Revel Blog

Coffee AI—Can Forecasting Brew a Better Cup?

Coral Drake | March 2, 2023 |

coffee shop Featured
Coffee AI—Can Forecasting Brew a Better Cup?

Coffee has been consumed since ancient times. It was popular throughout most of the world by the 17th century. There have been many evolutions in coffee's journey over time, but artificial intelligence (AI) may be among the most dramatic. 

If you’re wondering whether there’s truly space for AI forecasting in coffee, continue reading and decide for yourself! While relatively new, AI for coffee can be a useful tool to predict trends in coffee operations, with the potential to make a major impact in the industry. 

In some cases, AI is fundamentally changing how coffee is grown, produced, and consumed. Businesses utilizing AI coffee forecasting are able to predict trends in the industry and stay ahead of the competition. 

How is AI Forecasting Making Better Coffee?

Artificial intelligence and machine learning are revolutionizing how businesses function in just about every industry, and coffee is no exception. Coffee AI can make predictions about the industry that help coffee growers and sellers make better choices and investments. Here's what AI forecasting in the coffee industry is accomplishing, from planting and harvesting to brewing coffee. 

Planting—Farmers Predict the Best Places to Grow Coffee with the Help of AI-Powered Technology 

A device the size of a business card known as the IBM agri-pad is helping farmers predict where they should grow coffee beans to achieve the highest yields and best flavor. It only takes a drop of water or a little bit of soil for the device to report PH, nitrite, magnesium, chloride, and more; everything the farmer needs to know to decide whether to plant at that location. 

Here’s how it works: The farmer photographs the card with a designated smart device. AI analyzes the chemical makeup, uploading the results according to location. Not only does this kind of technology promise to enable individual farmers to forecast the best locations to grow coffee, but it also enables all of the farmers using the program to predict trends in coffee-growing areas.

With data like this, farmers can make accurate predictions about the best places to test soil and expand coffee-growing areas in the future. It also reveals changes to the soil quality over time. 

Coffee growers can anticipate regional changes depleting the soil’s natural nutrients or throwing off the soil’s PH balance. Conversely, they can quickly spot where efforts they’re making are likely to continue to be effective. 

Growing—Coffee AI Can Help Growers Anticipate (And Stop) Spreading Diseases

Computer science researchers at the Federal University of Uberlandia, Brazil are planning to train AI to recognize a fungus that attacks coffee leaves, resulting in the dreaded Rust Infection. Coffee production can drop by as much as 45% as a result of this disease, and it is very difficult to treat.

The proposed robot will be able to use advanced visual receptive fields and machine learning to recognize contaminated leaves and isolate plants in need of treatment. 

This kind of AI and the data it provides is valuable for reducing the negative impact on coffee plants and for predicting trends in coffee plants likely to become infected. It reveals valuable information to farmers about where to change planting patterns or create gaps between plants to keep disease from spreading. 

Tasting—Coffee AI Is Describing The Product Like A True Connoisseur, Showing Us Which Beans Are Winners 

A groundbreaking company known as Demetria is making vast strides with AI forecasting in the coffee industry. They are utilizing handheld near-infra-red (NIR) sensors to gather vast amounts of data. 

A “spectral fingerprint” is created as wavelengths of light from the sensors react differently to the organic compounds in the coffee beans. This creates a “description” of the beans’ chemical composition. 

Humans couldn't possibly interpret all of this data unassisted. Instead, Demetria utilizes a coffee AI program to interpret the data and describe the coffee in detail. 

In the past, certified tasting experts physically tasted coffee to determine the quality and flavor of the beans. Not surprisingly, such a process was fraught with the subjectivity of the tasters, despite their advanced training and the industry-standard coffee tasting wheel which they used to describe the taste they experienced. 

Using AI enables bean flavor profiles to be described more consistently than was previously possible. Coffee AI is constantly being trained to understand the nuances of flavors that human tasters experience. 

