To meet the demand of future building requirements, and to improve productivity and competitiveness, there is a need to modernize and revise the current practices in the wooden single-family house industry. In several other sectors, intensive work is being done to adapt to the anticipated fourth industrial revolution. The manufacturing industry has already begun its transformation with concepts such as smart manufacturing and Industry 4.0. So far, smart manufacturing has not been discussed to any significant extent for the wooden single-family house industry, even though it might be a way for this industry to improve productivity and competitiveness. The research presented in this thesis aims at increased knowledge about what smart manufacturing means for the wooden single-family house industry. This requires investigating what smart wooden house manufacturingis, what challenges that might be associated with it, and how smart wooden house manufacturing can be realized. At the core of this thesis is the conceptualization of smart wooden house manufacturing—when realized, it is expected to contribute to improve the competitiveness of the wooden single family house industry. The findings presented here are based on three Research Studies. Two studies were case studies within the wooden single-family house industry. The third study was a traditional literature review. The findings revealed two definitions and 26 components of smart wooden house manufacturing. At large, smart wooden house manufacturing emphasizes digital transformation with a focus on digital information flow, how to add information, information compilation, and information distribution between systems/programs and departments. Some of the challenges associated with smart wooden house manufacturing are, e.g. culture, competence and manual transfer of information between systems. The findings indicate similarities of smart wooden house manufacturing within certain components of industrialized house building and Industry 4.0, these components could enable the realization of smart wooden house manufacturing. / För att möta efterfrågan på framtida byggkrav och för att förbättra produktiviteten och konkurrenskraften finns det ett behov av att modernisera och revidera nuvarande tillvägagångssätt inom träsmåhusindustrin. I flera andra sektorer arbetas det intensivt med att anpassa sig till den förväntade fjärde industriella revolutionen. Tillverkningsindustrin har redan påbörjat sin omvandling med koncept som smart manufacturing och Industry 4.0. Hittills har smart manufacturing inte diskuterats i någon större utsträckning för träsmåhusindustrin, även om det kan vara ett sätt för denna industri att förbättra produktiviteten och konkurrenskraften. Forskningen som presenteras i denna avhandling syftar till ökad kunskap om vad smart manufacturing innebär för träsmåhusindustrin. Detta kräver undersökning av vad smart trähustillverkning är, vilka utmaningar som kan vara förknippade med det och hur smart trähustillverkning kan realiseras. Kärnan i denna uppsats är begreppsframställningen av smart trähustillverkning—när det realiserats förväntas det bidra till att förbättra konkurrenskraften för träsmåhusindustrin. Resultaten som presenteras här är baserat på tre forskningsstudier. Två studier var fallstudier inom träsmåhusindustrin. Den tredje studien var en traditionell litteraturstudie. Resultaten avslöjade två definitioner och 26 komponenter av smart träshustillverkning. Sammanfattningsvis betonar smart trähustillverkning digital transformation med fokus på digitalt informationsflöde, hur man lägger till information, sammanställning av information och informationsfördelning mellan system / program och avdelningar. Några av utmaningarna associerade med smart trähustillverkning är t.ex. kultur, kompetens och manuell överföring av information mellan system. Resultaten indikerar likheter mellan smart träshustillverkning inom vissa komponenter av industriellt husbyggande och Industry 4.0, dessa komponenter skulle kunna möjliggöra realiseringen av smart trähustillverkning.

Yes / Today’s factories are considered as smart ecosystems with humans, machines and devices interacting with each other for efficient manufacturing of products. Industry 4.0 is a suite of enabler technologies for such smart ecosystems that allow transformation of industrial processes. When implemented, Industry 4.0 technologies have a huge impact on efficiency, productivity and profitability of businesses. The adoption and implementation of Industry 4.0, however, require to overcome a number of practical challenges, in most cases, due to the lack of modernisation and automation in place with traditional manufacturers. This paper presents a first of its kind case study for moving a traditional food manufacturer, still using the machinery more than one hundred years old, a common occurrence for small- and medium-sized businesses, to adopt the Industry 4.0 technologies. The paper reports the challenges we have encountered during the transformation process and in the development stage. The paper also presents a smart production control system that we have developed by utilising AI, machine learning, Internet of things, big data analytics, cyber-physical systems and cloud computing technologies. The system provides novel data collection, information extraction and intelligent monitoring services, enabling improved efficiency and consistency as well as reduced operational cost. The platform has been developed in real-world settings offered by an Innovate UK-funded project and has been integrated into the company’s existing production facilities. In this way, the company has not been required to replace old machinery outright, but rather adapted the existing machinery to an entirely new way of operating. The proposed approach and the lessons outlined can benefit similar food manufacturing industries and other SME industries. / Innovate UK—Knowledge Transfer Partnerships (KTP010551)

