Automotive Case Study Research Articles

Data enrichment and customer insights for hyper-personalised experiences: An automotive case study

In our increasingly interconnected world, companies are seeking to connect with their customers through multiple touch points, both offline and online. For companies specialising in products and services, success is no longer about gaining one-time momentum but rather about creating deeply personalised interactions with customers along their purchase journey. Meeting the customer’s needs requires providing them with the right information at the right place and the right time. This paper argues that those companies that are collecting and enhancing consumer data will find themselves developing the most effective targeting strategies. The paper examines how data enrichment plays an integral part in providing a better customer experience and why hyper-personalisation should be at the core of your commercialisation strategy. By way of illustration, the paper describes how Tesla Motors embarked on a programme of data enrichment and personalisation. The paper concludes with key takeaways for companies looking to achieve sustainable engagement in a rapidly evolving digital landscape.

Manufacturing resilience through disruption mitigation using attention-based consistently-attributed graph embedded decision support system

With global supply chains experiencing significant disruptions, there is increasing emphasis on enhancing manufacturing resilience by mitigating supply chain-propagated impacts on production entities in a holistic manner. However, existing mitigative strategies typically rely on implicit experience, which may be biased and suboptimal. Furthermore, the lack of consideration towards the chain of custody, multi-disruption management, and transportation networks results in unintuitive systems. To overcome these challenges, this study proposes a graph embedding-based mitigation decision support system (GEM-DSS) using a heterogeneous industrial knowledge graph to facilitate disruption management. Addressing the prevailing industrial challenge of knowledge representations and graph incompleteness, an attention-based consistently-attributed graph embedding (ACAGE) model is proposed, with the capability to learn and factor in multi-attributes and multi-structures within the knowledge graph. Through supplier selection and material substitution mitigation strategies, two computational pipelines are designed as part of a dual-stage search mechanism to generate feasible solutions. The experimental results of the ACAGE model for knowledge graph completion outperform several attributed graph embedding baselines, achieving 0.999 AUC-ROC and 0.990 AUC-PR on the constructed real-world industrial knowledge graph. The functionality and applicability of the GEM-DSS are demonstrated in an automotive case study with dynamic considerations via a cognitive computing methodology. This study aims to provide valuable insights for both academia and industry towards enabling manufacturing continuity in the face of global supply chain challenges.

Increasing trust in AI through privacy preservation and model explainability: Federated Learning of Fuzzy Regression Trees

Federated Learning (FL) lets multiple data owners collaborate in training a global model without any violation of data privacy, which is a crucial requirement for enhancing users’ trust in Artificial Intelligence (AI) systems. Despite the significant momentum recently gained by the FL paradigm, most of the existing approaches in the field neglect another key pillar for the trustworthiness of AI systems, namely explainability. In this paper, we propose a novel approach for FL of fuzzy regression trees (FRTs), which are generally acknowledged as highly interpretable by-design models. The proposed FL procedure is designed for the scenario of horizontally partitioned data and is based on the transmission of aggregated statistics from the clients to a central server for the tree induction procedure. It is shown that the proposed approach faithfully approximates the ideal case in which the tree induction algorithm is applied on the union of all local datasets, while still ensuring privacy preservation. Furthermore, the FL approach brings benefits, in terms of generalization capability, compared to the local learning setting in which each participant learns its own FRT based only on the private, local, dataset. The adoption of linear models in the leaf nodes ensures a competitive level of performance, as assessed by an extensive experimental campaign on benchmark datasets. The analysis of the results covers both the aspects of accuracy and interpretability of FRT. Finally, we discuss the application of the proposed federated FRT to the task of Quality of Experience forecasting in an automotive case-study.

Product Development and Design Framework Based on Interactive Innovation in the Metaverse Perspective

Based on the theory of user needs and the product development mode and framework of mobile Internet interactive innovation, a new “reality → virtual → reality” interactive innovation product development mode is constructed. It draws on the unique characteristics, systematic technical system, and comprehensive scientific and technological layout of the Metaverse. On this basis, a framework for product development and design based on interactive innovation from the Metaverse perspective is innovatively proposed. In the Metaverse scenario, interactive innovation knowledge can be easily and effectively transformed into design knowledge, and all groups of users truly participate in the whole process of product design. Moreover, the development of interactive innovative products in the Metaverse scenario can be combined with artificial intelligence (AI) technology to further automate the statistical analysis of user needs and preferences so as to meet the dynamic needs of users and accurately develop products that fit user needs and enterprise standards. In addition, users, designers, and enterprises can make joint decisions on product design solutions and development forms, and the Metaverse technology can also optimize the products with continuous iteration and obtain the optimal solutions. An automotive case study illustrates the feasibility and effectiveness of the model for product development innovation and enterprise digital transformation.

