, Giovanni Cascone, Simona Maria Carmela Porto
A Regression Model for Tool Wear and Breakage Diagnosis

Many studies have addressed tool wear diagnosis, breakage detection, and tool vibration removal in machining large workpieces, such as moulds and aircraft parts. However, there have been difficulties in spreading commercialised tools to the field. This is because the optimisation of NC-data is based on machining experience and expertise. In particular, the expertise required to maintain the state of optimisation hinders the spread of commercial tools on actual machining floors. For this reason, NC data-based research has been conducted in CNC machining. In this paper, we propose a machining status diagnosis method using NC data. The machining load generated during machining is stored in synchronisation with the equipment–tool–material, then the correlation with the machining load can be expressed as a regression model, and a tool wear/damage detection method using this is presented. Thus, it is possible to provide auxiliary information for data-based management of individual tools of a CNC part mass production plant. In particular, the proposed method can be used as a standard for tool wear and adaptive control, even in the one-time machining of moulds, aircraft, and mechanical parts, and it can also be said to be a method to predict tool life. Therefore, this method can be considered an NC data-based on-site diagnosis method that can increase machining efficiency through repetitive learning.

S. G. Kim, E. Y. Heo, H. G. Lee, W. Kim, B. B. Choi, H. W. So, D. W. Kim
An Adaptive Control for NC Machining Using Reference Control Load Curves

Excessive cutting force accelerates tool wear, reduces the roughness of machined surfaces, and in severe cases leads to tool breakage and material waste. While cutting conditions would ideally be adjusted to account for both material/spindle speed/feed rate, it has proven challenging to predict and optimize for the dynamic characteristics of machining, such as tool vibration and wear, and spindle thermal deformation. Compounding the challenge, predicted tool paths may be accompanied by machining errors. In this paper we propose an advanced adaptive control method capable of balancing the cutting load, improving tool life, and reducing machining time. The proposed active adaptive control method (1) synchronizes the spindle load and NC-data and stores it, (2) analyses the stored data to create a reference control load curve capable of balancing the cutting load, (3) adjusts the tool feed rate using this reference control load curve, (4) engages a rapid traverse mechanism when the cutting load is small, and (5) applies an approach feed rate when the tool approaches a workpiece, reducing the impact when the former meets the latter. The proposed reference control load curve can be regenerated during machining to account for tool wear, chatter, and changes to workpiece shape that occur during machining. Our experimental results confirmed that the proposed method reduces machining time and increases tool life.

S. G. Kim, E. Y. Heo, H. G. Lee, W. Kim, D. W. Kim

Manufacturing Systems

Learned Manufacturing Inspection Inferences from Image Recognition Capabilities

Many complex electromechanical assemblies that are essential to some vital function of certain products can be time consuming to inspect to a sufficient level of certainty. Examples include subsystems of machine tools, robots, aircraft, and automobiles. A model-based definition with manufacturing tolerance specifications can address any design-related severe failure modes. However, out of tolerance conditions can occur due to either random common cause variability or undetected nonstandard results, such as foreign object debris. Application of various image recognition techniques potentially can save time by some automation of inspections. However, some of the most meaningful 3D recognitions may not be sufficiently reliable, and it can be an extensive process to train the recognition of all possible anomalies comprehensively enough for inspection certainty. This paper introduces a schema and method that can learn the likelihood that a specific autonomously inspected feature will be within tolerance specifications. These learned manufacturing inspection inferences from image recognition capabilities (LeMIIIRC) may be performed by accepting data inputs that can be obtained during the image recognition training process followed by machine learning of the likely results. The fundamental method is demonstrated by a realistic example with hypothetical manufacturing data.

Douglas Eddy, Michael White, Damon Blanchette
A Scalable Cloud-Based UAV Fleet Management System

In recent years, unmanned aerial vehicles (UAVs), or drones, have been widely used for a variety of civilian missions. Applications include surveillance, mapping, cinematography, search and rescue, goods delivery, security patrol, structure inspection and precision agriculture. As the drone fleets grow in size and heterogeneity, mission control and asset management become increasingly complicated. To address this challenge, we propose a scalable cloud-based UAV fleet operations management system. The system implements a centralized mission control approach by leveraging IoT infrastructure, real-time databases and mathematical optimization techniques. The proposed software system is able to facilitate fleet mission planning and execution for both real and simulated drones. This paper provides an overview of the architecture, design and functionality of the system as well as a testbed platform for future algorithm development, with pointers to detailed discussions of different components in the literature.

Zhenyu Zhou, Yanchao Liu
New Recycling Procedure of SMD Components for Reuse in E-Textiles in Accordance to the Green Deal Policy

The paper addresses the research, development and verification of new remanufacturing method for reuse of adhesive bonded components on electrically conductive textile ribbons used in e-textiles. The main purpose of this method is to save the new raw materials, reduce the waste to a minimal level, and make the process more economical. Also, in case of a lack of components, the method can prevent production from stopping. The first part of the article describes the environmental strategies and agreements in the world like Paris Agreement, European Green Deal or Green New Deal. This part also defines the article’s objectives. The second part describes the principles of circular economy and remanufacturing methods. The following part is focused on smart textiles, especially e-textiles and conductive textile ribbons. The next part describes the methodology used in this paper. Section 4 describes the designed method for remanufacturing components from textile ribbons. The last part describes the realized verification experiment. The results present that the method designed and verified in this paper is usable and remanufactured samples are functional like new.

Martin Hirman, Andrea Benešová, Jiří Navrátil, František Steiner, Jiří Tupa
Wastes Identification Through Kaizen Events: A Case Study in the Automotive Sector

The efficient use of lean tools and techniques leads to the reduction of non-value-added activities in production systems. Continuous Improvement (CI) efforts in a workshop format, a.k.a. Kaizen Event (KE), is one of these lean tools. Measuring the gain from KEs has always been a challenge and as a result, firms spend much effort fixing issues that are non-critical or have low or no effect on factory performance, therefore, it is necessary more research on the metrics and the outcomes KE, including waste metrics. This paper presents a case study within a company in the automotive electronics sector to characterize and present outcomes of eight KEs, within which a total of 136 wastes were identified. Categorizing these wastes by groups, results reveal that the “operator motion” is the waste category most frequently noticed by the teams, while “automatic assembly” is the most impactful one in terms of cycle time reduction. While this case study makes a significant contribution in providing empirical evidence of waste in an organization, more research is needed to develop context-specific tools to narrow down the wastes once they have been identified.

Angelica Muffato Reis, Sérgio Dinis Teixeira de Sousa, Lino Costa
Compact Test Machine for Secondary Packaging

Goods such as bottles, cans, jars, etc. are often bundled by applying secondary packaging. A dimensionally stable bundle facilitates the construction of a stable pallet load, to be evaluated according to the Eumos 40509 standard. This study investigates whether it is possible to develop a test method and corresponding test machine that quantifies the shape retention of a bundle in accordance with the test requirements of the aforementioned standard and the boundary conditions of an industrial environment. The method that has been developed applies alternating horizontal inertial forces whose magnitude remains constant for 300 ms with a compact machine. The control system of the corresponding test machine allows that the forces are applied accurately within the specs of the standard. The deformation of the bundles as a function of the number of load cycles is measured and allows the shape retention to be quantified.

Jannes Roman, Ward Nica, Gilles Verschueren, Peter Slaets, Marc Juwet
Implementation of Circular Economy in ETO Organisation: Use of Digital Lean and Risk-Based Thinking Perspective

There is a rising regulatory and societal concern for the environment, forcing companies to include the circular economy (CE) concept in their business models and management practices. The transition towards a CE is inescapable and requires engineering to order (ETO) organisations to re-think their strategy and the way they create and deliver value. “Digital lean” is a promising concept that provides major opportunities for enterprises to move towards more sustainable industrial value creation by incorporating CE goals. This manuscript first presents a short literature review that indicates a gap when it comes to the symbiosis of lean, digitalisation and CE. According to the literature review, there is a shortage of descriptive lean and CE industrial case studies to illustrate best practices in ETO organizations. To address the gap, this paper demonstrates the practical case study results of an ETO organisation, to illustrate how to use digital-lean and risk-based thinking in a CE transition phase. The paper highlights lean practices, supported by digitalisation, that enable project-oriented organisations to simultaneously improve the efficiency of the product life cycle process and process performance. A risk-based thinking perspective is used to evaluate risk in a case study of an organisation, when improving or establishing processes and controls for transition to CE. The findings presented in this paper increase knowledge of how ETO organisations can adapt to CE with the use of a digital lean and risk-based perspective. The findings are useful for practitioners of ETO organisations who are in the CE goals’ implementation transition phase.

