The data gathered in our 2023 Frictionless AI Global Survey Report of more than 2,000 professionals around the world revealed a mountain of insight and shed a comprehensive light on how organizational data and artificial intelligence (AI) strategies are adopted and implemented around the world. That data also revealed the obstacles that make these strategies hard to implement, and what roadblocks stand between companies and project success.
In this article, we’re going to dive into the survey insights specifically from the aerospace industry. This data from a representative sample of aerospace organizations around the world will examine where and how the industry is implementing data and AI initiatives, and where these initiatives are coming up short technologically.
The Landscape of Organizational Data and AI Strategy Adoption in the Aerospace Industry
To begin, let’s analyze the landscape of organizational data and AI adoption within the industry to see how it’s adopting, using, and plans to use these technologies.
First, the aerospace sector considers itself behind when it comes to enterprise data and AI strategy implementation. It was the least likely industry to call itself a “leader” in its approach to using AI and data to propel digital transformation at 27%, compared to a survey average of 40%. It was also the most likely industry to say “we’re behind the curve and trying to catch up to our competitors” at 27%, which was 15 points higher than the survey average.
In addition, the aerospace industry was, by far, the least likely industry to say they’d start to implement AI for large-scale projects soon; only 25% said they expected to do so within the next 12 months or sooner, compared to the survey average of 59%. The industry was also the most likely to say they’d start to implement AI for large-scale projects “within the next two years or later” at 14%, compared to the survey average of 8%. That said, the industry was also by far the most likely to say they’d start implementing AI for large-scale projects “within the next 1 to 2 years” at 34%, compared to a survey average of 15%. These data points could indicate an abundance of hesitance and/or caution from the aerospace industry, and they may also reflect the realities of the industry’s long, stringent, and expensive safety standard compliance process. Many companies within the aerospace sector have substantial order backlogs, and as such, they must increase production to avoid penalties and deliver products on time.
The data also revealed useful information regarding how the aerospace industry currently uses enterprise data and AI initiatives. When asked what the top benefits of using data analytics are, two responses stood out far above the other industries: “more productive employees” and “we can better measure employee productivity.” This may reflect companies increasingly needing to do more with less, due to the difficulty of acquiring enough talent.
When asked which areas of the organization could better benefit from a data analytics strategy, the industry was the most likely to say “optimizing workflows” at 58%, compared to the survey average of 45%.
Technological Friction in the Aerospace Industry
Now let’s examine some of the technological problems that are hindering these strategies. Let’s recap a few definitions first. Any issues, including delays, mistakes, misalignment, and miscommunications that lead to project delay or failure is called friction. Technological friction is friction that stems from technology infrastructure — this includes hardware and software resources, cloud and high-performance computing (HPC) resources, appliances and plugins, vendors, and more. Often, technological friction acts as a bottleneck by limiting projects’ speed, scale, and/or scope. Friction is the top reason AI projects fail.
Since the aerospace industry is an industry that deals in literal rocket science, it’s obvious that technological friction is a massive hinderance for any organization.
To begin, let’s first see what the data revealed about data and AI project limitations, broadly speaking. It indicated that aerospace was the most likely industry to say it faces limitations that slow down AI initiatives often or very often at a combined total of 68%, compared to the survey average of 56%. It was by far the most likely industry to say it faced such limitations “often” at 34%, compared to the survey average of 19%. The industry’s three biggest obstacles that limit AI deployment ability were “lack of trust in models,” “siloed data,” and “lack of skills specific to AI within my organization.” Many organizations within the industry struggle with siloed data because of the need for confidentiality and highly specialized departments.
The industry was also second most likely industry to say “a third or more” of their AI projects failed within the past two years at 52% (survey average was 44%) and “a quarter or more” at 83% (survey average was 57%). On the data analytics side, aerospace had the highest percentage of responses that said more than half of their data science projects never made it to production within the past two years at 42%, compared to the survey average of 33%. It also had the highest percentages for “more than a third” (76% vs. survey average of 55%) and “more than a quarter” (87% vs. survey average of 67%).