A journey with ASMETA from requirements to code: application to an automotive system with adaptive features

Modern automotive systems with adaptive control features require rigorous analysis to guarantee correct operation. We report our experience in modeling the automotive case study from the ABZ2020 conference using the ASMETA toolset, based on the State Machine formal method. We adopted a seamless system engineering method: from an incremental formal specification of high-level requirements to increasingly refined ASMETA models, to the C++ code generation from the model. Along this process, different validation and verification activities were performed. We explored modeling styles and idioms to face the modeling complexity and ensure that the ASMETA models can best capture and reflect specific behavioral patterns. Through this realistic automotive case study, we evaluated the applicability and usability of our formal modeling approach.

An automotive case study

An automotive case study

Determination of Hotspot Temperature Margin for Rectangular Wire Windings Considering Insulation Thermal Degradation and Partial Discharge

Increasing current density is a common way to improve the power density of electrical machines (EMs) which also boosts the winding temperature. Higher winding temperature accelerates the thermal degradation of insulation and also increases the risk of partial discharge (PD). Commonly, the winding hotspot temperature constraint is decided based on an over-engineering approach (i.e., the winding hotspot temperature is 30°C below the wire thermal class) constraining the EM’s performance. In this paper, the margin of winding hotspot temperature is precisely determined focusing mainly on the insulation degradation issues through experimental investigation. The partial discharge inception voltage and partial discharge extinction voltage have been measured under different temperature conditions for samples with different thermal aging statuses. Rectangular wires which find application in automotive motors are selected as test samples. The results link the temperature with the PD risk and the thermal lifetime of the insulation, and the winding hotspot temperature margin is accordingly determined. An automotive case study is presented to prove the applicability of the investigation outcomes.

Big data and machine learning in cost estimation: An automotive case study

This paper presents a case study on the applicability of machine learning and big data technology for product cost estimation, using data on the material cost of passenger cars. The study provides contributions to six research aspects. First, we show which machine learning algorithms are appropriate when dealing with product cost estimation for highly complex products with more than 2000 parts and hundreds of cost drivers. Second, our case study provides a novel approach to increasing the predictive accuracy of cost estimates for subsequent product generations. Third, we show that the accuracy is up to 3.5 times higher when using big data compared to an intermediate size of data. Fourth, machine learning can outperform cost estimates from cost experts during the early stage of new product development, even when dealing with highly complex products. Then, we evaluate the use cases, issues, and benefits of machine learning and big data from the perspective of cost experts. Specifically, the case study shows that machine learning can reliably select the most important cost drivers (fifth aspect) and calculate the average cost of cost drivers over thousands of product configurations (sixth aspect). However, cost experts must be knowledgeable about the product and remain careful when interpreting machine learning outcomes, as they can yield misleading outcomes in some exceptional cases. In conclusion, machine learning and big data empirically proved to be able to generate additional value in many aspects for managing costs during the early phase of new product development.

Towards seamless collaboration of humans and high-payload robots: An automotive case study

Towards seamless collaboration of humans and high-payload robots: An automotive case study

Automated cluster generation and labelling of peer groups for marketing reporting

In today's data-driven marketing landscape, clustering data helps businesses better understand themselves and their customers. However, clusters derived from machine learning can be difficult to interpret and obtain buy-in from stakeholders. This paper details a method for automated cluster generation and labelling using machine learning. Two automotive case studies are provided where clustering enhanced business value and gained stakeholder buy-in. The first details segmenting dealerships based on their media environment to produce higher quality media models for lead generation. The second entails the creation of peer groups to enhance performance reporting across a diverse set of dealerships.