Daria Larsson, R. M. Chandima Ratnayake, Arne Gildseth
A Lean Digital Approach to After-Delivery Services: A Case Study from the Multi-supplier Retail Industry

A Lean approach is frequently applied in the internal processual settings of many companies but is less common in the after-delivery service context, which is dominated by office and knowledge work. The multi-supplier retail (MSR) industry may benefit from a Lean digital approach to utilize effectively business-critical data from the after-delivery service (ADS) process. Through a case study of an ADS system in the MSR industry, this paper investigates how a Lean digital approach can be operationalized and support the improvement of customer service and product quality. During a two-year period, approximately 20,000 after-delivery cases incorporating (among others) different suppliers and product defects or non-conformities have been registered by consumers and processed through the ADS system. ADS processes in the given industry have arguably received less attention in the literature than the systems and processes leading up to the point of sale. To address this gap, this paper provides insights into how MSR companies may benefit and learn from a product defect or non-conformity that generates an ADS case by adopting a Lean digital approach. While the findings demonstrate synergies between Lean and digitalization to support the minimization of performance variation, visualization, and data-driven problem solving and decision making, digital solutions can also generate new forms of waste and complexity if their development is not aligned with the employees’ digital maturity and needs and made compatible or integrable with other internal and external systems.

F. P. Santhiapillai, R. M. Chandima Ratnayake
Participatory Modelling of Information Technology Equipment Vulnerability Using Causal Loop Analysis

The purpose of this study was to model the relationships between the vulnerability factors that relate to unintentional compromising electromagnetic emanations from Information Technology Equipment (ITE). ITE vulnerabilities of this type can occur in many contexts including office and manufacturing environments using for example thin client and laptop technologies. The study applies a Causal Loop Analysis to the vulnerability factors to show the inter-relationships between them. This enables a clearer understanding of their relative significance when assessing vulnerability likelihood. The resulting causal loop analysis will be used in future work to develop a decision support tool for cyber security practitioners, when assessing the likelihood of unintentional emanations from ITE leading to the compromise of sensitive information.

Maxwell Martin, Funlade Sunmola, David Lauder
Re-purposing a Welding Frame for a Tube Cutting Machine

The metalworking industry is a competitive market with a constant search for productivity and quality. In some cases, re-purposing of existing equipment can significantly increase the company's productivity. This work arises from the need to perform complex cuts in large tubes, namely fish mouth cuts, which currently is performed manually. The work is based on the re-purposing of an existing tube welding gantry to a semi-automatic tube cutting machine. Cost and return on investment (ROI) analyses are also required to assess the viability. The re-purposing of the existing frame on the factory floor was successfully accomplished, and the ROI of the proposed solution was approximately three years.

A. R. Vidal, R. D. S. G. Campilho, F. J. G. Silva, I. J. Sánchez-Arce
Researching Sensors for Dynamic Forces on Cobots

Cobots are supplied with sensors and safety features, allowing close human-robot interaction. However, most cobots have a rather small payload and a small range. Inertial forces occurring if the cobot is mounted on a track and forces caused by deformable loads may trigger the safety procedures of the cobot. In this study two methods to measure dynamic forces on a cobot are considered. In the first approach a dynamometer is mounted underneath the base of the cobot. In a second approach external forces at the end-effector are estimated by six torque sensors integrated in the cobot joints. A theoretical system model of the robot is used. As a reference a six axis force sensor is mounted on the end-effector. As expected, the forces obtained by the sensor on the end-effector are most accurate and reliable. Both other methods can deliver forces that can be used in the future to modify the cobot’s control system and avoid false triggering of safety procedures.

Gilles Verschueren, Ward Nica, Jannes Roman, Marc Juwet
A Methodological Framework to Assess Mental Fatigue in Assembly Lines with a Collaborative Robot

In the current manufacturing assembly lines, collaboration between human and robot plays a significant role in the final output of the assembly, be it performance, quality or overall reliability. In that regard, smooth collaboration between human and robot is required to minimize the probabilities of human-system error, potential loss of performance and quality, and minimize the risk of decision making mistakes. Mental fatigue, and more importantly the cognitive load, of human operators is a crucial aspect in decision making, potential of error during the task and the overall flow of the assembly process. This paper reports about the development of a methodological framework to assess mental fatigue during a collaborative assembly task. In this framework, general complexity of the process and assembly task is investigated, as knowledge of the dynamic and static complexity can be helpful in reducing mental fatigue and cognitive load. We validated the applicability of the proposed frame in a real based collaborative assembly process.

Panagou Sotirios, Fruggiero Fabio, Mancusi Francesco
Development of an Affordable Cost Estimation Tool for Machined Parts

Cost estimation of manufacturing processes is quite important, as it can be used to provide budgets to costumers, by providing these budgets accurately and in a fast manner greatly contributes for good relations between the company and the costumers. However, providing these accurate budgets is quite difficult and time-consuming for some processes. There is currently a lack of simple and inexpensive cost estimation tools, especially for manufacturing processes such as the machining one. In the present work, a cost estimation tool for milling operations was developed, calculating these costs based on the machining times and the initial amount of needed material. The developed tool can determine the total operation time needed to completely produce a certain part, providing information regarding these times, from the preparation stages to the final finishing operations. This tool was validated, by producing several machined parts and comparing the experimental values that were obtained to those provided by the developed cost estimation tool. The results were quite satisfactory, determining machining times quite accurately, however, for finishing operations there is still a slight associated error, primarily caused by the complexity of the produced parts.

V. F. C. Sousa, F. J. G. Silva, T. Pereira, L. P. Ferreira, J. C. Sá, P. Nogueira
Communication in Decentralized Production Systems – An Analytical Approach

Moving away from strictly centralized, hierarchical to decentralized decision-making structures in production processes generates various potentials. This includes high flexibility in reaction to unplanned events and an increased focus on individual customer requirements up to lot size 1. The decentral approach causes fundamental changes in the communication structure of the autonomously acting actors involved. The analysis of these changes can form a basis for determining decentralization and an evaluation of decentralized decision-making structures. This paper proposes an approach to operationalize the autonomy of actors in decentral-controlled production processes based on their communication structure. First, it creates an understanding of decentralization in production processes and gives a formal description. Next, it identifies a suitable indicator of social network analysis and the justification for the necessity of an additional factor. Afterward, it describes the aggregation to actors’ autonomy, the decentralization of process steps, and the decentralization of the process. Finally, the paper concludes with a simulation study that validates the results.

Hanna K. Theuer
Improvements in the Warehouse of a Danish Transport and Logistics Company Through a WMS Implementation

The purpose of this study was to solve a problem in the warehouse belonging to a Danish transport and logistics company. The detected issue was the inefficiency of the system used to manage warehousing and storage operations. Thus, this study aimed to achieve improvements to the performance of the warehouse and increase efficiency through the implementation of the warehouse management system (WMS) internally developed by company. The method used to achieve the goal was the implementation of the WMS ILIAS. From the collected results it was found that, of the 63 checked functionalities, the implemented software only possessed 27 (about 43%). With the process changes, there was also registered improvements in the picking time, in the storage utilization, the average dock to stock time, order accuracy, on the number of orders picked per hour and for the movements executed per hour. The inventory accuracy also reached a high level of accuracy.

Maria Teresa Pereira, E. Santos, J. C. Sá, Marisa Oliveira, L. P. Ferreira
Improving the Layout of the Supply Warehouse to the Land Cruiser 70 Series Assembly Line

This paper presents a warehouse layout improvement to supply an assembly line of an automotive company, using Lean methodologies. For the layout change to be efficient, it was first necessary to reorganize the material in the Final Assembly supermarket. Once this task was completed, the racks were reorganized. The implementation of these two improvements resulted in a significant reduction in the number of trips with full dollies. As far as the picking of dollies supplied by the mizusumashi is concerned, a reduction of about 29.02% in the distance traveled by the picking operator was achieved. Regarding the dollies supplied without mizusumashi, it was possible to transform 10.40% of the trips with dolly into trips without a dolly, which reduces the ergonomic risk level of the task. As for the layout of the pre-assembly cell, the workbenches and racks were reorganized, allowing the employee to perform all tasks on time and alone within the stipulated takt time (42.5 min).

