/ TECHNOLOGY'S TIME MACHINE
By Rondekka Moore, VP of Global Business Analytics, Avnet
See how smart engineers hack the future through trend analysis.
Think of trend analysis as your engineering time machine. It transforms messy, real-world data points into a high-resolution snapshot of where technology is heading—not where the hype cycle claims it’s going.
Sure, spotting the next unicorn market might grab headlines, but experienced engineers know that historical patterns pack more predictive punch than any market forecast.
In our tech landscape where disruption is the only constant, trend data doesn’t just tell stories, it maps the DNA of change. For engineering minds hungry for certainty, it’s the difference between building on bedrock and chasing digital mirages.
Making more informed decisions
We often use the word “trend” to describe what we’re seeing at that moment. For example, if you notice electric vehicle charging stations in your neighborhood, is it because more people are buying EVs? Those two observations could be described as trends and may be related, but the natural, human assumption is that they’re also an action and reaction.
What that assumption doesn’t consider is what’s going on between the action triggering the apparent reaction. To the observer, it may appear that both trends are in sync. In fact, it may take the market months to respond to the increased demand for charging stations, or it may be preempting future demand based on the forecasted volume in EV sales.
Ultimately, the business decision to invest in charging stations, or anything else, is based on how business leaders interpret data. Data that has value can be used to make business decisions, and it is also now being used to train artificial intelligence models. With good training data and sufficient context, artificial intelligence (AI) models can help with generating forecasts leveraging historical data. These in turn can help business leaders make more informed decisions.
With good training data and sufficient context, AI models can help with generating forecasts leveraging historical data. These in turn can help business leaders make more informed decisions.
It is important to understand that historical data may or may not have biases. The stock market is a good example. When you look up stock prices on your favorite site, that data is biased towards the “winners.” Companies that have merged or otherwise been removed from the index are no longer in the data. Leveraging the context of the question at hand and working with subject matter experts help mitigate these issues.
Sometimes, data points can be entirely determined by external factors that are not immediately apparent to the observer. Correlations between data points for different trends may indicate causality. Sometimes that causality can lead us to discover trends that can influence future investment decisions. This is the real work of the analytics teams, helping the business move forward.
There are factors behind why OEMs are investing in new applications. Some of those factors will have more weight than others. Experience provides us with an insight into the importance of those factors. Information technology is a great toolset, but no substitute for experience.
Today, traditional AI has a hard time identifying and then weighing all the factors involved in making business decisions. Access to the right data is important. Identifying trends is making the right connections to show causality. Beyond the headlines of language models, causal AI is one of the most important areas of AI that can bring business value today. It’s an area we are leveraging more and more at Avnet.
At Avnet, we are constantly generating data. Avnet has been trading since 1921 and we have gathered a lot of market data in those 103 years of business. We use this data to monitor how well we’re performing, but it also helps us make business decisions that meet the market’s needs.
By combining our own data with data from the vertical markets we service, along with global trends, we can better anticipate–and forecast–what’s next.
Avnet trend data
We have used Avnet data sources and Statista figures to help create the articles in this issue of What’s Next.
As an example of how data sources can corroborate indicative trends, here we show figures for global revenue figures for the satellite industry, in hundreds of millions, against the total number of Avnet design wins in the satellite industry (please note, no vertical axis has been applied to the design wins).
The charts indicate there is a two-year lag between design wins and revenue. This mean 24-month design cycle is typical across all verticals in the electronics industry.
Satellite manufacturing sector revenue worldwide from 2001 to 2022 (in billion U.S. dollars). Source: Statista - Satellite Industry Association, June 20, 2023
Industrial automation is an enabling application area. As the Industrial IoT and edge AI continue to redefine manufacturing, industrial automation is seeing constant change. As a result, it also shows consistent growth. The graph below compares the historic and forecast worldwide growth in the value of the industrial automation market (shown in blue, in $billions) versus Avnet’s historic design wins from the years 2018 to 2022 inclusive (in green, no scale applied).
Again, we can see the correlation between the two datasets. Avnet uses this correlation to inform our business decisions that help strengthen our value proposition to customers. If the only constant in life is change, then trend data analysis brings meaning to what can otherwise seem chaotic.
Size of the global industrial automation market from 2020 to 2025 (in billion U.S. dollars). Source: Statista, November 24, 2021
ABOUT THE AUTHOR
Rondekka Moore
Vice President, business analytics and operations, Avnet
Rondekka Moore is vice president, business analytics and operations. Ron is a 30-year Avnet veteran. He holds a degree in Electrical Engineering Computer Science from the University of California, Berkeley, as well as certificates in Data Science and Predictive Analytics from the University of California, Irvine.