This the multi-page printable view of this section. Click here to print.
- 1: Development of a methodology for implementing Predictive Maintenance
- 2: Design and Evaluation of a Blue Ocean Strategy for an Open-Core IIoT Platform Provider in the Manufacturing Sector"
- 3: Industrial image processing for quality control in production lines: Development of a decision logic for the application case specific selection of hardware and software
- 4: Deep learning for industrial quality inspection: development of a plug-and-play image processing system
- 5: Development of a decision tool to select appropriate solutions for quality control depending on the defects occurring in the manufacturing process in the automobile branch of the technical-textiles industry
- 6: Implementation of Time Series Data based Digital Time Studies for Manual Processes within the Context of a Learning Factory
1 - Development of a methodology for implementing Predictive Maintenance
Objective of this thesis: The goal of this thesis is to develop a methodology to implement Predictive Maintenance (PdM) economically viable into a company. The methodology is then validated in the Digital Capability Center (DCC) Aachen.
Solution process: Maintenance strategies and machine learning algorithms are researched together with methods for optimizing productions lines. This knowledge is then summarized and validated in the DCC Aachen.
Key results: Because of high costs and effort PdM is only economically viable on machines and components with high revenue losses due to breakdown and where the failure is almost independent from uptime and wear. In the DCC Aachen the wind up bearing at the warping machine is identified as a component for a PdM implementation, but a combination of machine learning and existing sensors is not enough for a economically viable implementation.
Key word: Predictive Maintenance, maintenance strategies, machine learning
2 - Design and Evaluation of a Blue Ocean Strategy for an Open-Core IIoT Platform Provider in the Manufacturing Sector"
Final blue ocean strategy and competitive positioning of UMH
The ongoing trend of digitization does not exclude the manufacturing sector which poses challenges for both involved parties: machine manufacturers and producing companies. While larger corporations and certain industries, such as the automotive industry, may be already well advanced in digitization, small and medium-sized enterprises (SMEs) often still face difficulties. For many producing companies, the added value of digitization is unclear and they also have strong security concerns about making their data available for analyses. On the other hand, machine manufacturers face problems in their transformation from hardware manufacturers to service providers (Bitkom Research & Ernst & Young 2018, p. 21; VDMA & McKinsey 2020, p. 22; Vargo & Lusch 2008, p. 254ff). Due to these difficulties and the currently high implementation costs, IIoT (industrial internet of things) platforms have so far been deployed rather sporadically. However, they can offer great potential for optimization, for example of service processes or the overall equipment effectiveness (OEE) and will play an important role in maintaining competitiveness in the future (VDMA & McKinsey 2020, p. 27ff).
For this reason, more and more startups and third-party providers are currently establishing businesses that are trying to solve the challenges and problems of both sides with a wide variety of approaches. Currently, the market is quite opaque, which makes it difficult to compare providers on the market and thus to compete. This thesis is written in cooperation with the Aachen-based startup developing the IIoT platform “United Manufacturing Hub” (UMH; UMH Systems GmbH). Its objective is to set UMH apart from the existing red-ocean market with the development of a blue ocean strategy. By redistributing the development focus to attributes that are most relevant to customers in the market and reducing efforts in less relevant areas, the goal is to create a new, non-competitive market (Kim & Mauborgne 2015, p. 24ff). UMH has set itself the task of making the digital transformation as easy as possible for machine manufacturers and producing companies as their end customers. To do this, it is important to know the needs and problems of the customers and to obtain their assessment of the solution approaches. As a starting point for market analysis, this thesis focuses on machine manufacturers as customers of the platform.
The research question is divided into sub-questions, which together contribute to answering the primary question (Karmasin & Ribing 2017, p. 24f). While the topic is elaborated on the example of UMH, the underlying questions can be generalized and are not sufficiently addressed in the existing literature. The concepts further described in chapter 2 provide useful insights into IIoT, open-source platforms, as well as blue ocean strategies, but there is limited literature on the linkages between those topics (e.g., Frank et al. 2019; p. 341ff; Shafiq et al. 2018, p. 1076ff) and none describing a blue ocean strategy in an IIoT platform context. Therefore, the primary research question (PQ) is:
PQ: Which blue ocean strategy has the best potential to set industry standards and establish an IIoT platform in the manufacturing sector?