Maria Teresa Pereira, Vera Pinto, J. C. Sá, Marisa Oliveira, L. P. Ferreira
Circular Waste Management by Start-Up Companies in Indonesia Using a Bottom-Up Circular Economy Business Model

The lack of government regulation to support circular economy (CE) initiatives has led to the emergence of start-up companies in Indonesia that focus on processing, recycling, and creating new valuable products or materials from household waste. This research seeks to identify trends and people’s future interest in these businesses by collecting and analyzing data from Instagram. The analysis uses a multiple case study approach by combining quantitative and qualitative data analysis methods. Based on the findings of this analysis, it was found that acceptable waste types, education about waste separation, collaboration on waste collection or treatment facilities, service area expansion, and the global issue of waste recycling were the predominant discussion topics in waste management startup Instagram accounts. This study argues that collaboration among start-ups with other parties such as government and industry is necessary to expand their service area coverage and capability. The authors also recommend that start-ups actively educate members of the public to separate their waste and communicate with their local community using examples of best practice from around the world.

Noorhan Firdaus Pambudi, S. M. Samindi M. K. Samarakoon, Togar Mangihut Simatupang, Nur Budi Mulyono, Liane Okdinawati
On the Necessity for Planning and Integration of Manufacturing Tasks to Align with Health Sector Needs and CE Goals: A Systematic Literature Review

The healthcare supply chain is significantly unsustainable due to the linear resource consumption. Implementing circular economy (CE) principles in the health sector supply chains are in an evolution phase. Manufacturers and service providers need to strengthen the integration for enhancing healthcare supply chains to minimize linear resource consumption. There is an opportunity for improvement, especially from a planning integration perspective, to satisfy circular economy goals and face the volatility of consumer demand, like in pandemic conditions. Lean thinking should be applied in supply chains to achieve waste minimization and improved operational efficiency. A systematic literature review is conducted to develop the strategic framework for aligning the manufacturing and health sector planning and integration. The resulting framework explains strategic, tactical, and operational levels for planning and integration. The framework also illustrates the interaction between internal and external value chains to give a clear flow of integration among stakeholders.

Kartika Nur Alfina, R. M. Chandima Ratnayake, Dermawan Wibisono, Nur Budi Mulyono, Mursyid Hasan Basri
Increasing Quality Control of Ultrasonically Welded Joints Through Gaussian Process Regression

Due to the recent advances in digitisation of the manufacturing industry and the generation of manufacturing data, there is increasing interest to integrate machine learning on the shop floor to improve efficiency and quality control. Ultrasonic welding is an emerging joining process used in various manufacturing industries, and is an energy efficient, cost-effective method of joining similar or dissimilar materials. However, the quality of the joint achievable is heavily dependent on process input parameters. In this study, a Gaussian Process Regression (GPR) model is developed to map the relationship between process parameters and joint performance for ultrasonically welded aluminium joints, with a view to improving quality control in a manufacturing setting. Initially, a 33 full factorial design of experiments is conducted to investigate the influential parameters, then a GPR model is trained on the experimental data. In-process sensor data is also used to infer process performance. To assess the prediction performance of the model, ten unseen parameter combinations are predicted and compared to their respective experimental result. The model demonstrates a high level of accuracy producing a Pearson’s correlation coefficient of 0.982 between the predicted and actual results for all data. The mean relative predictive error for unseen data is 7.35%.

P. G. Mongan, E. P. Hinchy, N. P. O’Dowd, C. T. McCarthy
Prioritization of Industry 4.0 Technologies to Increase Maturity in Lean Manufacturing from the Perspective of Enterprise Interoperability

In face of the deep changes in the global industrial scenario stemming from the concepts of Industry 4.0 and Lean Manufacturing philosophy, the adoption of new priority technologies associated with these concepts ensure that a company is permanently competitive in the market. The main goal of this article is the construction of diagnostic and decisional evaluation models, in the Lean Manufacturing domain and from the perspective of interoperability, for prioritizing the adoption of Industry 4.0 technologies, and using multi-criteria decision-making methods (MCDM) as support. For the diagnostic evaluation, a hybrid approach of the AHP and DEMATEL multi-criteria methods was carried out, using the Lean Manufacturing elements as criteria for the decisional evaluation. From the AHP result, the criteria diagnosed as fragile were selected and subsequently applied in DEMATEL, a method that allows the evaluation of the influences generated between the criteria. The aggregation of the weight structure resulting from the AHP and Dematel methods allows the characterization of the diagnostic weighting to define the weights of the PROMETHEE II decision matrix, guiding the prioritization of alternatives, i.e. technologies of Industry 4.0. The approach was applied in a company of the metal mechanic sector, and the results indicated that Enterprise Resource Planning (ERP) integrated with Business Analytics (BA) was ranked as the technology that should be implemented with priority to raise the level of organizational maturity on Lean Manufacturing. The integration of these concepts offers several advantages for the organization such as waste reduction, continuous improvement support (Kaizen), and information sharing.

Giovana R. D. N. Martins, Luiz F. P. Ramos, Eduardo F. R. Loures, Vanessa A. dos Santos, Lucas M. B. do Amaral, Fernando Deschamps
Design and 3D Printing Fabrication of a Low-Cost Lightweight Robot Manipulator

There is an increasing demand for low-cost and lightweight robot manipulators to operate in the most diverse domains. In the last few years, in parallel with additive manufacturing advancement, we witness the emergence of innovative robot designs fabricated using 3D printing. However, the development of a functional robot is still a complex process, requiring skills in multidisciplinary knowledge areas. This paper presents a novel low-cost serially linked lightweight robot manipulator. It features 6° of freedom (DOF) plus a 1 DOF gripper, an innovative structural design addressing topological optimization, and its structure is fabricated using a conventional 3D printing machine. Mechatronics hardware is based on standard off-the-shelf actuators and control boards. Experimental results demonstrated the robot’s efficiency, namely in what concerns to repeatability (0.4 mm). The robot presented a payload capacity of 0.4 kg, which is quite remarkable since the robot’s total mass is around 0.7 kg.

Francisco Cruz, Mohammad Safeea, Mihail Babcinschi, Pedro Neto
Multi-agent Based IEC 61499 Function Block Modelling for Distributed Intelligent Automation

A major challenge for traditional systems is the lack of capabilities to automatically discover alternative solutions and actively deploy corresponding functions to intelligently adapt to changes in dynamic environments. In this paper, we continue our previous research of developing a two-layer architecture for modelling industrial cyber-physical systems and focus on the IEC 61499 function block based low-level physical module design. We propose an architecture model to integrate function blocks with intelligent agents to support self-management capabilities for low-level physical modules to quickly respond to changes. In the proposed architecture, a self-manageable service model is introduced for IEC 61499 function block modelled systems and designed as a multi-agent model with Self-Manageable Service Execution Agent, Self-Configuration Agent, Self-Healing Agent, Self-Optimization Agent, and Self-Protection Agent. The proposed modelling framework is tested with preliminary experiments on Raspberry Pi using the agent modelling tool SPADE and function block modelling tool Eclipse 4diac.

Guolin Lyu, Robert W. Brennan
Prioritization of Industry 4.0 Technologies Based on Diagnosis and Performance Indicators Associated with Lean Manufacturing Under Interoperability Requirements

Due to the transformations stemming from Industry 4.0, many organizations are adopting technologies to make their factories more connected and interoperable. Enterprise interoperability is characterized by the ability to exchange information between one or more systems, departments or companies. The classic interpretation of EIA (Enterprise Interoperability Assessment) proposes concerns and barriers that characterize the constraints to obtain an ideal organizational performance. The literature is insufficient to present approaches that relate the implementation of new technologies with other business domains. Thus, to keep the company competitive through waste elimination practices and optimization processes, the Lean Manufacturing (LM) Production system becomes a source of performance indicators, which will be parameters for the adoption of technologies that must meet interoperability requirements. This work proposes a framework to obtain a technology prioritization model based on a diagnostic approach considering LM performance indicators within the scope of Interoperability in a company of the metal mechanic sector. Firstly, the PROMETHEE II multi-criteria decision analysis (MCDA) method will be used to identify the technologies that provide good results to the LM performance indicators of the organization. The identified technologies will, secondly, be evaluated using the ELECTRE I method, which will indicate those that best meet the organization's needs, jointly considering LM criteria and interoperability requirements. The results show that the technologies related to data synchronization, tracking, integration, process management, schedule monitoring, kanban update, and standardized interfaces are decisive, since they provide good results to the LM performance indicators and relate them to the organization's interoperability requirements.