Given recent advancements in technology and recognizing the evolution of smart manufacturing, the implementation of digital twins for factories and processes is becoming more common and more useful. Additionally, expansion in connectivity, growth in data storage, and the implementation of the Industrial Internet of Things (IIoT) allow for greater opportunities not only with digital twins but closed loop analytics. Discrete Event Simulation (DES) has been used to create digital twins and in some instances fitted with live connections to closely monitor factory operations. However, the benefits of a connected digital twin are not easily quantified. Therefore, a test bed demonstration factory was used, which implements smart technologies, to evaluate the effectiveness of a closed-loop digital twin in identifying and reacting to trends in production. This involves a digital twin of a factory process using DES. Although traditional DES is typically modeled using historical data, a DES system was developed which made use of live data with embedded machine learning to improve predictions. This model had live data updated directly to the DES model without user interaction, creating an adaptive and dynamic model. It was found that this DES with machine learning capabilities typically provided more accurate predictions of future performance and unforeseen near future problems when compared to the predictions of a traditional DES using only historic data

The manufacturing industry is continuously evolving as new practices and technology are adopted to improve productivity and remain competitive. There have been three well established manufacturing revolutions in recent history and some say that the fourth is occurring currently by the name of Smart Manufacturing, Indusrie 4.0, and others. This latest manufacturing revolution is highly dependent on industrial connectivity. This research aims to gage the ability of Industrial Innovation Platforms (IIPs) to reduce complexity of implementing base-line industrial connectivity for small and medium-sized enterprises (SMEs). The results of this study would be especially relevant to decision makers in industrial SMEs who are considering implementing industrial connectivity as well as providing insights into approaches for establishing base-line industrial connectivity. The research methodology consists of three main steps: 1) creation of IIP and non-IIP connectivity solutions that enable connectivity of the vast amount of industrial equipment, 2) Gathering measures from solutions in accordance with metrics identified for complexity evaluation, 3) discussion and interpretation of data To have a more complete analysis, quantitative and qualitative data was used and evaluated to address the varying elements of the broad task of establishing industrial connectivity. The research showed that IIPs can reduce complexity for select industrial equipment. Some industrial equipment have robust and streamlined connectivity solutions provided by the IIP. In these cases, the IIP almost certainly will reduce the complexity of establishing connectivity. Other industrial equipment have a solution provided by the IIP which requires piecing together and some component modifications. In these cases, the IIPs reduce complexity of establishing connectivity dependent on circumstances. Lastly, when no form of solution is available through the IIP for the industrial equipment, the IIP's has no ability to reduce complexity other than hosting the server used in connectivity. These findings open additional avenues of research which could improve the understanding of benefits IIPs may provide to SMEs.