Currently, most machine manufacturers rely on in-house developed IIoT platforms (Bender et al. 2020, p. 10f), although using an external platform would reduce duplication costs and provide access to existing applications and customers (Evans & Schmalensee 2008, p. 673). This suggests that currently available external IIoT platforms do not sufficiently cover customer needs. To better understand machine manufacturers’ needs and their motivation, the first sub-question (SQ) is therefore:
SQ1: What functionalities do the manufacturers’ platforms include and how were they implemented? Why have machine manufacturers decided to develop their own platform?
The four actions framework in the blue ocean literature suggests that product attributes need to be raised or created to increase the customer value and create new demand while others are reduced or eliminated to achieve cost leadership (Kim & Mauborgne 2015, p. 51). To assess and extend UMH’s solution approaches, the second sub-question is:
SQ2: 11 What functions or features are currently missing from existing platforms on the market? Which attributes must be raised to fulfill the desired customer benefits?
Finally, making the core of the software stack open source is a relevant part of UMH’s disruptive business model. Open source reduces costs and dependence and promotes among other things value creation. Dedrick and West (2004, p. 5f) found that the perceived reliability of Linux-operated servers was lower than that of servers with a proprietary operating system, which could also be the case for an open-source IIoT platform. To examine the effects of the open-source approach the third sub-question is:
SQ3: How does an open-source approach affect the value curve and how is it perceived by machine manufacturers?
To answer these research questions, this thesis first reviews the state of research on digitization and IIoT, platforms, and technology adoption of a market. Next, UMH and its open-core concept are presented based on an implemented proof of concept at the Digital Capability Center (DCC) Aachen. UMH’s competitors are then clustered into infrastructure providers, proprietary IIoT platforms, and system integrators, for which value curves are generated that show the current focus of the providers on the market. Hypotheses are formulated about the requirements of the IIoT market based on the literature, a conversation with Bender and Lewandowski (2021; authors of the underlying paper Bender et al. 2020), and an existing market research by UMH (2020). The hypotheses facilitate the preparation of the three-part interview guideline, each dedicated to answering one sub-question. Finally, the interviews with eight development managers at machine manufacturing companies are evaluated. Based on the findings, the hypotheses are assessed and the blue ocean strategy for UMH open core and premium is derived, thus answering the primary research question.
3 - Industrial image processing for quality control in production lines: Development of a decision logic for the application case specific selection of hardware and software
This publication was made by Michael Müller as a Master Thesis for the “Institut für Textiltechnik der RWTH Aachen University” in cooperation with Kai Müller (ITA / RWTH Aachen ) and us.
Cognex camera connected with the United Manufacturing Hub open-source stack
Objective of this thesis: The goal of the work is the development of a decision logic for the application case-specific selection of hardware and software for image processing systems for for quality control in industrial production. On the hardware side the components components camera, lens and illumination system are considered. On software side, it is decided, depending on the application, whether conventional algorithms or ventional algorithms or methods of Deep Learning are more suitable.
Solution process: Within the scope of a literature search, relevant descriptive variables for standardized for the standardized characterization of technologies and use cases. Furthermore, interdependencies between individual components and properties of the use case will be identified. By means of a market research, a database with concrete product product information. Based on these steps, a set of rules for the selection of hardware and software technologies is derived and tested on a use case in the application case at the Digital Capability Center Aachen. The decision-making logic for selecting hardware components will finally be user-friendly computer application.
Key results: To select suitable hardware components, a five-stage decision logic is developed and implemented as a software application, which suggests suitable components to the user depending on the specified use case and prioritizes them according to list price. In a simulative evaluation, this achieves complexity reductions between 73 and 98% and cost savings between 46 and 93%. A decision between Deep Learning and conventional algorithms can be made based on the given development circumstances as well as the complexity of image features.
Key word: Digital quality control, Technical textiles, Mobiltech, Industry 4.0, Technology selection
4 - Deep learning for industrial quality inspection: development of a plug-and-play image processing system
MQTT is used as a central element in the open-source architecture for image processing systems
Objective of this thesis: The objective of the thesis is the development of a robust and user-friendly software for an industrial image processing system, which applies deep learning methods. The user of this software will be able to quickly and easily put an image processing system into operation due to its plug-and-play capability and standardized interfaces. The system software is based exclusively on royalty-free software products.