Milena da Rocha Moro, Eduardo de Freitas Rocha Loures, Anis Assad Neto, Luiz Felipe Pierin Ramos, Vanessa Santos, Lucas Montanari Bento do Amaral
Modeling and Simulation to Assess the Role of Culture for Successful Lean Transformations

Lean manufacturing has been embraced by many companies as the business model to promote operational excellence. However, a high percentage of lean implementation attempts have failed. This failure is often attributed to the sole focus on the hard side (tools) of lean, at the expense of the soft side (culture). Most of the research investigating the culture-related aspects of lean either attempt to raise awareness about culture or merely promote a conceptual approach. This paper proposes a simulation-based method that can be used to identify and assess the role of various factors essential to forming an organizational culture conducive to successful lean transformations. The culture of an organization emerges from interaction among members, interaction with the environment, and personal attributes. Agent-based modeling is an effective technique for evaluating dynamic systems of interacting agents and therefore is potent for modeling complexities that arise from interactions of the system components. However, due to blurred indicators which exist in defining culture, many measures are described subjectively by linguistic terms. Therefore, the use of fuzzy logic is proposed to emulate subjective human decision making by team members during interactions. Preliminary efforts for fuzzy-integrated, agent-based modeling and simulation are presented in an attempt to demonstrate the potential of using the method to identify and assess factors influencing organizational culture for lean transformations.

Amir Najarzadeh, Fazleena Badurdeen
Sustainable Industry 4.0 Methodology for Improving SMEs’ Performance

Industry 4.0 concepts have been elaborated in response to an increasing rate of customized demands, in order to keep high industrial performance for enterprises. These concepts are based on the introduction of new technologies such as collaborative robotics, artificial intelligence, big data or internet of things, in the manufacturing performance improvement. Indeed, the addition of organizational methods in the improvement contributes to the company's positive digital transformation. For instance, lean manufacturing, with reduction of wastes and value-added management, corresponds to a methodology that could be exploited for increasing the performance of SMEs. This paper focuses on how to put sustainability as the kernel of company digital transformation and new technologies as a support for humans in the future manufacturing. Through a use case, this paper presents the concepts of smart manufacturing and flexibility 4.0 for sustainably optimizing the company performance. After a literature review on industry 4.0, flexibilization 4.0, smart manufacturing and lean manufacturing, the concepts developed in this frame will be exposed. Then, the intelligent system being developed for supporting the SME digital transformation will be presented. An application in the electronic card production sector will be shown.

Clément Soudé, Paul-Eric Dossou, Gaspard Laouenan, Baptiste Duquenne
Development of an Intelligent System for Supporting the Sustainable Digital Transformation of the SME Supply Chain

The covid pandemic has disturbed the logistics and industrial organization of companies. In Europe, this specific context, in addition to the war in Ukraine, increases the gasoil price, creating an augmentation of the freight transportation global costs of companies. Industry 4.0 and logistics 4.0 concepts, developed in advanced countries such as USA, Germany, or France, are used with success for improving the company’s performance. Despite the benefits of these concepts on the company transformation, numerous brakes exist for their implementation in SMEs. This paper presents a sustainable methodology more adapted for transforming digitally the SME supply chain. Sustainability is used in this methodology as the kernel and is combined with new technologies and organizational methods in the performance improvement. Indeed, an intelligent system is being developed for supporting the methodology implementation in SMES. In this paper, a focus is made on the decision aided module of this intelligent system. After a literature review, the sustainable methodology, and the architecture/development of the intelligent system will be shown. Then, the structure of the decision aided module will be exposed. Finally, an illustration case of SME supply chain digital transformation will be shown.

Paul-Eric Dossou, Cindy Dondji Nguefack, Zineb Daheur
Discussion on Achieving Resistance Spot Welding Knowledge Completion

Resistance sport welding (RSW) research usually suffers a lack of data to build an accurate estimation model. Besides physical experiment data, the text knowledge for welding mechanisms would be additional data. However, the text knowledge is often incomplete. Knowledge completion (KC) is expected to solve the problem. It aims to derive the missing (or unknown) knowledge from explicitly expressed (or known) knowledge. So far, the KC study has focused on web-based knowledge like DBLP and Freebase. Research on RSW KC is rare. The characteristic of RSW knowledge is different from that of web knowledge. Web knowledge uses triples for its representation and triples are a type of graph. Thus, Graph Neural Networks (GNNs) has recently become a promising KC algorithm. To assess whether the GNN-based KC is feasible for RSW KC, this paper performs three tasks: assessing RSW knowledge types; building possible scenarios for RSW KC; discussing the applicability of the GNN-based KC method to RSW KC by using GNN-centered problem formulations to resolve the scenarios. This paper highlights two considerable challenges. The first is that RSW knowledge requires causal relations in addition to the structural relations indicated in web knowledge. The second is that, although the problem of RSW KC could be formally formulated the same as web knowledge KC, additional research works are needed to confirm that the formulation goes in a sufficient level because the amount of data issue for GNNs remains.

Jaemun Sim, Kyoung-Yun Kim
CHAIKMAT 4.0 - Commonsense Knowledge and Hybrid Artificial Intelligence for Trusted Flexible Manufacturing

Flexible manufacturing plays an important role in Industry 4.0 for developing the factory of the future and requires enhanced planning, scheduling, and control. The quick and effective adaptation in the production line in response to customers’ requirements or face of unwanted situations will promote considerable flexibility in manufacturing. CHAIKMAT is a research project funded by the French National Agency of research that aims to add flexibility and transparency to manufacturing through trustful automatic decision-making. The project proposes a human-centric AI approach that investigates whether an available set of machines can perform a specific production process and then provides human experts with meaningful explanations of how the decision process is conducted. A hybrid predictive model, comprising of both semantic reasoning and machine learning system will help in real-time decision making through the automated analysis of two sources of information: a stream of machine-monitoring data describing the current state of the production line and a common-sense knowledge graph (MCSKG) that is modelled based on machine capability and process planning ontology model. Furthermore, this hybrid predictive model will also be able to explain its prediction so that the user can fully comprehend the rationale behind such a decision. In this paper, we will describe the architecture of the proposed system along with a detailed plan for verification. The paper also presents the state-of-the-art of AI applications in flexible manufacturing to establish how CHAIKMAT project aims to apply some of the novel AI methodologies to circumvent the existing gaps.

Arkopaul Sarkar, Muhammad Raza Naqvi, Linda Elmhadhbi, Dusan Sormaz, Bernard Archimede, Mohamed Hedi Karray
Preliminary Manufacturing Cell Design in Digital Factory

A design methodology is presented leading from product family specifications and estimated demand to basic 3D models of the corresponding manufacturing cell. Key steps involve mapping product specifications to alternative generic process plans, then mapping processes to machine tools by comparing their capabilities to product specifications. Optimization of machine layout is carried out by a standard genetic algorithm. The method is supported by geometric libraries of simplified yet parametric machine tool models. An anemometer and wind vane production case study is presented.

G.-C. Vosniakos, X. V. Gogouvitis, E. Panos
Design and Implementation of a New Layout in a New Production Area of a Cork Stopper Factory Following Lean Manufacturing Principles

The current market conditions and purchasing power put pressure on selling prices. To maintain corporate profits, it is important to increase manufacturing efficiency and lower costs. The industrial unit in which this project was developed aims to increase the annual production of cork stoppers and by doing so, maintain the high standards of the quality and accuracy in their processes. To respond to this growth, Lean Manufacturing was implemented, and it was decided to build a new production area. The goal is to create a new production area for cork stoppers capable of producing around half a million cork stoppers per day. The use of Total Flow Management (TFM) tools allowed for the design of the layout and to level production. With the construction of a continuous flow between processes, the storage areas were gradually reduced, and the volume of WIP stock decreased by 63%. The results have also shown a decrease in the complaint rate of 3.5%, which was reduced under the established threshold of 5%, fulfilling the objectives proposed by the company for the project.