Laser powder bed fusion is one of the most effective ways to achieve metal additive manufacturing. However, this method still suffers from deformation, delamination, dimensional error, and porosities. One of the most significant issues is poor printing accuracy, especially for small features such as turbine blade tips. The main reason for the shape inaccuracy is the heat accumulation caused by using constant laser power regardless of the shape variations. Due to the highly complex and dynamic nature of the laser powder bed fusion, improving the printing quality is challenging. Research gaps exist from many perspectives. For example, the lack of understanding of multi-physical melt pool dynamics fundamentally impedes the research progress. The scarcity of a customizable laser powder bed platform further restricts the possibility of testing the improvement strategies. Additionally, most state-of-the-art quality inspection techniques suitable for laser powder bed fusion are costly in economic and time aspects. Lastly, the rapid and chaotic printing process is hard to monitor and control. This dissertation proposes a complete research scheme including a fundamental physics study, process signature and quality correlation, smart additive manufacturing platform development, high-performance sensor development, and a robust real-time closed-loop control system design to address all these issues. The entire research flow of this dissertation is as follows: 1. This work applies and integrates three advanced sensing technologies: synchrotron X-ray imaging, high-speed IR camera, and high-spatial-resolution IR camera to characterize the melt pool dynamics, keyhole, porosity formation, vapor plume, and thermal evolution in Ti-64 and 410 stainless steel. The study discovers a strong correlation between the thermal and X-ray data, enabling the feasibility of using relatively cheap IR cameras to predict features that can only be captured using costly synchrotron X-ray imaging. Such correlation is essential for thermal-based melt pool control. 2. A highly customizable smart laser powder bed fusion platform is designed and constructed. This platform integrates numerous sensors, including but not limited to co-axial cameras, IR cameras, oxygen sensors, photodiodes, etc. The platform allows in-process parameter adjusting, which opens the boundary to test various control theories based on multi-sensing and data correlations. 3. To fulfill the quality assessment need of laser powder bed fusion, this dissertation proposes a novel structured light 3D scanner with extraordinary high spatial resolution. The spatial resolution and accuracy are improved by establishing hardware selection criteria, integrating the proper hardware, designing a scale-appropriate calibration target, and developing noise reduction procedures during calibration. Compared to the commercial scanner, the proposed scanner improves the spatial resolution from 48 µm to 5 µm and the accuracy from 108.5 µm to 0.5 µm. 4. The final goal of quality improvement is achieved through designing and implementing a real-time closed-loop system into the smart laser powder bed fusion platform. The system regulates the laser power based on the monitoring result from a novel thermal sensor. The desired printing temperature is found by correlating the laser power, the dimensional accuracy, and the thermal signatures from a set of thin-wall structure printing trails. An innovative high-speed data acquisition and communication software can operate the whole system with a graphic user interface. The result shows the laser power can be successfully controlled with 2 kHz, and a significant improvement in small feature printing accuracy has been observed. / Doctor of Philosophy / Laser powder bed fusion is one of the most effective ways to achieve metal additive manufacturing. However, this method still suffers from defects such as deformation, delamination, dimensional error, and porosities. Due to the highly complex and dynamic nature of the laser powder bed fusion, improving the printing quality is challenging. Research gaps exist from many perspectives, such as the lack of understanding of melt pool dynamics; the scarcity of a customizable laser powder bed platform; the need for suitable sensors; and the missing of a control system that can effectively regulate the rapid and chaotic printing process. This dissertation proposes a complete research scheme to address all these issues. The fundamental study characterizes the melt pool dynamics and discovers a strong correlation between the melt pool thermal and geometrical data, enabling thermal-based melt pool control. Following that, a highly customizable smart laser powder bed fusion platform is designed and constructed. The platform allows in-process parameter changes, opening the boundary to test various control theories. A novel structured light 3D scanner with an ultra-high spatial resolution was proposed to fulfill the quality assessment need. The final goal of quality improvement is achieved through designing and implementing a real-time closed-loop system into the smart laser powder bed fusion platform. The system regulates the laser power based on real-time thermal monitoring. The result shows the laser power can be successfully controlled with 2 kHz, and a significant improvement in printing accuracy is achieved.

Digitale Zwillinge können in verschiedenen Einsatzgebieten genutzt werden, insbesondere innerhalb einer digitalen Fabrik. Im Zuge verschiedener studentischer Projekte an der Smart Factory AutFab der Hochschule Darmstadt wurden state-of-the-art Software-Tools zur Implementierung verschiedener digitaler Zwillinge verwendet. Bei einer Linienplanung wurde das Potential für Brownfield-Anlagen deutlich. Verschiedenste Projekte haben mithilfe von digitalen Zwillingen virtuelle Inbetriebnahmen erfolgreich durchgeführt und inzwischen den größten Teil der Smart Factory abgedeckt. Über Materialfluss- und Energieverbrauchs-Simulationen konnten Engpässe in der Fertigung identifiziert werden, welche in einem Unternehmen zu großen Einsparungen führen können. Schließlich wurde ein neuer Weg von kollaborativer Entwicklungsarbeit gezeigt, welcher den Aufwand beim Aufbau von Systemen mit hoher Ähnlichkeit erheblich reduzieren kann.