Multidomain Synthesis of Optimal Vibration Suppression Systems

Abstract There are numerous design possibilities for vibration–suppression systems considering components from multiple domains (e.g., mechanical, hydraulic, pneumatic, electrical). Traditional vibration absorber design approach could only explore limited possibilities, of which the performance is away from optimal. Since the 2000s, network synthesis-based approach has been applied. It allows identification of optimal absorber properties represented by networks consisting of modeling elements (stiffness, damping, inertance), providing significant theoretical performance improvements. However, such improvements have not yet been realized in industry. This is because the following questions have not been answered: (1) What are the network-represented properties of the conventional absorber? (2) How can the optimal network-represented properties be realized considering multidomain physical components? This article provides a method for answering these questions by proposing a novel multidomain synthesis technique, allowing bidirectional transformation between networks and multidomain components. Building on this technique, a vibration–absorber design methodology is proposed, which can construct physical realizations of optimal absorbers considering multidomain components. Another contribution of this work is to propose a novel component, providing a hydraulic realization of compliance ‘embedded’ in a hydraulic network. This methodology is demonstrated using an automotive case study, where the constructed optimal hydraulic suspension provides 23% ride comfort enhancement over the conventional one.

GRU-Based Denoising Autoencoder for Detection and Clustering of Unknown Single and Concurrent Faults during System Integration Testing of Automotive Software Systems.

Recently, remarkable successes have been achieved in the quality assurance of automotive software systems (ASSs) through the utilization of real-time hardware-in-the-loop (HIL) simulation. Based on the HIL platform, safe, flexible and reliable realistic simulation during the system development process can be enabled. However, notwithstanding the test automation capability, large amounts of recordings data are generated as a result of HIL test executions. Expert knowledge-based approaches to analyze the generated recordings, with the aim of detecting and identifying the faults, are costly in terms of time, effort and difficulty. Therefore, in this study, a novel deep learning-based methodology is proposed so that the faults of automotive sensor signals can be efficiently and automatically detected and identified without human intervention. Concretely, a hybrid GRU-based denoising autoencoder (GRU-based DAE) model with the k-means algorithm is developed for the fault-detection and clustering problem in sequential data. By doing so, based on the real-time historical data, not only individual faults but also unknown simultaneous faults under noisy conditions can be accurately detected and clustered. The applicability and advantages of the proposed method for the HIL testing process are demonstrated by two automotive case studies. To be specific, a high-fidelity gasoline engine and vehicle dynamic system along with an entire vehicle model are considered to verify the performance of the proposed model. The superiority of the proposed architecture compared to other autoencoder variants is presented in the results in terms of reconstruction error under several noise levels. The validation results indicate that the proposed model can perform high detection and clustering accuracy of unknown faults compared to stand-alone techniques.

Compositional verification of embedded real-time systems

In an embedded real-time system (ERTS), real-time tasks (software) are typically executed on a multicore shared-memory platform (hardware). The number of cores is usually small, contrasted with a larger number of complex tasks that share data to collaborate. Since most ERTSs are safety-critical, it is crucial to rigorously verify their software against various real-time requirements under the actual hardware constraints (concurrent access to data, number of cores). Both the real-time systems and the formal methods communities provide elegant techniques to realize such verification, which nevertheless face major challenges. For instance, model checking (formal methods) suffers from the state-space explosion problem, whereas schedulability analysis (real-time systems) is pessimistic and restricted to simple task models and schedulability properties. In this paper, we propose a scalable and generic approach to formally verify ERTSs. The core contribution is enabling, through joining the forces of both communities, compositional verification to tame the state-space size. To that end, we formalize a realistic ERTS model where tasks are complex with an arbitrary number of jobs and job segments, then show that compositional verification of such model is possible, using a hybrid approach (from both communities), under the state-of-the-art partitioned fixed-priority (P-FP) with limited preemption scheduling algorithm. The approach consists of the following steps, given the above ERTS model and scheduling algorithm. First, we compute fine-grained data sharing overheads for each job segment that reads or writes some data from the shared memory. Second, we generalize an algorithm that, aware of the data sharing overheads, computes an affinity (task-core allocation) guaranteeing the schedulability of hard-real-time (HRT) tasks. Third, we devise a timed automata (TA) model of the ERTS, that takes into account the affinity, the data sharing overheads and the scheduling algorithm, on which we demonstrate that various properties can be verified compositionally, i.e., on a subset of cores instead of the whole ERTS, therefore reducing the state-space size. In particular, we enable the scalable computation of tight worst-case response times (WCRTs) and other tight bounds separating events on different cores, thus overcoming the pessimism of schedulability analysis techniques. We fully automate our approach and show its benefits on three real-world complex ERTSs, namely two autonomous robots and an automotive case study from the WATERS 2017 industrial challenge.