Sofia Amorim, Leonel Nunes, Carina Pimentel, Radu Godina, João C. O. Matias
On the Necessity for Digital Transformation in Agriculture Supply Chains: A Review from Task, Organization, Behavior, and Application Perspectives

Digital transformation has taken place in many industrial sectors to gain leverage by the technology implementation or to transform the whole organization or its parts. The digital transformation has grown to become popular, especially in agriculture supply chains related research. Research journals have reported more digital technology applications in the area, with some of the noted technologies such as the Internet of Things (IoT), big data, machine learning, and blockchain. Nonetheless, what structured principles and approaches to champion the digital transformation remained an open question. The process itself is not bound only to the technological aspects. Thus, this research reviews the transformation task definition, organization, behavior, and application aspects of the digital transformation within the agriculture supply chains to attain a practical and pragmatic perspective. The researchers performed a comprehensive literature survey to investigate the possible insights, current implementation state, key issues, and the necessity for further research. The overall findings reveal that the digital transformation is a dynamic process of fusing the technology into the agriculture supply chains. Also, the IoT, blockchain, and data mining technologies play a significant role in implementing the digital transformation in agriculture supply chains. Finally, a generic work for agriculture supply chains’ digital transformation process is proposed at the end of this study.

Roy Deddy Hasiholan Lumbantobing, R. M. Chandima Ratnayake, Togar Mangihut Simatupang, Liane Okdinawati, Nur Budi Mulyono
Diagnosis and TQM Strategies for Improving the Organizational Efficiency of a Consulting Company

Total Quality Management (TQM) is useful in both the industrial and service sectors. The purpose of this study was to increase profitability at a management consulting company by applying and improving internal operational excellence. This was accomplished using a case study. In order to overcome the problems, a prioritized plan of actions was developed by identifying the main sources of improvement. TQM strategies and the requirements of the ISO 9001: 2015 standard are followed. It was possible to observe some benefits on efficiency outcomes and small changes can provide to a company through large steps towards greater results. In a long-term perspective and considering that the performance and efficiency of the commercial has increased.

Eliana Pimenta, Carina Pimentel, Radu Godina, Susana Garrido Azevedo, João C. O. Matias
Virtual Supply Chain Network Platform Design and Development for Crisis Response

The COVID-19 pandemic exposed the vulnerability of the Canadian economy on many fronts. When the demand for lifesaving equipment increased globally, the supply chain networks were broken by the direct involvement of other countries. The rising competition and interruptions caused Canada to face significant difficulties in global markets to secure critical medical equipment and protective materials. Not only hospitals and healthcare workers but also the public and patients had no access to the needed equipment even though companies and organizations in the country have the required capacity and resources. In such emergency times, Canada should produce the essential equipment within the country. We propose a four-step strategic product manufacturing system to ensure crisis response. The first and second steps are creating a manufacturing capability database of Canadian companies and a library of product families, respectively. These two steps should be completed before the crisis. The third step involves emergency need analysis, equipment design and forecasting. Finally, the fourth step is developing a virtual supply chain network platform through which the procurement, production, and transportation activities will be scheduled based on the capability database, product families library, and requirements analysis in the most efficient and economical way possible. The research utilizes various tools such as forecasting, optimization, simulation, multi-criteria decision making, and engineering design tools.

Basak Tozlu, Ali Akgunduz, Yong Zeng
Dynamic Computer-Aided Process Control with Computer Vision for Industry 4.0

Because of health concerns and factory operational scale backs during the recent COVID-19 pandemic, we now need factory automation more than ever to maintain our productivity. However, most of our factories cannot operate remotely, and none can function without considerable human input and oversight. Trying to automate our factory highlights gaps in our technology, as it seems far behind our expectations, needs, and vision. Thus, this paper aims to fill this gap by showing how we have developed practical methodologies and applied technology to enhance legacy factories and their equipment. Specifically, we present the ORiON Production Interface (OPI) unit to run the factory as a smart networked edge device for virtually any machine or process. We have also implemented various computer vision algorithms in the OPI unit to detect errors autonomously, make decentralized decisions, and even control the quality. Although Industry 4.0 is a known concept to equip our factory to see, understand, and predict, we know that many machines today are closed source and cannot even communicate, let alone join a network. This research provides a workable solution to realize Industry 4.0 truly in existing factories with legacy equipment. Experimental results show that this system has a variety of applications, including process monitoring, part positioning, broken tool detection, etc. This novel intelligent networked system can enable our factories to be more innovative and responsive. It also allows for remote operations that can be unattended or lightly tended—a trend needed for the future.

Tsz Ho Kwok, Tom Gaasenbeek

Enabling Technologies

Combination of Data-Driven and Physics-Based Models for Thick Sintered Electrode Lithium-Ion Batteries

This work explored the use of physics-based and machine learning models in the context of a promising battery material system. The battery materials system was electrodes comprised of only electroactive material, which provides increased energy density at the cell level. The overall target of the modeling platform is to develop tools to aid in accelerating the experimental material discovery process. Machine learning models provide a route to more accurately predict the experimental electrochemical capacity of the materials in battery cells, although appropriate data training sets are needed. The combined application of the physics-based model and machine learning model resulted in the most accurate prediction of electrochemical cell outcomes.

Chen Cai, Shengyu Liu, Ziyang Nie, Kyoung-Yun Kim, Gary M. Koenig Jr.
Perceptions of a Digital Twin Application Case in the Auto Industry

Reality shows that, despite promises to facilitate the analysis of manufacturing systems, the use of state-of-the-art tools and techniques may become a challenging effort. Reliable data can drive significant analyses to help companies understand underlying issues and plan actions to improve processes. Digital Twins (DT) are models that could monitor production parameters and possibly run cause-and-effect-like investigations and future project events once fed with real-time and reliable data. In an attempt to bring light to recurring issues present in the daily lives of those who work on the implementation of Industry 4.0 projects in production lines, this article presents insights obtained from the case of a Digital Twin model fed with near real-time in the auto industry. It also proposes a minimum structure necessary for capturing, reading, and sending data and presents users’ perceptions from several functional units within a given company to build and test the developed model. Preliminary results show that companies should prepare for unexpected problems and limitations that span from the inadequacy of legacy hardware to obstacles related to human behavior in real-life implementation projects.

Suewellyn Krüger, Saulo Blan dos Santos, Milton Borsato
Machine Learning for Forecasting and Predicting Failures in Lithium-Ion Batteries

The adoption of Lithium-ion batteries (LIB) is increasing in many different sectors such as electric mobility, electronics products, and smart grids. This trend happens because LIBs are more stable, have a longer life cycle, and store more energy than other technologies. However, a couple of failures in the LIB can cause safety issues and degradation, mainly because these failures can happen slowly and progressively. Therefore, it is essential to diagnose, forecast, and predict failures to take action as soon as possible. Machine Learning (ML) techniques are commonly used to work with data-driven and diagnose failures in industrial machines to solve this problem. Thus, these tools also can be used in LIB products. Due to this motivation, this work developed an ML system with some different models in two steps. Firstly, the algorithm forecasted the central values from the LIB. Then, the output of the first step fed the second layer to predict and diagnose future failures. The work results indicate that a failure can be anticipated a long time before would be identified by the LIB management. Therefore, this approach can avoid failures and reduce the aging process.

Joelton Deonei Gotz, João Felipe Raffs Espolador, Gabriel Carrico Guerrero, Samuel Henrique Werlich, Milton Borsato, Fernanda Cristina Corrêa
Quality Control in Remanufacturing: Distinguishing Features and Techniques

Considering the existing amount of work on quality control in manufacturing companies, this work analyses its applicability to remanufacturing systems. After reviewing different quality control methods, we describe the distinguishing features of remanufacturing compared to manufacturing systems, such as increased variability of key remanufacturing inputs; the definition of customer requirements was done by a third party; customer requirements may be obsolete; and key product/process characteristics may need to be updated. Then, we conclude that existing quality planning and control techniques can be transferred to the remanufacturing domain. It ends by suggesting ways to assess the resulting quality control based on quality costs and identifies future research directions.

Sérgio D. Sousa, Duc T. Pham
Technological Insights of Interoperable Models for Integration of CAD/PLM/PDM and ERP Modules in Engineering Change Management

With the advancement of the technology, the computer aided design (CAD) has encountered developments which require the collaborative plan and the circulated configuration. The increasing necessity and intricacy of the product requires a new approach strategy to fulfil the worldwide market. The new collaborative approach can work for the designing party to share the entire work process with global market. This collaborative system can simplify the work of designer with suppliers, manufacturers, and clients across ventures. This paper has focused on the concept of Enterprise Resource planning (ERP) and product data management (PDM), with CAD packages. Insights are proposed to enable the integration of CAD/PDM with ERP. The insights propose a conceptual framework includes an architecture that will enables the collaboration of all modules simultaneously. For successful integration, all systems have been considered to have bidirectional information exchange. The framework enables the tracking of changes in Product design in other modules. This enables the quick response for production planning and manufacturing.