When asked what challenges are negatively affecting AI initiatives at their organization, the aerospace industry was the most likely industry to say “our data experts struggle to explain insights to non-technical domain experts” at 32% (survey average was 26%), “leadership is too fixated on the upfront costs to understand how investing in AI or machine learning would benefit our organization” at 37% (survey average was 28%), and “technical experts have difficulty communicating their needs to other departments using layman’s terms” at 37% (survey average was 31%). Since engineers within the aerospace and defense sector aren’t usually trained in data science, communication can be difficult. Thankfully, Altair RapidMiner and its Center of Excellence upskilling methodology helps teams overcome these types of both technical and organizational obstacles.
When asked if their organization makes working with AI more complicated than needed, 61% of respondents agreed, two points lower than the survey average of 63%. The industry was the most likely to “somewhat agree” at 46%, compared to the survey average 37%.
Interestingly, aerospace was by far the least likely industry to trust the accuracy of their organizations’ predictive analytics at 61%, compared to the survey average of 82%. Within this total, the data also shows that the industry was by far the least likely to say they were “very trusting” of their organizations’ predictive analytics at 17%, compared to the survey average 36%. This could be a major contributing factor in aerospace organizations’ hesitancy to adopt data and AI strategies for large-scale projects.
In all, the data suggests the aerospace industry is more hesitant than other industries when it comes to adopting enterprise data and AI initiatives. While the industry isn’t ruling out the adoption of such strategies, the data suggests the industry plans to adopt these technologies within a more distant timeframe compared to other industries surveyed. This hesitancy and caution may stem from factors inherent within the industry – such as extremely high safety standards and strict regulation – and a lack of faith in their organization’s ability to gather and utilize high-quality data. It’s also important to keep in mind that the aerospace industry has a long, pioneering history in the use of advanced software tools and related technologies; it’s possible that any “shortcomings” they perceive compared to other industries may be overstated. In any case, whatever the myriad of causes, it’s clear that technological friction is a key reason the aerospace industry hasn’t been able to adopt these technologies quicker, and it will continue to affect the industry in the near future.
In the following aerospace-focused article pertaining to the data gathered in the 2023 Frictionless AI Global Survey Report, we’ll look more closely at the other two types of friction that commonly affect organizations and industries – organizational friction, and technological friction. This will give us a complete picture of the adoption and state of data and AI friction within the aerospace industry.
For more information about Altair’s data analytics and AI solutions and Frictionless AI initiatives, check out the following resources:
- Report: Altair 2023 Frictionless AI Global Survey Report
- Webpage: Frictionless AI
- Article: Organizational Friction – Examining Data and AI Friction Part One
- Article: Technological Friction - Examining Data and AI Friction Part Two
- Article: Financial Friction - Examining Data and AI Friction Part Three
- Infographic: AI's Breakdown Lanes - The Three Key Areas of Friction
- Infographic: Why Are AI and Data Projects Coming Up Short?
- Infographic: Achieving Frictionless AI - When, Where, and How
In March 2023, Altair commissioned Atomik Research to conduct an international, online survey of 2,037 professionals employed throughout several target industries in 10 countries who work with data in some capacity to drive valuable insights for their organization. Target industries included aerospace, automotive, banking, financial services, and insurance (BFSI), consumer electronics, heavy/industrial equipment, and technology. The data in this article is an examination only of the data gathered from the 71 respondents who identified themselves as working within the aerospace industry. The margin of error for the survey is +/- 3% with a 95% confidence interval.
*Due to the sample group’s small population in comparison to the overall survey, Altair does not consider this to be a comprehensive representation of the aerospace industry, but does believe it constitutes a “representative cross-section” that gives readers valuable insight into industry trends.
To learn about text analysis in the aerospace sector, click here to register for Altair's webinar, "Unlock the Power of Text Analysis in Aerospace Manufacturing."