No description available.

Lagegeregelte Bewegungsachsen finden sich in nahezu allen aktuellen Werkzeug‐ und Produktionsmaschinen. Um eine hohe Qualität der hergestellten Produkte bei gleichzeitig möglichst hoher Effizienz sicherzustellen muss die Regelung der Bewegungsachsen genau auf das Regelstreckenverhalten abgestimmt sein. Informationen über die Regelstrecken können mit Verfahren der Identifikation ermittelt werden. Neben der Reglerparametrierung sind diese Informationen ebenfalls für Mechanismen der Maschinenüberwachung relevant, bei denen Fehlerzustände in den Maschinen und Anlagen (modellbasiert) frühzeitig erkannt werden sollen. Neben den Identifikationsmechanismen, welche derzeit im Bereich der Werkzeugmaschinen vorherrschend sind und meist auf Testsignalen basieren, können mit nichtinvasive Ansätzen wesentliche Parameter bestimmt werden, ohne in den Produktionsprozess einzugreifen zu müssen. Im vorliegend Beitrag wird dargestellt, wie derartige Verfahren im Umfeld der Werkzeugmaschine Anwendung finden können. Nach einer Einleitung wird im Beitrag der Stand der Technik zur Identifikation von Regelstreckenparametern an Bewegungsachsen beleuchtet. Das Hauptaugenmerk liegt dabei in der Unterscheidung von invasiven (Verfahren mit Testsignalen) und nichtinvasiven Identifikationsverfahren (ohne Testsignale). Aufbauend auf den Erkenntnissen der Literaturrecherche wird das eigene Vorgehen hinsichtlich Wahl der Modellstruktur und des Identifikationsverfahrens erläutert. Ausführungen zur Anregungsdetektion, gezielten Einflussnahe auf die Parameterschätzung, Fehlerbewertung und dem Einsatz der simulationsbasierten Optimierung schließen die theoretischen Betrachtungen zum Forschungsthema ab. Die erzielten Ergebnisse werden in einem weiteren Kapitel anhand eines Laborversuchsstandes und einer Werkzeugmaschine illustriert. Der letzte Abschnitt gibt eine Zusammenfassung des Forschungsvorhabens und erläutert zukünftige Schritte auf diesem Gebiet.

No description available.

<p dir="ltr">Differential privacy (DP) uses a probabilistic framework to measure the level of privacy protection of a mechanism that releases data analysis results to the public. Although DP is widely used by both government and industry, there is still a lack of research on statistical inference under DP guarantees. On the one hand, existing DP mechanisms mainly aim to extract dataset-level information instead of population-level information. On the other hand, DP mechanisms introduce calibrated noises into the released statistics, which often results in sampling distributions more complex and intractable than the non-private ones. This dissertation aims to provide general-purpose methods for statistical inference, such as confidence intervals (CIs) and hypothesis tests (HTs), that satisfy the DP guarantees. </p><p dir="ltr">In the first part of the dissertation, we examine a DP bootstrap procedure that releases multiple private bootstrap estimates to construct DP CIs. We present new DP guarantees for this procedure and propose to use deconvolution with DP bootstrap estimates to derive CIs for inference tasks such as population mean, logistic regression, and quantile regression. Our method achieves the nominal coverage level in both simulations and real-world experiments and offers the first approach to private inference for quantile regression.</p><p dir="ltr">In the second part of the dissertation, we propose to use the simulation-based ``repro sample'' approach to produce CIs and HTs based on DP statistics. Our methodology has finite-sample guarantees and can be applied to a wide variety of private inference problems. It appropriately accounts for biases introduced by DP mechanisms (such as by clamping) and improves over other state-of-the-art inference methods in terms of the coverage and type I error of the private inference. </p><p dir="ltr">In the third part of the dissertation, we design a debiased parametric bootstrap framework for DP statistical inference. We propose the adaptive indirect estimator, a novel simulation-based estimator that is consistent and corrects the clamping bias in the DP mechanisms. We also prove that our estimator has the optimal asymptotic variance among all well-behaved consistent estimators, and the parametric bootstrap results based on our estimator are consistent. Simulation studies show that our framework produces valid DP CIs and HTs in finite sample settings, and it is more efficient than other state-of-the-art methods.</p>

Several exact and approximate dynamic programming formulations have already been proposed to solve hotel revenue management (RM) problems. To obtain tractable solutions, these methods are often bound by simplifying assumptions which prevent their application on large and dynamic complex systems. This dissertation introduces HotelSimu, a flexible simulation-based optimization approach for hotel RM, and investigates possible approaches to increase the efficiency of black-box optimization methods in the presence of noise. In fact, HotelSimu employs black-box optimization and stochastic simulation to find the dynamic pricing policy which is expected to maximize the revenue of a given hotel in a certain period of time. However, the simulation output is noisy and different solutions should be compared in a statistically significant manner. Various black-box heuristics based on variations of random local search are investigated and integrated with statistical analysis techniques in order to manage efficiently the optimization budget.

No description available.

No description available.

No description available.

