Say Hello To Hey NAPA!

New AI tool from XBE aims to transform the way asphalt professionals learn

BY TY JOHNSON, MANAGING EDITOR

From corporate boardrooms to classrooms across the country, artificial intelligence (AI) has emerged as this year’s most buzzworthy topic, with signs that it will continue to innovate industries as its applications grow.

Learn more about AI and its innovative qualities, including its applications within NAPA’s own trained chatbot, Hey NAPA, in the following pages.

You can even experience the research power of AI yourself by visiting HeyNAPA.com. Simply sign in with your name and email address to gain access to decades of NAPA research publications via the tool, which accesses NAPA’s extensive knowledge to answer questions on all things asphalt.

Back when NAPA announced its 2023 Midyear Meeting would be held in Kansas City, large language model (LLM) artificial intelligence was still largely theoretical. Elements of AI had been playing roles in our web search and social media algorithms for years, but it wasn’t until the release of ChatGPT-4 on Nov. 30, 2022, that the general public began putting it to work.

XBE CEO Sean Devine, who calls himself an early adopter, quickly began putting the power of AI to work on his professional and personal goals, from editing content and collaborating on programming code to financial analysis and meeting transcriptions.

“I use AI tools for just about every purpose you can imagine,” he said during the opening general session at Midyear, where he was the featured speaker. “My personal estimate is that my productivity is up about 50 percent in the past nine months.

“And so I thought, if that’s true of me, I wonder who else it’s true of. What would happen if everyone in our operation increased their productivity by something like 50 percent?”

Pursuing an answer led to the creation of an XBE chat assistant called Hey Kayla, which allowed all employees to access a logistics and technology hivemind that reflected the goals and values of clients and the company itself, which leverages data for optimal horizontal construction.

Devine said providing the tool to his employees resulted in a 50 percent improvement in programming productivity, but he was most impressed by the increase in capability.

“We’re able to do harder things than we used to do,” he said. “Our customer communication has gotten better. Everyone has become a better writer.”

Fresh off the successful rollout of XBE’s chatbot, Devine said he soon found himself on a mission to share the news about an innovation he said may very well have a larger impact on our way of life than even the internet or smartphones.

HEY NAPA!

Devine delivered on that mission when he unveiled Hey NAPA, calling it XBE’s “gift to the industry.”

Building on the latest LLM technology, XBE worked with NAPA to develop the tool, which can deliver thoughtful answers to even the most technical of asphalt-related questions, all based on NAPA’s wealth of data found among more than 4,200 pages of publications (and growing). Hey NAPA is trained to stay on message as it relates to the interests of the asphalt community and provides links to primary source documents as it generates custom, concise answers to questions, which can be easily shared with others.

Acknowledging that some are still skeptical about AI, Devine said that the real danger may lie in getting left behind by the next-generation business tools it will power.

“You should focus on the intelligence,” he told the crowd at Loews Kansas City. “The artificial point, that’s the distraction: The intelligence is the secret.”

He said there is a lack of understanding of how intelligence factors into business bottom lines.

“We don’t talk about what intelligence is good for,” he said. “Intelligence is the number one predictor of job performance.”

And while humans historically have been unable to change our intelligence, Devine said the AI technology that exists today can make an impact on workers equivalent to an increase of 15 IQ points. He said just as the introduction of computers playing chess changed the game, it will be those in industry who can work best with AI who will emerge victorious due to what chess grandmaster Garry Kasparov called “human strategic guidance combined with the tactical acuity of a computer.”

“It’s the people plus the machines that do great things,” said Devine. “The AI doesn’t care if it survives. Don’t worry about it taking over. Don’t worry about its goals. It doesn’t have goals. Worry about your goals.”

---

Hey NAPA Tips & Tricks from Sean Devine

XBE CEO Sean Devine shared the following tips for using Hey NAPA.

• When using Hey NAPA, forget everything you’ve learned from simply searching on Google!
• Craft full sentences, be precise about what you’re seeking, and provide as much context as possible for better responses.
• While “please” and “thank you” are optional, if that gets you in the collaborative spirit, I recommend it.

---

Hey NAPA Q&A

Hey NAPA is the asphalt pavement industry’s newest education tool. Based on thousands of NAPA datapoints, this AI chatbot can answer questions on everything from Environmental Product Declarations (EPDs) to parking lot design. Asphalt Pavement’s Ty Johnson asked the chatbot a series of questions to introduce it to readers and share ideas for crafting the best question prompts for users. The responses from Hey NAPA are provided below in their entirety without editing. And in case you’re wondering, you will get a somewhat different response if you ask Hey NAPA the same questions due to the nature of the underlying AI algorithm.