Onkar Bhaskar Kadam, Amir Pirayesh, Omid Fatahi Valilai
Using Blockchain Technology for 3D Printing in Manufacturing of Dental Implants in Digital Dentistry

In recent years, the healthcare and health industries have seen revolutionary changes. With the advancement of manufacturing technologies, the health industry has adjusted its paradigm to provide patients with the highest and most minimally intrusive, new treatment alternatives. Following its inception in therapeutics and medicine management, Three-dimensional (3D) printing technology has evolved into a vital tool for the medical practitioners. 3D printing enables the creation of made-to-order models, implants, prosthesis, and specialized tools for a variety of different fields from health education to prosthetic regeneration and restoration shaping the digital dentistry paradigm. In digital dentistry, components are custom-made for each patient in small batches, and the finished products cannot be utilized on other patients. However, the spread of the idea of digital dentistry presents some challenges for the actors of the dental industry such as specialists, designers and manufacturers knowing that they have common obligations with regards to treatment definition and regulations. This paper has proposed the idea of using the Blockchain technology to tackle the aforementioned challenges of 3D printing in the dental industry. The proposed framework enables the data and resource sharing in a secure platform and utilizing the prosthetic rebuilding efforts and CAD/CAM features like laser sintering and 3D printing capabilities through an integrated network.

Sahil Sachin Shah, Amir Pirayesh, Omid Fatahi Valilai
Digital Twin Based Definition (DTBD) Modeling Technology for Product Life Cycle Management and Optimization

Model Based Definition realizes the unified transmission of information between digital product design and manufacturing process, but it cannot provide support for the full life cycle management of Digital Twin products. This paper proposed the Digital Twin Based Definition (DTBD) technology, and the definition, the application principles, the composition and characteristics, the way of interaction, the system hierarchical relationships, the division of sub-models and the collaborative relationship between them, are described in detail. The main procedures of developing DTBD model are presented. The model is built on MBD, and the evaluation of processing mode and efficiency of each module of the model will be obtained in the simulation system, which will be used as the basis for optimal design. The product operation data collected by sensors is processed by multiple information fusion algorithm to form the data group which is the basis for updating the Digital Twin attribute of the model to the actual operation state of the product. The operational status data of the models is shared with industrial Internet technology, and to acquire the long-life model operating conditions, the differences between model states are analyzed by data mining algorithm. Product fault characteristics obtained by multi-scale information extraction algorithm are fed back to the data list of the model for design optimization, and to avoid similar failures, fault data is shared. DTBD as an enabling technology improves MBD, and promotes the development of Digital Twin and intelligent manufacturing processes which cover design, processing, optimization and life cycle management.

Zhongyuan Che, Chong Peng, Zhongwen Zhang
A Hybrid Architecture of Digital Twin with Decision Support Layer for Industrial Maintenance

Decision-making has been an important field of study in the context of industrial maintenance, as the methodologies disseminated in this field aim to support the operators to choose the most suitable decision for the maintenance process. When subjects related to Industry 4.0 (i4.0), digitization and digital twin are included in the analysis, the complexity in decision-making increases due to the large set of variables that act directly in the process. Thus, it has been noted that the Multi-Criteria Decision-Making methods (MCDM) allied with Digital Twin (DT) concept have the potential to increase operational efficiency to select the most adherent actions to the process under this set of data and variables. Therefore, this paper aims to present a novel framework that merges the MCDM and DT concepts to support the daily tasks of decision-makers, performing criticality analysis in discrete manufacturing in the context of industrial maintenance. A framework of digital twin with a decision support layer for industrial maintenance is presented and tools are highlighted seeking the integration of multiple concepts and technologies with a focus on improving the shop floor operation and, consequently, facing and solving industrial maintenance decisions.

Cleiton Ferreira dos Santos, Rolando Jacyr Kurscheidt Netto, Ricardo Eiji Kondo, Eduardo de Freitas Rocha Loures, Eduardo Alves Portela Santos, Anderson Luis Szejka
Framework for Reducing the Complexity of Programming Robot Skills

In the field of manufacturing, modern approaches like skills based programming are applied to robot controllers using well adopted standards for communication like the Open Platform Communications-Unified Architecture (OPC UA) [1] standard. The current research focuses at defining OPC UA models to enable skill based programming for robot controllers. This defines a complex environment for a process programmer, who has to implement a manufacturing process represented as a skill at a high level domain. This work attempts to reduce the programming effort of implementing a skill by reducing the complexity for programming the process executed by the robot. This is achieved by introducing a framework assuming a little to no prior knowledge of OPC UA. This work will verify the reduction of complexity by analysing the code complexity and by analysing the knowledge domains used.

Michael Hofmann, Matthias Propst, Markus Ikeda, Andreas Pichler, Fabian Spitzer, Roman Froschauer
Visualization Concept for Representing Capability Matchmaking Results in a Virtual Environment

The traditional production system design and reconfiguration planning are manual processes. The lately developed capability matchmaking system aims to improve the production system design with a more intelligent design approach that automates the search for feasible resource combinations to specific product requirements. Virtual reality concepts and virtual manufacturing can bring more immersivity, perceptual intuition and interaction to the design process, and thus speed it up. 3D graphical visualizations of a production system and its resources can help in identifying problems in the reconfiguration of industrial equipment. The result from existing capability matchmaking system in XML format is not intuitive for the designer. Additionally, it is very difficult to analyze the proposed resources based on the textual description. To enhance the efficiency and performance of the existing capability matchmaking system, especially how to present and visualize the possible resource combinations inside the result is seen as an essential step towards virtual and smart manufacturing. This research provides an approach to visualize the result of capability matchmaking system in a virtual simulation environment with a use case example.

Rongwei Ma, Minna Lanz, Niko Siltala
Joint Industrial Preventive Maintenance and Production Scheduling: A Systematic Literature Review

Maintenance plays a fundamental role in the efficiency of productive processes, contributing to cost reduction, operating safety, and compliance with environmental constraints. From the standpoint of Preventive Maintenance (PvM), a disconnection can be observed in the industrial context between inspection intervals and production schedule programming. In this context, the current research effort has been in correlating the maintenance schedule and production planning. An exploratory survey in this problem space was undertaken by way of a systematic literature review. Current approaches face barriers for their applied usage in shopfloor, as the computational tools required for their optimization models and extraction information from process. In order to provide support for the gaps identified in current approaches, a framework is proposed that, using information extracted from event logs, generated from process and assets, and processed by process mining algorithms, integrates production demands, process behavior and asset status information. In the form of criteria assessed over the course of time windows alternatives, the ideal moment to stop production and perform the appropriate maintenance actions is established. Thus, a viable alternative is achieved to obtain, objectively and based on process information, the parameters required for integrating the maintenance scheduling with production planning.

Rolando Jacyr Kurscheidt Netto, Eduardo de Freitas Rocha Loures, Eduardo Alves Portela Santos, Cleiton Ferreira dos Santos
Quality Control of Die Castings by Machine Vision: A Case Study Exploiting Classic and Machine Learning Techniques

The aim of this work is to demonstrate that part dimensional inspection and defect detection in pressure-die castings is perfectly possible by exploiting image capturing hardware of reasonable or low cost accompanied by selecting, configuring and synthesizing already mature machine vision and machine learning algorithms rather than developing them from scratch. Reasoning for dimensional accuracy is based on edge identification and is purely algorithmic in nature reaching 100% accurate results. Reasoning regarding crack detection is based on regions of interest and it exploits Artificial Neural Networks, results being promising given the low cost equipment and small data set used.

