How Shall Data Science And AI-Driven CAE Reduce Overall Lead Time And Product Development Cost From A Durability Perspective?

<p style="text-align: justify;">&nbsp;</p><p style="text-align: justify;">Data Science and Artificial Intelligence (AI) is no longer a buzzword. It is a crucial contributor to the global automotive megatrends, from autonomous vehicles to alternate powertrain technologies such as Electric mobility, Fuel cell, and many more.</p><p><img style="display: block; margin-left: auto; margin-right: auto;" src="" alt="auto industry trends" width="754" height="452" /></p><h2><span style="font-size: 14pt;">Role of Data science and AI</span></h2><p>Data Science is a multidisciplinary approach to extract actionable insights from the large and ever-increasing volumes of data collected and created by today's organizations as per the definition of IBM. Here's the key role of data science and AI in the automotive industry:&nbsp;</p><p><img style="display: block; margin-left: auto; margin-right: auto;" src="" alt="role of data science and AI in the automotive industry" width="841" height="308" /></p><p style="text-align: justify;">Artificial Intelligence is the science and engineering of making intelligent machines and brilliant computer programs. It is related to the similar task of using computers to understand human intelligence, but AI does not have to confine itself to biologically observable methods, according to John McCarthy, Stanford University.</p><p style="text-align: justify;">Machine learning uses data and algorithms to simulate the way humans learn and gradually improve its accuracy. Deep learning attempts to imitate the human brain and enables systems to make predictions with incredible accuracy.</p><p style="text-align: justify;">There have been significant software advancements in the last 15 years in Computer-Aided Engineering (CAE) that is specific to automotive R&amp;D.</p><p style="text-align: justify;">Durability of components, systems, and vehicles are usually done as per Design Validation Plan (DVP). Nearly 80 to 120 load cases shall be simulated to clear the durability criteria at all levels. Here's a classical workflow of vehicle durability:</p><p><img style="display: block; margin-left: auto; margin-right: auto;" src="" width="846" height="356" /></p><p style="text-align: justify;">Based on use cases, Data science and AI-driven CAE could predict the durability &nbsp;simulation outcomes in minutes or seconds for subsystem level, be it stress, displacement, strain, Eigen values or modes, mode shapes, thermal stress/strain, etc.</p><h2 style="text-align: justify;"><span style="font-size: 14pt;">Benefits of Data Science and AI in Durability Simulation</span></h2><p style="text-align: justify;">Data science and AI could largely influence the Design of Experiments (DOE) to evaluate multiple design points in less time, specifically focusing on the vital design points that could produce better results with high accuracy, known as Reduced Order Modeling. This helps with quicker design decision-making, thus reducing the product cycle time and cost.</p><p style="text-align: justify;">Classical simulation practices in durability development, verification, and validation can now powered with Data Science and AI to create digital prototyping. State-of-the-art commercial CAE software companies had already linked the initial phase of data analytics and predictive modeling capabilities.</p><p style="text-align: justify;">Since it is futuristic, we need to train the Software-in-the-loop and Hardware-in-the-loop processes to understand the expected outcomes with phenomenal accuracy. This might take more time initially than we expected.</p><p style="text-align: justify;">However, it drastically reduces the future product development time from months to weeks, weeks to days, days to hours, hours to minutes, and even minutes to seconds.</p><p style="text-align: justify;">It depends on the data availability. In case data is not organized or unavailable, companies need to start capturing the data and follow the data science workflow to extract actionable insights for further developments.</p><p style="text-align: justify;"><span style="font-size: 10pt;"><em>This article was contributed by our expert <a href="">Venkat Anumula</a></em></span></p><h3 style="text-align: justify;"><span style="font-size: 18pt;">Frequently Asked Questions Answered by &nbsp;Venkat Anumula</span></h3><h2><span style="font-size: 12pt;">1. What is durability analysis? &nbsp;</span></h2><p style="text-align: justify;">Durability analysis determines the overall life requirements of a component or subsystem, system, and vehicle. For instance, specifying the product life as it lasts for three years. The product warranty is decided based on durability evaluation.</p><h2 style="text-align: justify;"><span style="font-size: 12pt;">2. What is the role of Artificial Intelligence (AI) in a simulation? </span>&nbsp;</h2><p style="text-align: justify;">AI speeds up the simulation process for new or modified designs in less time. For this to happen, we need to build an AI model based on the data availability and train the model with the existing simulation process.&nbsp;</p><p style="text-align: justify;">The trained model needs to be tested on large sample size, and validation must be done at several phases of AI model development. Once the model is validated during pre- and post-development, the AI could leap forward to predicting a new design's durability or fatigue life in minutes or even seconds.</p><h2 style="text-align: justify;"><span style="font-size: 12pt;">3. What are the factors affecting lead time? </span>&nbsp;</h2><ul><li style="text-align: justify;">3D CAD model &nbsp;</li><li style="text-align: justify;">Poor product design</li><li style="text-align: justify;">Non-use or unavailability of direct modeling or repair tools that are available for CAE&nbsp;</li><li style="text-align: justify;">The skill gap in both product design and CAE</li><li style="text-align: justify;">Lack of automation for standard simulations or repeated simulations &nbsp;</li><li style="text-align: justify;">Lack of new methodologies, simulation data capture, and knowledge base &nbsp;</li><li style="text-align: justify;">Lack of simulation tool for product designers to extract better design before sharing it with CAE teams &nbsp;</li><li style="text-align: justify;">Non-upgrade of simulation tools over a prescribed period of time &nbsp;</li><li style="text-align: justify;">Lack of readiness with future technologies&nbsp;</li></ul><h2><span style="font-size: 12pt;">4. What are the advantages of Artificial Intelligence (AI) in durability simulation?</span></h2><p>Based on the use cases, there's a high possibility of achieving the following key benefits after validating the AI model. &nbsp;</p><ul><li>Predicting the fatigue life or durability in less time (minutes or seconds) &nbsp;</li><li>Reduces simulation lead time and cost to a large extent resulting in quick decision making &nbsp;</li><li>AI models gain more confidence resulting in phenomenal accuracies gradually &nbsp;</li><li>AI models shall build an extensive knowledge base based on their learning and can recommend alternatives at the end of the prediction</li><li>AI human interface with voice recognition system lets humans talk to intelligent machines that could predict seamless results &nbsp;</li></ul><h2><span style="font-size: 12pt;">5. What are the main advantages of Data Science?</span></h2><ul><li>Ability to predict and extract intellectual insights from the data based on the patterns &nbsp;</li><li>Faster decision-making (business or technical) &nbsp;</li><li>Hi-fidelity and consistency with periodic checks &nbsp;</li><li>Reduced product development time and costs on a large scale &nbsp;</li><li>Readiness to self-diagnostics &nbsp;</li><li>Smart manufacturing and Industrial Internet Of Things compliances (IIOT) &nbsp;</li><li>Predictive machinery maintenance even before the breakdown is expected</li></ul><p style="text-align: justify;">&nbsp;</p>
KR Expert - Venkat Anumula