<p>  </p> <p>Smart manufacturing is a broad concept of manufacturing technology that employs the computer aided systems, digital information technology, artificially intelligent algorithms, etc., to realize high-level automation of the production. The rise of the smart manufacturing concept, which has also been treated as the fourth industrial revolution, has been increasingly advocated by the policy makers and investigated by the worldwide researchers. Though machining is one of the key processes in the manufacturing industry, there are only a few researches focusing on automatically scheduling and improving the machining process. The design of the machining parameters and tool path planning still requires engineers with significant knowledge and experience in manufacturing fields to juggle between product quality, machine tool maintenance, and production cost. This design process also requires high level of human intelligence to consider the type of material, machine tool setups, workpiece geometry, and cutting tool property to provide an optimal manufacturing process. The overall machining related processes cannot satisfy the requirement of the ultimate goal of the smart manufacturing – to fully automate the machining process without human’s involvements.</p> <p><br></p> <p>In order to solve this problem, we aim to employ advanced machine learning technologies to enable the machine tool to automatically build up the cutting physics and generate the optimized toolpath. The final optimized result can be conducted automatically and shows a near human level optimization design ability. The generated toolpath beats the result from other commercial software. The overall framework can be fully automated when the machine learning technology is mature. </p>

Smart manufacturing is a revolutionary domain integrating advanced sensing technology, machine learning methods, and the industrial internet of things (IIoT). The development of sensing technology provides large amounts and various types of data (e.g., profile, image, point cloud, etc.) to describe each stage of a manufacturing process. The machine learning methods have the advantages of efficiently and effectively processing and fusing large-scale datasets and demonstrating outstanding performance in different tasks (e.g., diagnosis, monitoring, etc.). Despite the advantages of incorporating machine learning methods into smart manufacturing, there are some widely existing concerns in practice: (1) Most of the edge devices in the manufacturing system only have limited memory space and computational capacity; (2) Both the performance and interpretability of the data analytics method are desired; (3) The connection to the internet exposes the manufacturing system to cyberattacks, which decays the trustiness of data, models, and results. To address these limitations, this dissertation proposed systematic engineering-driven machine learning methods to improve the system intelligence for smart manufacturing. The contributions of this dissertation can be summarized in three aspects. First, tensor decomposition is incorporated to approximately compress the convolutional (Conv) layer in Deep Neural Network (DNN), and a novel layer is proposed accordingly. Compared with the Conv layer, the proposed layer significantly reduces the number of parameters and computational costs without decaying the performance. Second, a physics-informed stochastic surrogate model is proposed by incorporating the idea of building and solving differential equations into designing the stochastic process. The proposed method outperforms pure data-driven stochastic surrogates in recovering system patterns from noised data points and exploiting limited training samples to make accurate predictions and conduct uncertainty quantification. Third, a Wasserstein-based out-of-distribution detection (WOOD) framework is proposed to strengthen the DNN-based classifier with the ability to detect adversarial samples. The properties of the proposed framework have been thoroughly discussed. The statistical learning bound of the proposed loss function is theoretically investigated. The proposed framework is generally applicable to DNN-based classifiers and outperforms state-of-the-art benchmarks in identifying out-of-distribution samples. / Doctor of Philosophy / The global industries are experiencing the fourth industrial revolution, which is characterized by the use of advanced sensing technology, big data analytics, and the industrial internet of things (IIoT) to build a smart manufacturing system. The massive amount of data collected in the engineering process provides rich information to describe the complex physical phenomena in the manufacturing system. The big data analytics methods (e.g., machine learning, deep learning, etc.) are developed to exploit the collected data to complete specific tasks, such as checking the quality of the product, diagnosing the root cause of defects, etc. Given the outstanding performances of the big data analytics methods in these tasks, there are some concerns arising from the engineering practice, such as the limited available computational resources, the model's lack of interpretability, and the threat of hacking attacks. In this dissertation, we propose systematic engineering-driven machine learning methods to address or mitigate these widely existing concerns. First, the model compression technique is developed to reduce the number of parameters and computational complexity of the deep learning model to fit the limited available computational resources. Second, physics principles are incorporated into designing the regression method to improve its interpretability and enable it better explore the properties of the data collected from the manufacturing system. Third, the cyberattack detection method is developed to strengthen the smart manufacturing system with the ability to detect potential hacking threats.