G.-C. Vosniakos, E. Manou
The Importance of Digital Readiness on Manufacturing SMEs’ Performance Aiming Towards Industry 4.0: A Case Study

It is increasingly advocated by authors that Industry 4.0 is the way forward to retain market competitiveness. However, the high failure rate of digital technologies’ implementation seems to fuel the slow progress of manufacturing SMEs to fully engage in its digital transformation. This phenomenon seems to come from a generalized absence of organizational changes which are necessary for the successful deployment of digital technologies. There seems to be a blur in the literature between digital maturity and digital readiness which deserves to be addressed. This paper targets the technical, technological, and organizational prerequisites to be implemented in manufacturing SMEs. The most frequently cited prerequisites in the literature are: Lean, agility, knowledge management, data accessibility, business strategy, dedicated financial resources, and cybersecurity. The objective of this paper is to study the effect of these prerequisites on the overall performance of a manufacturing company (production volume, manufacturing time, turnaround time, and reaction time) in order to prioritize them in a SME context. The prioritization is demonstrated by a Design of Experiments and a Monte-Carlo simulation, both of which were based on real data from a Quebec manufacturing SME. The results show that Lean offers a much greater potential than the other prerequisites. Additionally, business strategy, if not supported by the other prerequisites, has very little impact on the whole system. Several of the prerequisites observed during the simulation were implemented in the field to validate their impact in a real-life setting, which made it possible to support the conclusion of the study.

M. Charbonneau-Genest, Sébastien Gamache
Smart Particle Sensing Via Computer Vision for Precise Dedusting with Water Monitors

Water monitors are used for dedusting tasks in various working areas with heavy dust emissions. They are an important part of the safety infrastructure to prevent wear of the machinery and safeguard health. State of the art solutions are designed as simple mechanical set-ups that throw nebulized fluids unspecifically. Moreover, these solutions are characterized by a target conflict between power, coverage and dynamics that limit the operating range. Considering this, the branch still lacks smart technologies for operating such fluid systems efficiently regarding resource deployment, i.e. water consumption, and efficacy. From an explorative research design a specific use case was derived and led to a cooperative research project, where a prototype was developed, empirically examined in laboratory scale, and yet applied in exemplary working grounds. Key feature is a meshed computer vision setup supported by a linked microcontroller and computer unit. It provides real-time operating data from a water monitor and it`s surrounding for condition-based actions. By this, the data enables both for online detection of current dust clouds and forecasting them. As a result, the system automatically adapts its range of throw as well as the nebulization. Moreover, an AI considers further factors, as for instance sedimentation velocity and specific surface of the spread fluid, in order to react in the most effective way. The digitized set-up enables for a broader range of operations, other than present analog solutions. Considered as a safety enhancing solution, further research will be needed in long-term field studies regarding data rates and computing requirements.

Dr.-Ing. Dennis Bakir, Florian Engels, Robin Bakir
A Systematic Literature Mapping on the Process Reconfiguration of Smart Manufacturing Systems with the Integration of Multi-criteria Decision Models and Ontology Based Interoperability

Sector 4.0 technologies, such as cloud computing, big data, the internet of things, and cyber-physical systems, are transforming the manufacturing industry significantly. Companies must concurrently address consumer expectations and production capacities, resulting in an increase in system requirements. This growth in requirements necessitates the ongoing modification of the production characteristics’ structures, culminating in the development of Smart Manufacturing Systems that enhance the client value proposition. As market needs change over time, the system's capabilities must adjust to these changes, which may lead to issues such as an increase in production time, cost, and a decline in the quality of the final product. To tackle this problem and help in the decision-making process, assessment techniques may be used to examine and offer several alternatives depending on the current capabilities and needs. In addition, these models may be adopted and linked with ontological techniques since they can effectively represent and communicate information across multiple systems and domains, hence enhancing the accuracy and efficacy of decision-making. In this context, the primary objective of this study is to build a thorough literature mapping of current research on the Smart Manufacturing Systems decision-making process.

Matheus B. Canciglieri, Athon F. S. de M. Leite, Eduardo F. R. Loures, Anderson L. Szejka, Osiris Canciglieri, Yee M. Goh, Radmehr P. Monfared, Giovana Regina Dal Negro Martins
Cloud Manufacturing Services Adoption in Higher Education Institutions: Challenges and Framework for Developing Countries

Developing countries are lagging in the adoption and implementation of emerging technologies built on cloud computing. Information technology has been identified as being crucial for business activities in the 21st century, leading to rapid adoption and implementation in developed countries. However, regardless of globalisation, there exist various organizational and systematic challenges in acquiring cloud computing services for research purposes in higher education institutions in developing countries. Research institutions work in collaboration with industry, as beacons of industrialisation in improving work and productivity. The aim of this paper is to illustrate challenges and barriers towards the adoption of cloud computing services, and development of an agile framework for research institutions in developing countries. Multi-criteria decision-making method was utilised for selection of cloud resources, which were mapped on researchers end goal. The result is a cloud-based microservice architecture for a learning environment at institutions of higher education, which independently executes multiple research projects, with the transport manufacturing sector as a case study.

Alice Elizabeth Matenga, Khumbulani Mpofu, Olukorede Tijani Adenuga
Facilitating Trust in Food Supply Chains Through Blockchain Technology: A Systematic Review of Considerations for Alternative Food Networks

Trust is an important concept in our food systems. Global food systems are witnessing challenges regarding trust. Alternatives to global food systems, generally referred to as alternative food systems, emphasise trust through local and short food initiatives, commonly referred to as alternative food networks. Alternative food networks can benefit from the trust characteristics inherent in digital technology advances such as blockchain. A key characteristic of blockchain is trust. Knowing how blockchain trust characteristics and trust mechanisms can be used to facilitate trust in the supply chains of alternative food networks will be helpful. This paper presents a systematic review of the literature and an understanding of how trust is captured in architectures of blockchain-enabled food supply chains and highlights possible extensions towards facilitating trust in the supply chain of alternative food networks. A distinction is made between the trust characteristics inherent in blockchain and those that apply to alternative food network supply chains. A framework to integrate trust enabled by blockchain into supply chains of alternative food networks is proposed. Areas of future work are recommended, including a need to validate the proposed framework.

Patrick Burgess, Funlade Sunmola, Sigrid Wertheim-Heck
Intuitive Robot Programming by Capturing Human Manufacturing Skills: A Framework for the Process of Glass Adhesive Application

There is a great demand for the robotization of manufacturing processes featuring monotonous labor. Some manufacturing tasks requiring specific skills (welding, painting, etc.) suffer from a lack of workers. Robots have been used in these tasks, but their flexibility is limited since they are still difficult to program/re-program by non-experts, making them inaccessible to most companies. Robot offline programming (OLP) is reliable. However, generated paths directly from CAD/CAM do not include relevant parameters representing human skills such as robot end-effector orientations and velocities. This paper presents an intuitive robot programming system to capture human manufacturing skills and transform them into robot programs. Demonstrations from human skilled workers are recorded using a magnetic tracking system attached to the worker tools. Collected data include the orientations and velocity of the working paths. Positional data are extracted from CAD/CAM since its error when captured by the magnetic tracker, is significant. Paths poses are transformed in Cartesian space and validated in a simulation environment. Robot programs are generated and transferred to the real robot. Experiments on the process of glass adhesive application demonstrated the intuitiveness to use and effectiveness of the proposed framework in capturing human skills and transferring them to the robot.

Mihail Babcinschi, Francisco Cruz, Nicole Duarte, Silvia Santos, Samuel Alves, Pedro Neto
A Probabilistic Model to Estimate Automated and Manual Visual Inspection Errors

Primary risks associated with visual inspection are missing a defect (Type II error) and identifying a defect that is a false alarm (Type I error). The focus of this paper is on defect detection on cylinder head surfaces using optical inspection techniques. Several factors were included in the study for their impact on defect identification: lighting, lens angle, camera distance, defect dimensions, and defect accessibility. Using the influence diagram technique, we used quantified uncertainties to determine how these factors affect Type I and Type II errors. For some defects, there can be significant variation in inspection performance results due to subjective judgment of inspectors. We also analyzed the reduction in Type II errors using a machine learning methodology to identify and classify defects. The influence diagram model is populated with data from an experimental study and subject matter expert (SME) input. Presented results may be used to identify and assess factors that can influence success of hybrid inspection processes and mitigate associated risks of Type I and Type II errors.

