Most of us use artificial intelligence every day, whether we recognize it or not. But it’s a big jump to adopt it for handling our precast machines and processes. Here’s a look at what it can do.
Artificial intelligence, commonly known as AI, is rapidly changing the world around us.
While you may be familiar with the AI that powers voice assistants such as Siri and Alexa, you may be unaware of the other areas of your life where AI already has been implemented.
Used a search engine lately?
These applications use AI to learn from search results and evolve over time.
Tried to get some help online?
Businesses are increasingly using chatbots and virtual assistants to answer your inquiries and provide customer service.
Got a warning from your email program that a message may be malicious?
AI can recognize and thwart potential cyberattacks.
Behavior analysis. Facial recognition. Shopping personalization. Disease diagnosis. Fraud detection.
The list of areas where AI has permeated our lives grows every day.
And it includes the manufacturing of precast concrete.
DYNAMIC SYSTEMS
Over the past several years, companies that specialize in precast software solutions have been integrating AI into their offerings. They’re exploring the benefits of how AI can help precasters now, with an eye on an ever-evolving future.
For example, Progress Group was founded in 1961 as a company focused on the production of building materials. During the years, it expanded into the production of precast elements and the manufacturing of precast machinery. Since 2015, Progress Software Development has developed digital solutions for the precast and rebar industries worldwide. Five years ago, it began integrating AI into its software solutions.
“We develop software for the precast concrete and rebar industry from the designing to the planning to the manufacturing process – we have software for the whole process,” said Alexandra Kammerer, marketing manager at Progress Software Development.
AI also has created opportunities for new companies to emerge.
AICrete, a 2020 startup based in Richmond, Calif., launched its cloud-based quality control software for precast in October 2023 and introduced it at The Precast Show, presented by NPCA, this past February in Denver.
“It’s a dynamic system,” Michael Fletcher, vice president of sales and marketing for AICrete, said. “Ultimately, the vision is to get to real-time optimization.”
Fletcher, who holds a degree in industrial and systems engineering, is AICrete’s “concrete guy.” Prior to joining the company, he was responsible for the admixtures and VERIFI Technology for GCP Applied Technologies, headquartered in Alpharetta, Ga.
DATA DRIVEN
Use of AI in precast manufacturing hinges on the collection of data.
“AI has this great ability to crunch data and make predictions,” said Hugh Martin, P.E., NPCA’s Director of Technical Resources. “It can even make up for there being a lack of data by filling in gaps, but it has to have data to begin with.”
Many precasters don’t have tons of data at their fingertips. But that will need to change moving forward. Whether precasters collect their own data or use a service that provides it for them, they will need accurate data to receive optimal results.
“In the mixing of the cement is where so much of a precaster’s data is being collected,” Martin said. “It seems like this is the area most ripe for this technology.”
Parham Aghdasi, AICrete’s CEO and founder, takes it a bit further.
“The kind of data that’s required for training meaningful AI models for our industry is nonexistent,” said Aghdasi, who has a Ph.D. in structural engineering with a focus in concrete and strong background in AI. “What data there is, is disjointed. It’s in different systems that aren’t tied together.”
Kammerer hopes that precasters will realize how important the topic of digitalization and the resulting benefits of data usage are.
“Only with the data can you reach optimal productivity and efficiency in your production, especially when you have automated machines,” Kammerer said.
AI software is often developed to interface with other programs you may already have so they can share data.
Companies are already developing systems to integrate or supplement available data. AICrete, for instance, has a database of physical and chemical properties of raw materials, weather data and performance metrics.
QUALITY CONTROL
Many different factors affect concrete’s properties, such as the materials used, the amount of water in the mix and environmental conditions – weather, temperature, humidity and wind.
“It’s unintuitive for us, as human beings, to process such a large amount of data in real time and to decide how to make adjustments to produce the best product,” Aghdasi said. “AI is very good at processing large amounts of data and optimizing or fine-tuning concrete in real time, to ensure it meets the performance requirements, is cost efficient and environmentally friendly.”
