How AI Will Shape the Restaurant Industry
If you’ve been paying attention at all during the last few months, you’ve undoubtedly heard the words “Artificial Intelligence” (AI) and “machine learning” (ML) being bandied around. What are these technologies, and how will they impact the restaurant industry? That’s what we’ll explore in this article.
AI and ML Defined
First of all, what’s Artificial Intelligence? The Encyclopedia Britannica defines it as “the ability of a digital computer or computer-controlled robot to perform tasks commonly associated with intelligent beings, [...] such as the ability to reason, discover meaning, generalize, or learn from past experience. ”
In short, AI is simply training a computer system to do something ‘human-like’, such as classify an image, find a pattern among many data points, look up information, etc. Recently, generative AI (AI that “makes” content, usually by being trained on lots of existing similar content) has gotten a lot of buzz – and not all of it good. However, AI has been used for decades as an analytical tool and has a long history of helping businesses process data effectively and efficiently. Most of us use AI every day through things like Siri, Google Assistant, and Amazon recommendations (among many others).
AI is a broad field with many subsets. One of these subsets is machine learning, which is simply AI that’s designed to analyze and improve its own performance. This analysis can be done through human inputs (i.e. through feedback on the usefulness of search results) or by the ML system itself. If you read about ‘self-learning AI’, that’s another term for ML.
Other AI subsets include natural language understanding and generation, computer vision (image recognition), deep learning, neural networks, and robotics. In this article, we won’t differentiate between the various types of AI; we’ll just focus on the field as a whole and its impact on the restaurant industry.
10 Ways AI Will Shape the Restaurant Industry
Below, we’ll discuss how AI will (and is already) changing aspects of the QSR environment. Some of these are quite familiar and already becoming widely adopted; others are on the cusp of wider acceptance. Let’s start with one that’s already well-known: personalization.
Personalized Recommendations
As we mentioned above, AI is great at processing enormous amounts of data and finding patterns that humans may miss because of the sheer size of the dataset. The traditional example of this is making personalized recommendations to customers. AI can analyze customer data and make individual and unique recommendations based on that customer’s order history, location, and demographic info. It can go deeper and create recommendations based on customers with similar order histories (or other shared characteristics).
Personalization isn’t just limited to recommendations; it can also be used to tailor the user experience, especially digitally. Check out these articles on how Netflix uses personalization in movie recommendations and how Spotify is using AI to create personalized DJs for more background. Across industries, this kind of personalization has been shown to boost customer engagement, improve user experience, and lead to higher revenue.
Tillster AI & machine learning recommendation engine
Customer Engagement
Personalization is great for making customers feel valued, but it’s not the only way AI can improve the customer experience.
If you’ve reached out to online customer support, chances are good that you started by interacting with a chatbot. AI (specifically, machine learning) has made these interactions more comfortable and user friendly. And since chatbots can handle many more customer interactions at once, they’re both cost-effective and excellent at minimizing frustrating wait times. A Gartner report revealed that “when designed correctly, chatbots can improve customer experience and drive positive customer emotion at a lower cost than live interactions.”
As most restaurants struggle with an ongoing labor shortage, chatbots will alleviate some of the strain on both sides of the equation. By allowing customers to quickly find answers to their questions and concerns, they foster better customer engagement. And by removing some of the load from employees, they allow staff to focus on other essential tasks.
Speech Recognition
When we ask “Hey Siri, what’s the weather right now?”, we’re using another common application of AI: natural language understanding, or speech recognition. (While “speech recognition” and “voice recognition” are often used interchangeably, speech recognition refers to recognizing words spoken by different voices, while voice recognition refers to recognizing individual voices, i.e. as a security measure.)
Natural language understanding is a subset of natural language processing, which is a subset of AI. Essentially, it’s all about training computers to understand human language -- in this case, spoken language. This has been in the business world for several years; think of automated phone systems, voice-activated shopping or searching, and speech-to-text services on your phone or computer.
In the restaurant world, the most exciting application of speech recognition is certainly voice ordering. This will add a layer of convenience to the digital ordering experience; it will also make strides in keeping systems accessible for customers with motor difficulties. In turn, a well-functioning speech recognition system will add to customer satisfaction and engagement.
Inventory Management
Now let’s switch focus from customer-centric uses of AI to operational AI, starting with inventory management.
Inventory management (and by extension, supply chain management) is a well-established domain within business AI. There are a multitude of examples:
Amazon uses computer vision to track the movement of goods inside warehouses, which is more efficient than hand-held scanners.
Companies around the world – including restaurants – rely on AI-assisted inventory management tools to track incoming and outgoing inventory. Some can automatically order items, leading to less waste and out-of-stock situations.
During and after the pandemic, retailers used AI to cope with demand forecasting, inventory planning, and interrupted supply chains.
The potential savings here are obvious: less food waste due to overstock and spoilage, easier inventory management (because you can hand some tedious tasks to AI), and better customer satisfaction because you’re not running out of their favorite items.