Pallavi Dubey, John Jackman, Gül E. Kremer, Paul Kremer
Industrial Maintenance and the Digital Twin—An Architectural Assessment

The fourth industrial revolution is characterized by intelligent factories that connect the different elements of the production process such as machines, systems and people, integrating them vertically and horizontally, in order to improve production logistics, use of resources, reduce production defects and use of raw materials. The information flow emerges as the main engine of this industrial revolution and its most important asset. The maintenance sector, from this informational perspective, appears as one of the areas with the greatest impact potential in the adoption of industry 4.0-oriented technologies, playing a critical role in sustaining an organization's operations. The literature points to different architectures aimed at implementing the functions of industrial maintenance and which present an evolutionary orientation to meet the requirements of Industry 4.0. This evolution, currently, is directed towards the Digital Twin concept, which provides a new paradigm for the redesign of architectural functions and components with greater focus on simulation, data analysis, prediction function and decision making. The purpose of this article is to evaluate the different architectures aimed at industrial maintenance under the requirements of Industry 4.0 and provide a guiding position for them in relation to the Digital Twin.

Alexandre Helmann, Anis Assad Neto, Fernando Deschamps, Eduardo de Freitas Rocha Loures
A Passive Compliant Gough-Whitehall-Stewart Mechanism for Peg-Hole Disassembly

The insertion of a cylindrical peg into a cylindrical hole is often used by robotics researchers as a model for studying assembly operations. This is because many assembly operations can be represented as the mating of a male cylindrical object into a matching female object. The device was an inverted Gough-Whitehall-Stewart mechanism where the six legs were springs instead of actuators. This paper presents a modified version of the device where the legs do not meet in pairs at the platform but at points located remotely from it. This gives the device the properties of a remote-compliance-centre (RCC) mechanism, which has been shown to be effective for precise peg-hole assembly tasks. However, unlike currently available RCC mechanisms, which can only withstand high compressive forces, the proposed compliant device can resist both compressive and tensile forces, which makes it applicable to assembly as well as disassembly operations. Using small motion approximations, the paper derives the compliance matrix of the mechanism and determines the location at which it is diagonal, proving that the compliance centre is situated away from the platform. A sensitivity analysis has confirmed the correctness of the small motion assumptions and the RCC properties of the new compliance device.

Joey Lim, D. T. Pham
Taxonomy of Manufacturing Joining Operations Based on Process Characterization

Depending on the complexity of the assembly design and required production constraints, factories employ various types of joining operations as part of product fabrication. Manufacturers, who are in the business of assembling, gain their competence based on what types of processes and resources they use as well as how they are used in these joining operations. For data interoperability and exchange among the partners of distributed manufacturing, these joining operations need to be described formally to build a common set of vocabulary. Current ontologies in the related topics lack the details of the process characterization in their analysis and do not adopt foundational concepts to build such definitions. This paper presents an ontology-driven characterization of the joining operations to formalize a set of definitions on ontologically grounding concepts towards the construction of a taxonomy of the joining operations.

Arkopaul Sarkar, Dusan Sormaz, Mohamed Hedi Karray
Digital Multiphase Material Microstructures for Image-Based AI Methods

The availability, cost, and producibility of components using additive manufacturing (AM) have changed how parts are designed and manufactured. These changes have highlighted the need for real-time control of production parameters to adjust product quality and to reduce or eliminate post-processing. Improved quality can be achieved by controlling the as-produced microstructure. The need for improved part quality is particularly pressing for aerospace products made from titanium alloys, due to their high material and production costs. Since mechanical properties and material microstructure are closely related, microstructure prediction for additively manufactured products has been intensely pursued. Titanium alloy microstructures with a low ratio of columnar grains and a high ratio of more fatigue-resilient equiaxed grains are preferable for aerospace applications. In addition to controlling microstructure, improved process control can mitigate the formation of porosities and residual stresses during part fabrication. The objective of this research is to provide a fast method to produce digital multiphase microstructures that will support an AI (artificial intelligence)-based deep learning model for predicting the microstructure of a multiphase Ti6Al4V alloy produced by powder bed fusion (PBF). Digitally developed microstructures, through polycrystalline generation, can be used as training samples for a deep learning algorithm (DLA).

Eray Aksit, Karl R. Haapala, Ali Tabei
Integration of Mixed Reality (MR) and Structural Analysis Towards Industry 4.0

One of the core engineering activities during the design and development of new products, components, and structures is the structural simulation in order to validate the design and ensure that the parts will efficiently withstand the loading cycles. On the other hand, simulation is one of the nine pillar technologies for Industry 4.0. Under the current industrial revolution, engineers have focused on improving the Machine-2-Machine (M2M) communication extending to the development of suitable techniques for achieving greater interoperability. Moreover, the development of cutting-edge digital technologies, such as Mixed Reality (MR) has also been focused in order to facilitate the so-called Human–Machine Interface (HMI). Therefore, the contribution of this research work is based on the design and development of suitable framework for the integration of structural simulation tools with MR, in order to facilitate engineers setup simulations and visualize the results in a more robust and intuitive manner. The proposed framework can be realized as a Cloud platform, which supports the distribution of several services such as storage of CAD files, simulation results, and the MR application for the setup and visualization of the experiments.

Dimitris Mourtzis, John Angelopoulos, Nikos Panopoulos
Deep Learning-Powered Visual Inspection Using SSD Mobile Net V1 with FPN

There is an increasing demand to automate manufacturing inspection processes that can be time-consuming and subject to the expertise of individual inspectors. While there are advantages to having humans assess the quality of a part versus automated vision systems, such as greater flexibility of the inspector to adapt to new parts, this approach is also prone to errors. This study examines the capability of optical inspection techniques to reduce inspection errors in remanufacturing. We implemented an SSD Mobile-Net Algorithm that uses depth-wise separable convolutions to build lightweight deep neural networks and Feature Pyramid Network (FPN) to enhance feature extraction. The algorithm was evaluated using the GC10-DET benchmark data. The algorithm addressed the extreme imbalance between common and uncommon defect samples. The algorithm’s performance is compared with other object detection algorithms using average precision and mean average precision (mAP) metrics. Our model outperforms five class categories compared to other state-of-the-art models, especially in small size defect and class imbalance categories. The importance of data pre-processing is also discussed, including improving data quality while keeping the training model constant.

Pallavi Dubey, Elif Elçin Günay, John Jackman, Gül E. Kremer, Paul Kremer
Application of Anomaly Detection Algorithms in Lithium-Ion Battery Packs - A Case Study

Lithium-Ion batteries (LIB) store energy for many different applications, especially in the mobility and smart grid areas. LIB has several advantages like stability, longer lifetime, and capacity compared with other technologies. Although, LIB can be dangerous, mainly if it operates in unsafe conditions due to failures that can appear and cause accidents like explosions or fire. To maintain the safe operation, an electronic system named Battery Management System (BMS) manages and controls the main parameters to guarantee the safe operation of the LIB. Therefore, BMS collects and controls the main parameters of the LIB like the voltage, current, and temperature. Despite that, BMS has challenges predicting, preventing, and identifying unforeseen failures. How a failure can be considered an anomaly, data-driven models can identify a precocious potential failure. For this reason, this paper presents an anomaly detection system to identify an overheating failure in a LIB as earlier as possible. A LIB prototype was built to simulate the overheating failures, and three anomaly detection algorithms have been applied to drive the work. An edge-cloud computing architecture collected and stored the needed data in a dataset to elaborate the idea and demonstrate the excellent results to anticipate failures in LIB. For future works, the intention is to operate with this approach in online embedded in the vehicle with the BMS to detect a failure precociously.

Joelton Deonei Gotz, Gabriel Carrico Guerrero, João Felipe Raffs Espolador, Samuel Henrique Werlich, Milton Borsato, Fernanda Cristina Corrêa
Prediction of Disc Cutter Replacement of Tunnel Boring Machine Using Denoising Auto Encoder

While a tunnel boring machine (TBM) is working, rocks are crushed into pieces by disc cutters which often fail during construction. To replace disabled cutters timely, the condition of the cutters needs to be checked regularly. However, this is time-consuming and uneconomical. In this paper, a denoising auto encoder (DAE) model is proposed to judge whether TBM disc cutters need to be replaced. First of all, the field data related to dis cutter status are selected as inputs. Then, the cutter conditions can be learned automatically base on a DAE model. Test result on a water transport tunnel shows that the proposed model can obtain an average accuracy of 99.7% and an average F1 score of 99.4% on field data prediction. Compared with other machine learning and deep learning models, proposed method reduces the need of manual data denoising and feature extraction.

Yang Liu, Shuaiwen Huang, Di Wang, Guoli Zhu, Dailin Zhang
Flexible Automation and Intelligent Manufacturing: The Human-Data-Technology Nexus
Kyoung-Yun Kim
Leslie Monplaisir
Jeremy Rickli
Copyright Year
Electronic ISBN
Print ISBN