Solution process: For the development of the overall system, relevant standards, interfaces and software solutions are researched and presented. By dividing the sys- tem into sub-processes, functional requirements for the software are derived and implemented in the development with the general requirements in a system architecture. The implementation and subsequent validation is carried out in the model production for textile wristbands at the Digital Capability Center Aachen.
Key results: The central result is an overall process overview and a microservice architecture, with the help of which an industrial image processing system can be put into operation on the software side only by configuring the camera and entering the environment variables. Currently, cameras of the GenICam standard with GigE Vision interface and Cognex cameras are supported. The open architecture creates a basic platform for the development of further microservices and subsequent processes in the context of industrial image processing.
Key word: Machine vision, quality control, deep learning, microservice architecture, MQTT
5 - Development of a decision tool to select appropriate solutions for quality control depending on the defects occurring in the manufacturing process in the automobile branch of the technical-textiles industry
This publication was made by Aditya Narayan Mishra as a Master Thesis for the “Institut für Textiltechnik der RWTH Aachen University” in cooperation with Kai Müller (ITA / RWTH Aachen ) and us.
Objective of this thesis: The objective of this thesis is to develop a decision tool regarding the quality control in the manufacturing of technical textiles for the automotive industry. The tool shall enable the access to information about the problems being faced and the consequent defects occurring during the manufacturing of technical textiles in the automotive industry. Subsequently, it shall provide an overview of the corresponding solutions and measuring principles for each of the identified problems
Solution process: Firstly, a literature review is carried out to provide a deep profound understanding to the important quality parameters and defects in each of the manufacturing processes of technical textile. Based on the literature review, a questionnaire is created to perform a market analysis in form of expert interviews. With the help of the market analysis, industry insights to the current status and problems associated with the quality control of manufacturing the technical textile fabrics in the automotive industry are addressed. Afterwards, based on the problems acquired through the expert interviews, the solutions and measuring principles are identified and subsequently a concept for the decision tool is designed.
Key results: The results of this research provide an overview of the problems being faced regarding quality control during the manufacturing processes of technical textile in the automotive industry. In addition, information on the extent to which digital solutions for quality control are established in the industry is analyzed. Moreover, existing digital quality control solutions and measuring principles to tackle the identified problems in the industry are researched and identified.
Key word: Digital quality control, Technical textiles, Mobiltech, Industry 4.0, Technology selection
6 - Implementation of Time Series Data based Digital Time Studies for Manual Processes within the Context of a Learning Factory
This publication was made by Tobias Tratner (Xing, LinkedIn) as a Master Thesis for the Graz University of Technology & Deggendorf Institute of Technology in cooperation with Maria Hulla (Institute of Innovation and Industrial Management at TU Graz) and us.
Finished setup of the time studies in the LEAD-Factory
The steadily advancing globalization significantly shapes today’s business environment for companies. Therefore, companies are increasingly under immense cost pressure and need to improve their production efficiency and product quality to remain competitive. Industry 4.0 applications that result from the advancing digitization offer great potential for long-term cost savings. Implementing time studies for mechanical activities can identify potential for improvement in the production process and enable them to be rectified. This thesis is concerned with combining the subject areas Industry 4.0 and the implementation of manual time studies. For this purpose, a digital time recording of manual activities was implemented in the LEAD Factory, the learning factory of the Institute of Innovation and Industrial Management at Graz University of Technology. Therefore, a mobile sensor kit and an IoT platform, provided by the factorycube of Industrial Analytics, were used. Using sensors, existing data from an RFID system, and an energy monitoring system, all activities on a selected workstation in the LEAD Factory can be documented and analyzed. This automated time recording enables long-term measurements to analyze working times and possible anomalies. The collected data is stored in so-called time- series databases and processed using various methods. The data is displayed on a dashboard using a visualization program. One focus of the work was the design of the data processing architecture with two different time-series data models, as well as the conception and development of methods for data processing in the context of time studies. A relational and a NoSQL database system were used equally. The use of two very different approaches should show the possibilities of both systems and enable an assessment of the two systems. Based on a utility analysis, both approaches are evaluated and compared using selected criteria. Thus, a clear recommendation can be made for one of the two approaches. Making the results of the work available to an open-source community, they can be used as a basis for the implementation of similar applications. In addition, the work shows through a digital time recording the huge potential to improve productivity in case of using existing data in a production environment.