Artificial Intelligence and the Origin of Digital Twins | Technology
|Artificial Intelligence and the Origin of Digital Twins|
The Origins of Digital Twins and Artificial Intelligence
In the race to implement AI and digital twin technologies, there are important questions and processes that enterprises need to consider when evaluating products.
Artificial intelligence, machine learning, and the digital twins - why are we hearing so much about them and why do they suddenly seem serious? The simplest explanation is this: when something is so complex that a person cannot easily process it or has little time to make an important decision, the only option is to remove the human. This requires the ability to replicate the thought process that humans may go through, which requires a great deal of data and a deep understanding of the decision environment.
So why now? Over the decades, we have seen great progress mainly from the integration and shrinkage of electronics. Smaller products, lower power consumption, and offering a dramatic increase in functionality per square inch were hallmarks of technology advancements.
Software applications have also evolved over the decades, one of the most notable ways is the dramatic acceleration of the application adoption cycle. In just the last two decades, users have moved at alarmingly fast rates from treating applications as a novelty, to using them as a convenience, and then to expecting them to work flawlessly all the time. Is. With each adoption stage, user expectation increases, which means the product must evolve and mature at very fast, scalable rates.
The combination of hardware and software trends led to the convergence of product development needs. New "critical need" applications must suddenly have high potential for real-time processing, time-sensitive decision making, high to very high availability, and the expectation that platform-generated decisions are correct every time.
While most people think of AI primarily as an end user resource, AI has increasingly become essential to product design and development. Emulators have become essential for building complex interfaces and environments, from the early stages of chipset design or layout of circuits through to end-product validation. These emulators, known as digital twins, are a virtual manifestation of a process, environmental condition or protocol capable of serving as a "known good signal". In terms of testing, a digital twin can be a simple signal generator, a complete protocol generator, or a complete environmental emulator. Digital twins allow developers to rapidly create a wide range of test conditions to validate their product before shipping. High-performance digital twins typically have their own AI engines for troubleshooting and regression testing new product designs.
AI-powered evolution and the digital twin
The shift to AI-driven development and digital twins has become necessary due to the amount of functionality and autonomous decision making expected in new products. Basic design principles specify a product's features and functionality, then set up tests to verify them. The sheer number and complexity of interface standards make it nearly impossible to build by hand. Using Digital Twins, a comprehensive set of functional tests can be programmed in a very short time. AI functionality then automates the testing processes it discovers and predicts actions that may be needed. To better understand this, it is useful to understand the core of what makes any AI possible.
In its simplest form, software decision making begins with algorithms. Basic algorithms run a set of calculations, and if you know what the acceptable results are, you can build a finite state machine using the decision tree results. It would hardly be considered intelligent. However, by adding a notation of the state, and inserting a feedback loop, your basic algorithm can make the outcome decisions more a function of the current conditions than the current state. Combine this while evolving the decision tree into a behavior tree and you have formed the origin of AI.
The need for AI and digital twins is real, and when you question the veracity of one - yours or someone else's - go back to its origins, otherwise known as data. Data sources are the foundation of any digital evaluation tool, and they determine the potential for modeling accuracy of source algorithms. If multiple data-rich sources are available, the accuracy is likely higher. If only the original data is available, the resulting algorithm or digital twin will not be accurate. This is something you can assess yourself.
Here are the steps to assess the potential of any AI algorithm or digital twin
- Create a crude diagram of the closed-loop decision process – inputs, situation considerations, outputs – that the AI is supposed to replicate or the environment the digital twin should emulate. Write down as many variables as possible to brainstorm. Do not take more than 30 minutes in this step.
- In the case of an AI algorithm, look at the data sources used by the vendor. In the case of Digital Twin, refer to the vendor's system performance specifications and background knowledge. Their collective depth is proportional to the algorithm's capacity. This may take an hour or two of research.
- Ask the seller a ton of questions based on what you learned in Steps 1 and 2. Clarity - or lack thereof - will rapidly shape your understanding.
We are in the early stages of AI, which means a lot of products will make a lot of claims. Understanding what a product has to deliver will allow you to assess it. Understanding what data sources it processes will tell you how accurately it can deliver on what the vendor promises. Digital twins are far ahead in maturity – especially those that emulate specific elements rather than the entire ecosystem. Remember, however, that the more finite the environment, the more likely the digital twin will accurately replicate it.
We all want to understand how something works and how it produces results. With an understanding of the fundamentals of each AI system and digital twin, you can ask questions about their fundamentals. If you get stuck, use the steps as a guide for questions to ask the seller. Most will share all or some of the key background or parameters to help you understand. Not me, his competitors will.
Source: Jeff Harris, Information Week, Direct News 99