There are many benefits of utilizing AI forecasting in coffee, such as:

  • Farmers receive accurate descriptions of their beans which they can use to inform growing decisions for the next season.
  • Coffee sellers can use consistent flavor profile information for demand forecasting to better predict what their customers are likely to want and stock it for them.
  • Flavor profiles enable coffee shops or sellers to choose specific coffee flavors, predict trends in coffee that will succeed with particular customer bases, test the beans, and stock consistent flavor profiles across different bean varieties. 
  • Beans can be tested at any time, whether they are green or fully roasted coffee. Predictions made by coffee AI that a green bean isn’t going to make a good drink when it’s roasted saves excess roasting time and bean testing. 

Roasting—Automated Roasting Machines Are Replacing Human Roasters

AI roasting technology can cut the labor required to roast beans, a valuable advantage in the coffee industry, which is experiencing the same kind of labor shortages as most food and service industries. 

In fact, coffee AI roasters can adjust temperature and airflow as needed so as to roast the bean perfectly. In the past, roasters had to rely on intuition and experience to make these kinds of adjustments. Now, AI can do it.

Artificial intelligence in roasting goes further than the actual process of roasting the beans. Roasting machines utilizing artificial intelligence record important data points as the beans are roasting, including: temperature in the air, exhaust, and “automatic first crack detection,” a flagship feature. 

The machine uses this data to roast better coffee beans with less trial and error on the part of the roaster. This kind of AI forecasting in coffee both improves the product and decreases the human labor required. 

This data is also invaluable for informing future decisions about beans and roasting. The combined understanding of a bean's flavor profile with the help of near-infrared sensors and knowing exactly what temperature was required to roast it to the ideal flavor makes it more likely that a bean with a similar profile will be roasted appropriately in the future. 

These data points enable coffee roasters to forecast a wealth of predictions about which beans need what kind of roasting for what kind of time to achieve their ideal flavor profile. This kind of AI forecasting in the coffee industry is actively working to bring us the best coffee beans, faster than was possible before. 

Brewing and Serving—Robot Baristas?

Robots are replacing baristas at a coffee startup known as Artly. Artly uses artificial intelligence to control specialized robots and make very good coffee. 

Not surprisingly, business has been doing well since the pandemic made hiring and staffing real life baristas a challenge. Artly set out to design a robot that could make coffee like a human, without sacrificing quality. In fact, they wanted to make a robot that could make better coffee than a person could.  

Many would argue they’re getting close, or have even achieved a more consistently good cup of coffee than what you could expect at a traditional cafe. Artly robots can detect anomalies, plan paths, avoid obstacles, and utilize physical reasoning skills as they go about making an excellent cup of coffee. 

Deep learning AI and computer vision work together to continuously learn to make better coffee. It was primarily trained by reigning US barista champ, Joe Yang, but it keeps learning more from the baristas who train it. 

As we move into an increasingly contactless world that is no less obsessed with coffee, it's reasonable to expect we'll see AI baristas become ever more advanced. 

How is AI forecasting making better coffee right now? AI bots are learning from human baristas. However, it is reasonable to think that this kind of technology could grow to anticipate the techniques most likely to be well-received by customers, enabling it to forecast and test new techniques on its own. 

We aren't there yet, but it's reasonable to think that AI forecasting in the coffee industry will eventually be able to make important predictions about what customers are likely to want from their coffee, then brew them the perfect cup. 

Don't Miss out on the Benefits of AI Forecasting for Coffee Businesses

Whatever step of the coffee industry you're part of, whether you're interested in fair payment and ethical working conditions for coffee growers, worried about creating the absolute best coffee and sourcing the beans most likely to sell, or trying to staff a cafe in a labor deprived economy, AI forecasting has valuable contributions to make to the coffee industry. 

In fact, failing to take advantage of AI forecasting in the coffee industry could be a real missed opportunity that puts you behind the competition. In a highly concentrated industry like coffee, having the edge is the difference between staying open and not. Increasingly, it's becoming clear that AI coffee forecasting is a valuable edge to have.