In order for the companies to be competitive in today’s market, it is vital to adapt quickly to the market trends. The steady shift towards mass customization from mass production has been challenging many industries globally, which demands the use of digital tools and technologies in various areas to improve performance throughout the supply chain processes. One of these areas is short-term scheduling of jobs on the shop floor. Short-term scheduling of jobs plays a very vital role in many production systems. Optimisation of short-term scheduling help the companies in improving their operational Key Performance Indicators (KPIs), thus saving both money and resources. Today’s complex production systems with multiple constraints, system level interactions and the dynamic environment have challenged the traditional static scheduling approaches. These complex production systems require new scheduling approaches which can consider all the dynamics and should be capable of real-time reconfiguring in case of uncertainties in the shop floor. In this thesis, a case study was performed in a steel manufacturing company over the period of five months to validate Discrete Event Simulation (DES) as a tool for short-term scheduling of heavy plates in a dynamic environment. The challenges related to DES for short-term scheduling during model design, development and implement phases were also identified. In addition to this, the requirements to implement DES model for short-term scheduling in a dynamic environment were also discussed. Furthermore, a Systematic Literature Review (SLR) was also conducted to support the empirical findings from the case study. The idea of this study was to generate an optimal schedule by minimizing overall makespan and maximizing resource utilization using DES model. The findings from both SLR and DES model has clearly proven that DES as a digital tool is exceptional for short-term scheduling in a dynamic environment, nevertheless there are still some challenges associated which needs to be investigated further. The same model can also be used for other purposes such as analysing and identifying bottlenecks in the whole production system. / För att företagen ska vara konkurrenskraftiga på dagens marknad är det viktigt att snabbt anpassa sig till marknadstrenderna. Vägen mot mass-anpassning från massproduktion har utmanat många industrier globalt, vilket kräver användning av digitala verktyg och tekniker inom olika områden för att förbättra prestandan under hela leveranskedjans processer. Ett av dessa områden är kortsiktig schemaläggning av arbeten på fabriksgolvet. Kortsiktig schemaläggning av arbeten spelar en mycket viktig roll i många produktionssystem. Optimering av kortsiktig schemaläggning hjälper företagen att förbättra sina operativa nyckeltal, vilket sparar både pengar och resurser. Dagens komplexa produktionssystem med flera begränsningar, systemnivåinteraktioner och den dynamiska miljön har utmanat de traditionella statiska schemaläggningsmetoderna. Dessa komplexa produktionssystem kräver nya schemaläggningsmetoder som kan ta hänsyn till all dynamik och bör ha möjligheten att omkonfigurera i realtid vid osäkerheter på fabriksgolvet. I denna avhandling genomfördes en fallstudie i ett ståltillverkningsföretag under fem månader för att validera Diskret händelsestyrd simulering som verktyg för kortsiktig schemaläggning av grovplåtstillverkning i en dynamisk miljö. Utmaningarna relaterade till Diskret händelsestyrd simulering för kortsiktig schemaläggning under modellens design, utveckling och implementeringsfaser identifierades också. Utöver detta diskuteras också kraven för att implementera Diskret händelsestyrd simulering för kortsiktig schemaläggning i en dynamisk miljö. Dessutom genomfördes en systematisk litteraturstudie för att stödja de empiriska resultaten från fallstudien. Tanken med den här studien var att generera ett optimalt schema genom att minimera den totala schemalängden och maximera resursutnyttjandet med hjälp av Diskret händelsestyrd simuleringsmodellen. Resultaten från både den systematiska litteraturöversynen och Diskreta händelse simuleringsmodellen har tydligt bevisat att Diskret händelse simulering som ett digitalt verktyg är exceptionellt för kortsiktig schemaläggning i en dynamisk miljö även om det fortfarande finns några utmaningar som måste undersökas ytterligare. Samma modell kan också användas för andra ändamål som att analysera och identifiera flaskhalsar i hela produktionssystemet.

This master thesis presents a practical framework for integrating Simulation-Based Optimization into facility layout studies. While various methodologies exist for improving facility layouts and the utilization of Simulation-based Optimization, there is a gap in practical frameworks offering a systematic approach to combine these methods effectively. Existing frameworks lack specificity, require substantial prior knowledge, and offer limited insight into the methodologies employed. Consequently, there is a need for a comprehensive, step-by-step framework accessible to a broader range of practitioners. The proposed framework addresses this gap by providing a practical step-by-step approach that guides practitioners through a Simulation-Based Optimization integrated facility layout study. This framework facilitates the development of alternative layouts and system configurations using trusted methods. To assess the proposed framework, a case study was conducted at a manufacturing facility with the purpose of enhancing the production throughput. The case study followed the steps outlined in the framework in a real-world scenario, which provided valuable insights into the practicality and usefulness of the framework. The framework's effectiveness hinges on careful execution of each step, particularly given its front-loaded nature. Neglecting any step could lead to inaccuracies in subsequent stages, undermining the overall validity of the study. Adequate time allocation, especially in data collection and simulation model development, is critical to ensuring robust results. While the framework's applicability has been demonstrated in a high-mix low-volume production environment, its broader effectiveness across different settings remains to be explored. Nonetheless, the framework's intuitive flow and reliance on established methodologies enhance its usability and potential to improve production system throughput. Ultimately, the study contributes a tangible solution to the research question, offering practitioners a valuable tool for streamlining production.