Partial-order reduction in reachability-based response-time analyses of limited-preemptive DAG tasks

Response-time analysis (RTA) has been a means to evaluate the temporal correctness of real-time systems since the 1970 s. While early analyses were successful in capturing the exact upper bound on the worst-case response-time (WCRT) of systems with relatively simple computing platforms and task activation models, nowadays we see that most existing RTAs either become pessimistic or do not scale well as systems become more complex (e.g., parallel tasks running on a multicore platform). To make a trade-off between accuracy and scalability, recently, a new reachability-based RTA, called schedule-abstraction graph (SAG), has been proposed. The analysis is at least three orders of magnitude faster than other exact RTAs based on UPPAAL. However, it still has a fundamental limitation in scalability as it suffers from state-space explosion when there are large uncertainties in the timing parameters of the input jobs (e.g., large release jitters or execution-time variations). This could impede its applicability to large industrial use cases, or to be integrated with automated tools that explore alternative design choices. In this paper, we improve the scalability of the SAG analysis by introducing partial-order reduction rules that avoid combinatorial exploration of all possible scheduling decisions. We include systems with dependent and independent task execution models (i.e., with and without precedence constraint). Our empirical evaluations show that the proposed solution is able to reduce the runtime by five orders of magnitude and the number of explored states by 98% in comparison to the original SAG analysis. These achievements come only at a negligible cost of an over-estimation of 0.1% on the actual WCRT. We applied our solution on an automotive case study showing that it is able to scale to realistic systems made of hundreds of tasks for which the original analysis fails to finish.

An automotive case study on the limits of approximation for object detection

The accuracy of camera-based object detection (CBOD) built upon deep learning is often evaluated against the real objects in frames only. However, such simplistic evaluation ignores the fact that many unimportant objects are small, distant, or background, and hence, their misdetections have less impact than those for closer, larger, and foreground objects in domains such as autonomous driving. Moreover, sporadic misdetections are irrelevant since confidence on detections is typically averaged across consecutive frames, and detection devices (e.g. cameras, LiDARs) are often redundant, thus providing fault tolerance.This paper exploits such intrinsic fault tolerance of the CBOD process, and assesses in an automotive case study to what extent CBOD can tolerate approximation coming from multiple sources such as lower precision arithmetic, approximate arithmetic units, and even random faults due to, for instance, low voltage operation. We show that the accuracy impact of those sources of approximation is within 1% of the baseline even when considering the three approximate domains simultaneously, and hence, multiple sources of approximation can be exploited to build highly efficient accelerators for CBOD in cars.

Systematic modeling of complex time-variant gear systems using a Power-Oriented approach

This paper addresses a methodology for the systematic modeling of complex gear systems. The methodology is based on the use of a unified general model, working for all complex gear systems: time-variant as well as time-invariant, having parallel or oblique rotation axes. The model equations are automatically written following the outlined procedure and applying the presented algorithms, making this approach less prone to mistakes with respect to other approaches. Next, a reduced model assuming rigid gear connections and introducing no loss of information is proposed, which directly gives the kinematic relations between the gears angular speeds and input torques. In order to show some case studies, the proposed methodology is applied to three systems of interest for vehicle dynamics and powertrain modeling. The considered case studies are a differential structure having a bevel gearing system with non-perpendicular gear shafts, a vehicle differential and a full toroidal variator, which is suitable for applications such as KERS (Kinetic Energy Recovery System) and IVT (Infinitely Variable Transmission). Furthermore, the control of a full toroidal variator acting as a KERS with reference to an automotive case study and the comparison of the proposed modeling methodology with two other approaches are addressed.

MPC-Based Process Control of Deep Drawing: An Industry 4.0 Case Study in Automotive

