“As standards continue to evolve, applications spanning multiple cloud environments are increasing, and user expectations for “all-in-one” products are increasing, all of which make today’s new Electronic products increasingly complex. The increase in complexity directly affects the design, development, simulation and test coverage of new electronic products, putting a lot of pressure on development teams.
“
By: Jeff Harris, Vice President, Global Corporate and Product Marketing, Keysight Technologies
Jeff Harris, Vice President, Global Corporate and Product Marketing, Keysight Technologies
As standards continue to evolve, applications spanning multiple cloud environments are increasing, and user expectations for “all-in-one” products are increasing, all of which make today’s new electronic products increasingly complex. The increase in complexity directly affects the design, development, simulation and test coverage of new electronic products, putting a lot of pressure on development teams.
Among the many development directions of testing, the most mainstream direction is still to automate the design and testing process, and intelligently gain insight into the entire workflow DD, that is, automated artificial intelligence. However, according to a recent Forrester survey commissioned by Keysight, 89% of companies are still using manual processes and only 11% have fully automated their test matrix. While adoption of full automation remains low, companies do see value in automation, with 75% partially automating processes and nearly half expecting full automation within the next three years.
Artificial intelligence, machine learning and digital twins are gaining traction in the development of complex electronic systems
In December 2021, Keysight commissioned Forrester Consulting to evaluate the use of technologies such as data integration, analytics, artificial intelligence (AI) and machine learning (ML) in a typical product development cycle. Forrester surveyed more than 400 development leaders and asked a series of questions about their current use of AI and ML in product development.
Initially, we heard from the majority of organizations that they were satisfied with the development methodologies they are currently using, with 86% being satisfied or very satisfied. However, these same organizations said that 84 percent of their projects and designs employed either complex multi-layered subsystems or integrated systems, most of which were not tested.
Although the company seemed satisfied at first, we learned through our survey that they felt pressure when asked about increasing the automation and intelligence of their electronic design process, especially in the future.
Currently, only 10% of companies have fully automated design and testing in the development process, but the COVID-19 pandemic has forced businesses to accelerate the adoption of remote development and automated testing sequences. The development team is also striving for continuous collaboration between people working in different locations, so the use of digital twins is likely to increase further.
Digital Twins and Simulation: A New Model for Electronic Design
Hardware developers have long relied on simulation environments to design hardware before prototyping. Using a software-driven simulator or digital twin enables them to measure the impact of different operating environments, conditions, and protocol evolutions against a known good reference system, reducing the number of design iterations. Likewise, software developers can build and deploy new features incrementally, using methods such as Scrum and testing them during virtual simulation, which also helps reduce the number of design iterations.
Communication protocols and cloud platforms continue to evolve, and software and firmware are constantly updated, resulting in increasingly complex interactions between electronic products, posing a real challenge to developers, because every evolution and update will bring a series of new changes that require rigorous testing . By maximizing the use of test automation and continuously updated digital twins, development teams are able to test more changes and reduce the risk of specific design issues.
Automating artificial intelligence in electronic design workflows
Automation is fast becoming a must. Currently, fully manual test plans based on manual data entry, some Python or graphical programming, and Excel spreadsheets can only satisfy a small subset of possible user scenarios. Designers need to manually update test plans every time a new version of the software is released, further delaying the electronic design cycle.
However, while test automation software can solve some of these problems and is therefore indispensable, it is not enough. The effectiveness of test automation is determined by the analytics and insights they generate. Respondents to the Forrester survey revealed that more than half of their test cases were “beyond necessary.” Test automation helps reduce test time, but does not address issues such as test coverage, test quality, and coverage. Using analytics and insights, designers will be able to automate AI in the design workflow and execute a wider range of test sequences, ensuring excellent test speed while still covering the ideal test coverage.
As a software model, automated AI builds on Keysight’s vast array of measurement technologies and simulation capabilities to provide developers with rapid insights that help them bring designs to market faster and minimize risk. Whether measuring power and ground, waveform signal quality, high-speed data I/O, network integrity or application delivery, we must consider how we can help our customers speed up the development process.
What are the successful hallmarks of automated AI?
In the past, when strategizing for new development projects, people often found that “fast, good, and savings can’t be all at the same time.” If all else remains the same, this conclusion may still hold today. However, by integrating automated AI into your development workflow, you may be able to do all three:
• Faster: enables faster time-to-market
• Better: Provide better products and make customers more satisfied
• Less: Make product development processes more agile and efficient
Development teams have paid off with this approach. Whether product development involves emerging electronics using the latest wireless communication standards, high-speed data transfer, complex cloud networks or distributed application software delivery, the focus is the same. Build your lab design and test solution to deliver insightful analysis at every stage. This is complemented by AI and ML so they can always explore new room for improvement. Automating the development environment as you do in the manufacturing phase will minimize development time while ensuring the best possible product performance.