Fletcher said AI is especially useful for quality control staff who are usually “in the weeds.”
“The manual process of optimizing a concrete mix is very linear,” Fletcher said. “Literally, you change one thing at a time and see what happens.”
AI optimizes concrete using a multi-armed bandit approach.
“It can run tens of thousands of combinations in the blink of an eye and predict outcomes,” Fletcher said.
According to Martin, studies show the more water that’s used in the concrete mix, the lower the quality and vice versa. Project owners require stringent tracking of the amount of water in the concrete mix. They want to know how much water came in with the aggregate and how much water was added to the cement.
This information, available via the batch plant and batch ticket, can also easily be obtained using AI.
PROACTIVE MAINTENENCE
Precasters schedule routine machinery maintenance to try to keep their machines in tip-top running condition, but even then the machinery sometimes breaks down. And any precaster knows how costly it is to have a machine down.
AI can track trends and proactively predict when machine maintenance should occur based not on time or the number of jobs run but on the amount of stress the machine has been under. As Martin notes, self-consolidating concrete is very fluid and doesn’t require as many amps to mix, while dry-cast concrete, which requires more power from the mixer, will stress the machine more.
Using AI’s calculations for maintenance can be a lifesaver.
EFFICIENCY AND OPTIMIZATION
“AI can help you know what needs to change to reach a certain goal,” Martin said.
For instance, if the strength doesn’t meet the requirements, AI can tell you if you need to add more cement as you decrease the water or if you need to change your percentage of fine and coarse aggregate.
On the other hand, what if, after 28 days, your concrete strength is 30% to 40% higher than it needs to be? AI can help you optimize the mix plan so the concrete will meet the requirements but not overproduce, so you can save money.
The statistics look impressive.
Fletcher said their precast clients save, on average, $3.07 per cubic yard.
Progress Software Development customers say they experience time savings of 45% and 30% higher productivity rate by using their software
THE SUSTAINABILITY FACTOR
Sustainability is an umbrella that covers many facets relating to our environment. Project owners and government agencies are demanding that precasters meet more strict sustainability requirements.
“Having automated quality checks not only ensures the quality of the elements but also makes the process smoother and decreases the amount of waste during production,” Kammerer said.
Tracking and quantifying sustainable practices is becoming more prevalent. Some systems integrate directly with Environmental Product Declaration (EPD) software, allowing them to generate EPDs to quantify data such as carbon reduction.
“Our average carbon reduction is around 44 pounds per cubic yard,” Fletcher said.
Cement manufacturers are being mandated by the government to reduce the amount of CO2 that occurs when crushed limestone and calcium silicate are superheated. To comply, those manufacturers are changing cement ingredients and processes. Currently, they must reduce CO2 emissions by 10%, but Martin said that percentage will ultimately increase to 20%, 30% and 40%.
“That means our members are going to have to relearn how to make concrete every few years, because the cement is going to keep changing,” Martin said.
COMMON CONCERNS & MISCONCEPTIONS
Even with all the benefits AI offers precasters, it’s still too early for AI adoption to be mainstream. Fletcher said precasters often ask him who else is using it and whether it works with specific applications.
The most common objection Martin’s heard is this: If it ain’t broke, don’t fix it.
Another objection is the cost. Collecting the data may require more sophisticated equipment, and that’s a business expense in which some precasters aren’t ready to invest.
Aghdasi finds the most common misconception is that artificial intelligence will replace people in the workplace. Instead, he believes AI gives workers “superpowers.”
“AI will augment people’s capabilities,” Aghdasi said, “making them more efficient and increasing productivity.”
FORWARD THINKING
Advancements and improvements are being made to AI continually, and it is progressing at a rapid pace.
“We didn’t come to the party saying this is everything you need,” Fletcher said. “We ask our customers, ‘What do you need it to do?’”
Kammerer sees a bright future for AI in the world of precast.
“It’s a new topic at the moment, but I think AI will optimize many more processes, continue to increase efficiency, and improve the sustainability of our resources in the future,” Kammerer said.