Deep drawing is a metalworking procedure aimed at getting a cold metal sheet plastically deformed in accordance with a pre-defined mould. Although this procedure is well-established in industry, it is still susceptible to several issues affecting the quality of the stamped metal products. In order to reduce defects of workpieces, process control approaches can be performed. Typically, process control employs simple proportional-integral-derivative (PID) regulators that steer the blank holder force (BHF) based on the error on the punch force. However, a single PID can only control single-input single-output systems and cannot handle constraints on the process variables. Differently from the state of the art, in this paper we propose a process control architecture based on Model Predictive Control (MPC), which considers a multi-variable system model. In particular, we represent the deep drawing process with a single-input multiple-output Hammerstein-Wiener model that relates the BHF with the draw-in of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$n$ </tex-math></inline-formula> different critical points around the die. This allows the avoidance of workpiece defects that are due to the abnormal sliding of the metal sheet during the forming phase. The effectiveness of the proposed process controller is shown on a real case study in a digital twin framework, where the performance achieved by the MPC-based system is analyzed in detail and compared against the results obtained through an ad-hoc defined multiple PID-based control architecture. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —This work is motivated by the emerging need for the effective implementation of the zero-defect manufacturing paradigm in the Industry 4.0 framework. Especially in the deep drawing process, various quality issues in stamped parts can lead to significant product waste and manufacturing inefficiencies. This turns into considerable economic losses for companies, particularly in the automotive sector, where deep drawing is one of the most used cold sheet metal forming techniques. In most applications, only sample inspections are performed on batches of finished-product, with subsequent losses of time and resources. For the sake of improving the workpiece quality, innovative strategies for real-time process control represent a viable and promising solution. In this context, the proposed MPC-based process control approach allows the correct shaping of the metal sheet that is getting deformed during the forming stroke, thanks to the draw-in monitoring at various locations around the die. The draw-in is indeed one of the most effective forming variables to control in order to provide a correct BHF during the forming stroke. A useful and easy-to-implement non-linear metal sheet deep drawing process model is provided by this paper to perform an innovative process control strategy. A comprehensive methodology is applied in detail to an automotive case study, ranging from process modeling (model identification and validation based on experimental data acquisition) to MPC implementation (controller tuning and testing and software-in-the-loop system validation). The presented method can be easily implemented on any real deep drawing press, providing the multivariable constrained process with a suitable control system able to make the stamped parts well formed.

How did supply chain networks handle the COVID-19 pandemic? Empirical evidence from an automotive case study

PurposeThe coronavirus disease 2019 (COVID-19) pandemic unveiled resilience deficits in supply chains. Scholars and practitioners aim to identify supply chain resilience (SCRES) measures suitable for this unique disruption; however, empirical evidence on a pandemic's specific characteristics, resulting challenges, and suitable countermeasures is scarce.Design/methodology/approachA single-case study on an automotive supply chain network (ASCN), including eight nodes, was conducted. Based on current research and interviews with 35 experts, characteristic pandemic challenges for the ASCN were identified. Moreover, promising SCRES measures were determined along the most prominent SCRES levers. The findings lead to five central propositions and advance organizational information processing theory in the context of SCRES.FindingsThis study’s results confirm unique pandemic characteristics along the supply chain disruption's duration, severity, propagation, and volatility. The resulting unprecedented challenges made the ASCN apply novel SCRES measures, particularly regarding collaboration and risk management culture. However, well-known visibility and flexibility strategies were also suitable. Overall, agility and collaboration measures showed the highest capacity to address characteristic pandemic challenges. A lack of preparedness impeded some measures' application, calling for enhanced proactive risk management following the pandemic.Originality/valueThis paper addresses several research calls by providing in-depth empirical evidence on hitherto conceptually researched pandemic characteristics, challenges, and suitable SCRES measures from a network perspective. The study uncovers the different perceptions of individual tiers, emphasizing the need to analyze supply chain disruptions from multiple angles.

Operator training framework for hybrid environments: An Augmented Reality module using machine learning object recognition

As market demands are characterized by more customized products with shorter lifecycles, it is obligatory for modern operators to manage recurrent product or manufacturing system changes. In contrary to previous years, adaptation to such changes prerequires memorization of more information, and familiarization with more complex systems and resources in a shorter period of time. This manuscript presents a novel operator training framework based on Augmented Reality (AR) technology. More specifically, intuitive instructions enhanced with machine learning-based physical object detection are used for making steeper learning curves and providing hands-on experience to operators. The implemented application also supports a walkthrough mode where users can get familiarized with Information and Communication Technologies (ICT) data streams besides fenceless Human-Robot coexistence in collaborative schemes. An automotive case study is used for evaluating the performance of the training framework through a Human-Robot Collaboration (HRC) assembly scenario.

System reconfiguration for reverse logistics: A case study

Manufacturers need to effectively address returned products in their system, and challenges occur in the remanufacturing processes due to the uncertainty related to the quantities and quality of the returned components. This research focuses on identifying the challenges encountered in a remanufacturing Reverse Logistics (RL) system and are illustrated with an automotive case study. Lean Six-sigma techniques are used to find these issues encountered in the RL process or system, and a linear programming approach is taken to reconfigure the system to improve the throughput. Additional analyses need to be performed to explore the influence of different production strategies and system layouts.