Industry
Manufacturing
Future-Ready Manufacturing: Partnering in your journey to a sustainable manufacturing enterprise
Highlights
- Generative artificial intelligence (GenAI) can solve some of the most pressing challenges in the automotive industry, making automakers a lot more competitive and future ready.
- Some areas where GenAI will have a significant impact include research and development (R&D), product engineering, customer experience, vehicle safety, and supply chain.
- Manufacturers, at the same time, need to be aware of the risks and challenges of GenAI. Data breach, algorithmic biasness, and doubts over outcomes raise legal and ethical queries which need to be addressed by the industry.
On this page
On this pageinpage
The GenAI effect
The automotive industry is evolving at an unprecedented pace with increased focus on electric, connected, and autonomous vehicles.
Technologies such as artificial intelligence (AI), cloud, virtual reality, and IoT are acting as catalysts to that change. Generative artificial intelligence (GenAI) has the capability to truly transform the way automotive businesses are run. It can help with forecasting demand with a great deal of accuracy, manage inventory a lot better, and take customer experience several notches up. When integrated with manufacturing process, GenAI will shorten design cycles, fuel product innovation, and boost sales and marketing efforts. By addressing the challenges of longer design cycles, vehicle safety, warranty payouts against fraudulent claims, and higher service repair time in the absence of the right diagnostic mechanisms, GenAI will enable automakers to drive profitable growth and keep pace with market dynamics.
Exploring the vast scope
Automakers can leverage GenAI across the product lifecycle to create maximum value.
In fact, some original equipment manufacturers (OEMs) are already experimenting with it. Mercedes-Benz, for instance, integrated ChatGPT with its in-car voice control infotainment system, MBUX Voice Assistant, to elevate driving experience. The technology, integrated in about 900,000 Mercedes vehicles through Azure OpenAI service, helps drivers to inquire about their destination or answer complex questions. Toyota Research Institute (TRI) unveiled a GenAI technique last year to amplify vehicle designers’ capabilities. With this, Toyota designers can add initial design sketches and engineering constraints into their creative process, thus cutting down the iterations needed to reconcile design and engineering considerations.
Figure 1 shows a few use cases for GenAI across the automotive value chain.
Figure 1: GenAI use cases for the automotive industry
- Design optimization, product development, and core product engineering: GenAI can be used to create multiple design options faster, based on input criteria. It accelerates computational fluid dynamics to minutes, from hours or days, allowing the design cycle to complete faster. During design analysis and design optimization, AI-enabled attributes like cost capacity and material availability can be used to tune up the products and lower the number of design iterations. Overall, GenAI will help automakers reduce the number of parts needed, improve design efficiency, and be more compliant with regulations.
- Connected supply chain: GenAI can analyze data on suppliers, production processes, inventory, and logistics to optimize the production and supply chains. With conversational experience in process planning, engineers can design these processes using AI-based models. That will help them with strategies to forecast demand, reduce lead times, improve product quality, nurture sustainability, and ensure a well-organized inventory management. Visualization of new process models can also be iteratively created using GenAI models.
- Sales process: GenAI can help automotive companies improve the customer journey during the sales process by automating a large part of backend processing. It can streamline the submission of loan documents, improve fraud detection capabilities, and enable personalized solutions.
- Customer experience (CX) management: GenAI can help in creating a comprehensive profile of the customer. As customers progress through life stages, their requirements and preferences change; this technology can help the company to understand them throughout that journey. GenAI can analyze the preferences, patterns, and behavior data to build a customer persona that would guide how the automotive company engages the customer across touchpoints. GenAI has the potential to transform customer experience by reforming navigation with optimized routes, establishing personalized journeys based on user preferences, and tailoring driving experiences based on user behavior. For instance, it can improve the buying experience and deepen brand loyalty by suggesting vehicles matching a customer’s needs. For aftersales, GenAI can be used to support customers by suggesting more accurate maintenance timing and automatically scheduling appointments.
- Service desk operations: GenAI can transform service desk operations by leveraging large language models (LLMs) to provide more contextual, precise, and useful knowledge base article search. It can curate knowledge articles based on past conversation or transcripts, thereby eliminating the need for running a manual search, which is time-consuming and may not be entirely error-free.
- A safety net: GenAI can also transform automotive safety by leveraging real-time data. That will enhance accident detection rates, upgrade advanced driver assistance systems (ADAS), better flexible cruise control, and improve the ability of the system to learn the patterns on the sign boards. This blend of adaptability and data-driven insights will significantly improve road safety standards. The in-built virtual assistant can help the driver converse with the car to guide him through the process during emergencies.