Predictive Maintenance
Predictive maintenance – fixing something before it actually breaks – is familiar to anyone who owns a home, car, etc. In the past, this has usually taken the form of regularly scheduled servicing. AI takes this one step further by predicting potential equipment failures and alerting maintenance personnel to them before they happen. This avoids unexpected downtime due to repairs and can also make the restaurant a safer and more efficient place to work.
How does AI do predictive maintenance? By collecting data points, just like everything else. The poster child for predictive maintenance is the mining industry, where equipment failure can cost lives. In addition to data coming from sensors placed on or near mining equipment, an AI system also analyzes workloads, weather conditions, routine upkeep cycles, and other factors. It can detect when a failure is likely to occur and send teams a warning to fix the malfunction.
Fraud Detection
Fraud can happen in restaurants in a variety of ways: skimming cash from sales, re-charging customers for the same items, fictitious ingredient orders, exaggerated supply costs, fraudulent customer transactions, and more. Every year, restaurants lose an estimated 4-5% of sales to fraud. How can AI help?
For inspiration, let’s turn to the banking industry. AI has been used for years to detect patterns of fraudulent activity. They do this in several ways: by using AI to analyze billions of transactions and flag suspicious behavior; by verifying customer identities in compliance with federal Know Your Customer regulations; and by building profiles of customer behavior and using that to check for activity that departs from those patterns. (If you’ve ever gotten a notification because you used your credit card on an overseas vacation, that would be fraud detection at work; the transaction was flagged because your card is not in its usual geographic location!)
Admittedly, the financial sector has to meet stricter fraud standards than the restaurant industry. But fraud still takes a chunk out of restaurateurs' hard-earned income; anything that can prevent those losses is a gain.
Route Optimization
You’re on your way to a meeting in an unfamiliar city. You enter the address into your map app of choice and start driving. Halfway through, you get an alert that the route has changed to avoid heavy congestion. You can thank AI for that time-saving route update.
It’s not too hard to see how this kind of help – known as route optimization – can be extremely helpful to food delivery drivers. Not only can AI-powered apps help drivers avoid unexpected slowdowns, they can help them plan their route to save the most time and fuel. This can reduce delivery times, resulting in hotter, fresher food (and happier customers).
Quality Control
One of the shortest paths to customer dissatisfaction is to present them with a dish that looks nothing like its luscious picture on your menu – or worse, one that’s made with substandard ingredients. Given what it can already do in retail and other industries, using AI to ensure only high-quality food leaves the kitchen isn’t far away.
Remember the computer vision we mentioned earlier? It essentially uses AI to recognize images (photos, video, or real life) and, in this case, compare them against a benchmark. This can be used along with machine learning to verify planogram compliance, or the layout of items in a store display. This could also be used to optimize in-store displays for QSRs and others. Some other food-related AI uses include:
Analyzing customer taste trends and coming up with new dishes.
Monitoring food production for quality and predicting food decay.
Robotic food production and delivery.
Labor Scheduling
The food service industry has been facing a labor shortage. This makes scheduling employees – particularly during peak times – extremely important. AI can help by analyzing restaurants’ activity as well as other factors that can influence local demand (weather, nearby events, holidays, etc.). This not only saves human schedule-makers from having to weave all those factors into their decisions, it also helps them deal with unexpected challenges (like the COVID-19 pandemic).
By ensuring that enough workers are available during peak periods, AI can help keep the customer experience positive. It can also reduce long wait times. On the flip side, it can ensure that fewer workers are scheduled during slow times, which improves efficiency and saves money.
Price Optimization
Finally, we’ll briefly mention one more area AI can help restaurants increase their profitability – via optimized pricing.
This is a fairly standard practice in many industries. Think of dynamic pricing, which online retailers use to adjust product prices based on demand, customer location, amount in stock, etc. There’s also price forecasting (predicting how prices will change based on various factors) and price sensitivity analysis (which models how customers will react to different price points for the same product). In a restaurant situation, price optimization can include factors like location, customer preference, market trends, and competitor pricing.
Once again, AI can save humans a lot of time and frustration by doing this analysis and quickly presenting various options. And by making it easier to achieve the right price point, it helps brands avoid the trap of over- or under-charging for their product. This improves revenue and profitability.
An Exciting Future for AI and the Restaurant Industry
AI has already made its mark on the restaurant industry. And it’s poised to continue to transform how we perform essential parts of our business.
Although there’s a lot of speculation about AI taking over human jobs, this technology actually works best alongside humans. It can do the uninspiring routine tasks and detailed analyses that people often find hard or boring, freeing us up to concentrate on more productive tasks.
And we haven’t covered all the possible use cases for AI in food service; the future is definitely going to be interesting. Why not examine whether AI can solve some problems in your restaurant operations? As you start to adopt AI in some areas, you’ll see just how much it can benefit your entire company.