“As standards continue to evolve, applications spanning multiple cloud environments are increasing, and user expectations for “all-in-one” products are increasing, all of which make today’s new Electronic products increasingly complex. The increase in complexity directly affects the design, development, simulation and test coverage of new electronic products, putting a lot of pressure on development teams.
“
By: Jeff Harris, Vice President, Global Corporate and Product Marketing, Keysight Technologies
Jeff Harris, Vice President, Global Corporate and Product Marketing, Keysight Technologies
As standards continue to evolve, applications spanning multiple cloud environments are increasing, and user expectations for “all-in-one” products are increasing, all of which make today’s new electronic products increasingly complex. The increase in complexity directly affects the design, development, simulation and test coverage of new electronic products, putting a lot of pressure on development teams.
Among the many development directions of testing, the most mainstream direction is still to automate the design and testing process, and intelligently gain insight into the entire workflow DD, that is, automated artificial intelligence. However, according to a recent Forrester survey commissioned by Keysight, 89% of companies are still using manual processes and only 11% have fully automated their test matrix. While adoption of full automation remains low, companies do see value in automation, with 75% partially automating processes and nearly half expecting full automation within the next three years.
Artificial intelligence, machine learning and digital twins are gaining traction in the development of complex electronic systems
In December 2021, Keysight commissioned Forrester Consulting to evaluate the use of technologies such as data integration, analytics, artificial intelligence (AI) and machine learning (ML) in a typical product development cycle. Forrester surveyed more than 400 development leaders and asked a series of questions about their current use of AI and ML in product development.
Initially, we heard from the majority of organizations that they were satisfied with the development methodologies they are currently using, with 86% being satisfied or very satisfied. However, these same organizations said that 84 percent of their projects and designs employed either complex multi-layered subsystems or integrated systems, most of which were not tested.
Although the company seemed satisfied at first, we learned through our survey that they felt pressure when asked about increasing the automation and intelligence of their electronic design process, especially in the future.
Currently, only 10% of companies have fully automated design and testing in the development process, but the COVID-19 pandemic has forced businesses to accelerate the adoption of remote development and automated testing sequences. The development team is also striving for continuous collaboration between people working in different locations, so the use of digital twins is likely to increase further.
Digital Twins and Simulation: A New Model for Electronic Design
Hardware developers have long relied on simulation environments to design hardware before prototyping. Using a software-driven simulator or digital twin enables them to measure the impact of different operating environments, conditions, and protocol evolutions against a known good reference system, reducing the number of design iterations. Likewise, software developers can build and deploy new features incrementally, using methods such as Scrum and testing them during virtual simulation, which also helps reduce the number of design iterations.
Communication protocols and cloud platforms continue to evolve, and software and firmware are constantly updated, resulting in increasingly complex interactions between electronic products, posing a real challenge to developers, because every evolution and update will bring a series of new changes that require rigorous testing . By maximizing the use of test automation and continuously updated digital twins, development teams are able to test more changes and reduce the risk of specific design issues.
Automating artificial intelligence in electronic design workflows
Automation is fast becoming a must. Currently, fully manual test plans based on manual data entry, some Python or graphical programming, and Excel spreadsheets can only satisfy a small subset of possible user scenarios. Designers need to manually update test plans every time a new version of the software is released, further delaying the electronic design cycle.
However, while test automation software can solve some of these problems and is therefore indispensable, it is not enough. The effectiveness of test automation is determined by the analytics and insights they generate. Respondents to the Forrester survey revealed that more than half of their test cases were “beyond necessary.” Test automation helps reduce test time, but does not address issues such as test coverage, test quality, and coverage. Using analytics and insights, designers will be able to automate AI in the design workflow and execute a wider range of test sequences, ensuring excellent test speed while still covering the ideal test coverage.
As a software model, automated AI builds on Keysight’s vast array of measurement technologies and simulation capabilities to provide developers with rapid insights that help them bring designs to market faster and minimize risk. Whether measuring power and ground, waveform signal quality, high-speed data I/O, network integrity or application delivery, we must consider how we can help our customers speed up the development process.
What are the successful hallmarks of automated AI?
In the past, when strategizing for new development projects, people often found that “fast, good, and savings can’t be all at the same time.” If all else remains the same, this conclusion may still hold today. However, by integrating automated AI into your development workflow, you may be able to do all three:
• Faster: enables faster time-to-market
• Better: Provide better products and make customers more satisfied
• Less: Make product development processes more agile and efficient
Development teams have paid off with this approach. Whether product development involves emerging electronics using the latest wireless communication standards, high-speed data transfer, complex cloud networks or distributed application software delivery, the focus is the same. Build your lab design and test solution to deliver insightful analysis at every stage. This is complemented by AI and ML so they can always explore new room for improvement. Automating the development environment as you do in the manufacturing phase will minimize development time while ensuring the best possible product performance.
The Links: NL12876BC15-01 LQ064V3DG01 IGBTS