With the rise of smart manufacturing technologies and a shift towards a new industrial revolution, brings forth many new challenges, one of which is how to adapt and integrate these technologies into existing assembly lines. Scania CV AB has joined this race and, with the help of smart manufacturing solutions, works on increasing efficiency amongst its assembly lines. This thesis is aimed at creating concept designs using different smart manufacturing technologies in the assembly line of a pedal car, to evaluate and adapt the concept suited for a real assembly line. The thesis starts with studying the different smart manufacturing technologies to better understand them and the scientific methods used. This follows up with the methodology where several scientific methods such as morphological matrix and weight based decision making matrix are used to generate and evaluate different concept designs. This is followed by a qualitative analysis that helps in selecting the concept design that best suits the needs of the assembly line under consideration. The different concepts are visualized and the evaluation based on different parameters are discussed. This thesis lays a foundation to realize that an aggregate of an optimized process plan, a continuous improvement strategy and the right use of smart manufacturing technologies contributes to the productivity of the assembly line in the long run. / Med en ökning av smarta tillverkningsteknologier och ett skift mot en ny industriell revolution, kommer nya utmaningar. Av dessa utmaningar ifrågasätts hur man anpassar och kombinerar dessa olika teknologier gentemot existerande monteringslinjer. Scania CV AB har tagit del i denna resa och, med hjälp av smarta tillverkningslösningar, jobbar ständigt mot att effektivisera sina monteringslinjer. Detta examensarbete fokuserar på att använda olika och smarta tillverkningsteknologier i en monteringslina för tillverkning av en trampbil. Detta görs genom att evaluera och anpassa olika koncept som är lämpade för en verklig monteringslina. Examensarbetet börjar med en undersökning av nuläget för att få en bättre uppfattning om smarta tillverkningsteknologier samt de vetenskapliga metoder som används. Även här så undersöks referensprodukten - trampbil samt de olika programvarorna AviX, ExtendSim och LayCAD. Genom att ha satt tydliga arbetsmetoder och följt upp mot dessa så diskuterar och hänvisar nästa kapitel resultaten. Resultaten visar hur olika koncept har tagits fram samt vilka teknologier som går ihop med dessa. Fördelar, nackdelar och risker hos respektive teknologi har benämnts. Ett systematiskt arbetssätt har tagits fram mot hur man anpassar konceptet för monteringslinan av en trampbil till en verklig monteringslina, samt hur man har jobbat runt de restriktioner som uppkommit. Examensarbetet slutar med ett kapitel där slutsatserna, på en överskådlig nivå, tas fram. Hur dessa teknologier har kombinerats och evaluerats, samt att koncepterna som har tagits fram har lagt en grund för framtida projekt att följa upp emot.

The industry has during the past decades been changing towardsdigitalization at a rapid pace, adapting new frameworks and digitalsolutions, with the thrive to improve efficiency, and output quality.This thesis covers how a hyped industry concept, digital twin, incontext of smart manufacturing, could be applied in this changingclimate. Focus is put on what strategies a Product LifecycleManagement (PLM) principal could adapt when approaching theutilization of digital twins, in a customer setting. The research project, using a qualitative exploration format,incorporated a thorough review of journal articles and standards,interconnected with conducted interviews with industry experts, inorder to develop two strategies on how to approach the conceptof digital twins. The two-part strategies were formed on the basis of Mintzberg(1978) emergent and deliberate strategies. The former, consisted ofIDEF0 function modeling diagramming, where a digital twin businessprocess was portrayed, based on the literature review and interviewdata. The latter, approached the digital twin application challenge ina theoretical manner, based on the concept’s high risk anduncertainty, incorporating organizational structure theory andinnovation theory. The two strategies showed two different pathsto approach the digital twin phenomena and how to, boththeoretically and practically, adapt digital twins in a customer setting.