- Vehicle infotainment: GenAI-based virtual assistant models can seamlessly integrate with vehicle infotainment systems to offer a variety of voice-activated functionalities. The in-car virtual assistant can enable drivers to ask the car what to do when there is a fault and guide them through the process during emergencies.
- Enterprise operations: Enterprise operations involve application management services (AMS) and infrastructure management services (IMS). GenAI can reduce the run operations and maintenance cost by applying predictive self-healing techniques.
The complexities of the GenAI architecture
The three key building blocks of the GenAI architecture are data processing, machine learning (ML) models, and feedback loops.
The architecture should be designed such that it generates new and original content based on input data or rules. It should include digital platforms, utility building blocks, external applications, and core building blocks, along with stakeholders. Gen AI feedback and improvement, generative model, and data processing are the key layers related to GenAI.
The architectural building blocks of GenAI may vary slightly based on the selected use cases and the respective manufacturing enterprise architecture decisions. Figure 2 shows the standard building blocks of such a layered architecture.
Figure 2: A reference architecture for GenAI in automotive manufacturing
The manufacturing stakeholders, which include customers, dealers, distributors, and partners, sit at the top of the architecture. This is followed by the user interface (UI) layer, services layer, and business layer which already exist within an enterprise. The Gen AI feedback and improvement layer, which sits below the business layer, will continuously improve the architecture’s accuracy and efficiency with user feedback. Digital platforms, utility building blocks, cloud, and external apps will be the standard building blocks of the architecture.
The generative model layer, the penultimate one, will be responsible for creating new content or data through various ML models. These models will leverage deep learning or reinforcement learning, depending on the use case and data type to be generated. Finally, the data processing layer will be used to collect, prepare, and process the data to be used by the ML models. Within that base layer, the data collection block will gather data and the data preparation block will clean it. The feature extraction block will identify the most relevant features of the data.
Navigating with caution
While GenAI offers several benefits, the automotive industry also needs to be cognizant of the risks and challenges.
The most common risks are data security, biasness of the system, and trust in the outcome. All of these can have adverse and serious consequences, including the loss of human life. Some of the risks are:
Data privacy: GenAI often requires access to personal information to authenticate authorized users. Sensor data is collected to help the vehicle understand where it is relative to other objects on the road. AI uses the dataset associated with a particular vehicle to personalize and enhance navigation features to plan personalized routes for drivers. If a hacker gains access to this data, information about the owner or passengers, such as where they live and work and the specific locations they frequent, could be compromised and misused.
Manufacturers should also restrict the collection and retention of personal data to only what is needed for the system to function properly in order to reduce the risk of potential privacy breaches. Before using data for training GenAI, personal information should be pseudonymized to ensure individuals cannot be identified from generated outputs.
Manufacturers using GenAI tools should clearly communicate to data subjects (vehicle occupants) their data collection, data storage, and data usage practices, and should only process personal data for disclosed purposes. GDPR and the EU AI Act could impact how data is handled, processed, protected, secured, and used. Manufacturers must thoroughly analyze the impact of such regulations on their AI-based tools or products. Machine learning models may need to be retrained when the data subject withdraws their consent.
Algorithmic bias and fairness: If the training data sets in the models contain biased information, the resulting AI systems will be biased in their decision-making. This can produce discriminatory decisions, which can have serious legal and ethical implications. A human-in-the -loop is needed to identify outlier prompts and responses requiring human intervention.
Data and model security: AI systems rely on large amounts of data and security of that data is paramount. Hackers can manipulate the inputs to the models to produce incorrect outputs. This can lead to incorrect decisions, which can have negative consequences. Therefore, it is important to design and develop secure AI models.
Trust: A biased and compromised AI model is bound to have concerns over the authenticity and accuracy of its responses. Other concerns include user identity verification and the potential for misuse of such products. Human-in-loop systems coupled with rigorous real-world testing can go a long way in addressing these concerns.
Intellectual property (IP): Manufacturers will have to understand and find out how they can protect their IPs by using GenAI models. Copyright or IP breaches from third parties will be a bigger cause of concern, as GenAI models can source information from a vast pool of data over the internet.
Explainability (interpretability): Many existing AI systems are not designed for explainability, and they are sometimes opaque in their decision-making processes. It is important to develop explainable AI systems, to build trust and confidence, that provide clear and transparent explanations for their decision-making processes.
A perfect future
A connected, collaborative, and cognitive ecosystem will help automotive companies take the next big leap in the manufacturing process.
And GenAI will play a pivotal role in it. Automakers embracing the new technology will be able to innovate faster, reduce manufacturing time, and improve their go-to-market process, which will help them gain an edge in the competitive landscape. By harnessing GenAI in a smart and ethical way, automakers can drive the future of mobility and realize the dream of a car as an extension of the living room.