HI, HEY NAPA! DOES IT HELP YOU ANSWER MY QUESTIONS IF I TELL YOU THE REGION WHERE I WORK AND WHAT I DO FOR A LIVING?

Yes, providing your professional background can help me tailor the responses to your specific needs and context. Whether you’re a contractor, engineer, project manager, or other professional involved in asphalt paving, or even if you’re just curious about the topic, I’m here to assist. When users share their location and occupation, it can greatly assist in providing more tailored and relevant responses.

In the United States, for instance, different regions have varying climate conditions that can affect asphalt-paving practices. The standards and regulations related to asphalt manufacturing and construction can also vary from state to state. Understanding where a user is based can help provide region-specific advice.

Knowing a user’s occupation can also help in providing more relevant information. For example, a contractor might be interested in the practical aspects of asphalt application, while an engineer might require more technical details. A project manager might be interested in timing, cost, and coordination aspects, while a sales professional might want to understand the benefits and selling points of asphalt compared to other materials. In summary, the more context a user provides, the more specific and helpful the responses can be.

OFTENTIMES AI CHATBOTS ARE ASKED OFF-TOPIC QUESTIONS AS PART OF TESTING OR USER INPUT. WHAT HAPPENS IF I ASK YOU A QUESTION ABOUT SOMETHING NOT RELATED TO ASPHALT PAVEMENT?

As an AI developed by XBE in partnership with NAPA, my primary function is to provide information and answer questions related to asphalt pavements, asphalt paving, and the business of asphalt manufacturing and construction.

If a user asks a question that is not related to these topics, I will politely inform them that I am specifically designed to assist with asphalt-related inquiries. I will then encourage them to ask a question within my area of expertise. This ensures that the information I provide is always relevant, accurate, and helpful to the user’s needs in the realm of asphalt paving and pavements.

WHAT’S THE BEST THING ABOUT ASPHALT PAVEMENT?

There are many excellent qualities of asphalt pavement, but if I had to choose one, it would be its combination of durability and flexibility.

Asphalt pavement is highly durable, able to withstand heavy traffic and various weather conditions. It’s designed to flex and “give” with slight settlements or frost heave, which makes it less likely to crack over time compared to more rigid materials like concrete.

Furthermore, asphalt is 100 percent recyclable, making it a sustainable choice. When asphalt pavement is removed, it can be crushed and reused in new asphalt mixtures. This recycling process conserves natural resources and can save money.

Finally, asphalt pavement is smooth and quiet, providing a comfortable and safe ride for motorists. It’s also quick to install and easy to repair, resulting in less traffic disruption.

So, while it’s challenging to highlight just one aspect, the blend of durability, flexibility, sustainability, and user comfort makes asphalt pavement an excellent choice for a wide variety of paving needs.

“Providing your professional background can help me tailor the responses to your specific needs and context.”

—Hey NAPA

HOW CAN ARTIFICIAL INTELLIGENCE TOOLS LIKE HEY NAPA USE BALANCED MIX DESIGN KNOWLEDGE TO OPTIMIZE ASPHALT PAVING?

Artificial intelligence tools like “Hey NAPA” can utilize balanced mix design knowledge to optimize asphalt paving in several ways:

1. Optimization of Mix Design By analyzing the principles of balanced mix design, AI can provide guidance on the optimal proportions of aggregate, binder, and additives to achieve a mix that balances durability and resistance to distresses such as rutting and cracking.
2. Predictive Analysis AI can use historical data to predict the performance of different mix designs under various conditions. This predictive ability can guide the selection of the most effective mix design for a specific project, considering factors like traffic loads, climate, and pavement structure.
3. Quality Control AI can analyze real-time data from the production and paving processes to identify any deviations from the balanced mix design. This allows for immediate corrective actions, ensuring the quality and performance of the paved asphalt.
4. Resource Optimization By optimizing the mix design, AI can help reduce the use of costly materials, minimize waste, and improve the overall efficiency and profitability of asphalt paving operations.
5. Learning and Improving AI systems learn from data. As more data is collected from various projects, the AI becomes more accurate and efficient in optimizing asphalt mix designs, contributing to continuous improvement in asphalt paving practices.

In essence, AI can be a powerful tool for leveraging balanced mix design knowledge to optimize asphalt paving, improving quality, efficiency, and sustainability in the process.