What does the drive-thru of the future look like? One thing for sure is that Artificial Intelligence (AI) will play a very significant role. This article explores this topic and brainstorms how AI might be applied to each step in the typical quick-service drive-thru process.
As with every other technology, business objectives and strategy should be considered first. Once the business objectives are clear and a business strategy is in place to accomplish those objectives, only then should an AI strategy be defined to support and enable meeting those objectives. How might that look for a QSR drive-thru? Let’s say the business objectives are as follows:
- Reduce average drive-thru time by 25%
- Increase guest satisfaction in drive-thru above 95%
Once there is a clear understanding of the business objectives, you should then set out to define your current drive-thru process in great detail. From there, you can evaluate how AI technologies might be applied to each step of the process in support of accomplishing the business objectives.
Some aspects of the AI Strategy will include foundational activities like building the big data infrastructure needed to properly support AI capabilities (AI algorithms are only as good as the test data they are fed). Other aspects of the AI Strategy will have a direct correlation to the business objectives (ex: optimizing menus based on the guest’s past history will speed up their decision process and in turn help reduce the average drive-thru times).
AI Technologies for the Drive-thru
Let’s start off with a high-level example of a typical QSR drive-thru:
Using the typical QSR drive-thru in the diagram above, let’s now take a look at each step of the process and brainstorm some ways that AI technologies might be applied:
Off-premise Ordering
Some QSRs will allow a guest to place their order while off-premise (via mobile, online, or automobile) and designate the order for pick-up at a specific location’s drive-thru. Here are just a few ways AI technology can be applied to the off-premise ordering process:
- Menu: Order history, item filtering, promotions, and availability as described in the Menus section below could also be applied to the off-premise ordering process using AI technologies
- Ordering: Item modifiers, up-selling, and cross-selling as described in the Ordering section below could also be applied to the off-premise ordering process using AI technologies
- Payment: AI technologies can be applied to off-premise payment to help authenticate the guest and identify fraud patterns
- Location Selection: AI algorithms can be applied to recommend a location for the order based on distance, drive-time, location drive-thru hours of service, location bandwidth/wait times, and/or item availability
- Order Firing: AI algorithms can be used to define the optimal time to fire an order to the kitchen in order to minimize wait time once the guest arrives at the location
Guest ID
There are many ways to leverage AI technologies to identify guests as they approach a drive-thru location. Here are some that can be applied during the approach:
- Mobile ID: mobile devices(s) associated with an off-premise order can be identified as they approach a location using geo-fencing and other proximity technologies
- Vehicle ID: license plate scanning of vehicles (using AI to translate the license plate image into text) as they enter the parking lot and drive-thru line can be associated with a registered guest or used as an un-registered guest ID
After the vehicle enters the drive-thru, additional guest identification can be performed using the following AI technologies:
- Facial ID: facial recognition technology is improving in accuracy and can cameras can be positioned on both sides of the drive-thru and at the face-level to assist in the accuracy of identifying all inhabitants of a vehicle
- Voice ID: voice recognition technology can be used to identify guests in a vehicle
- Predictive ID: AI algorithms can use pattern recognition to predict the most probable inhabitants of a vehicle once the vehicle ID is successful
Any and all of the above technologies can be used to identify the vehicle driver and possibly the other vehicle inhabitants.
Menus
Now that we have performed guest identification, menus can be customized on a per-vehicle basis and for any specific-guests identified or associated with that vehicle:
- Order History: past order history of the guest(s) can be leveraged by AI algorithms to identify and propose the most probable order and items
- Item Filtering: any known preferences and allergies can be used by AI algorithms to filter out items
- Promotions: AI algorithms can be used to determine which promotions are most probable to be successful with the specific guest(s)
- Availability: AI algorithms used to create the custom menu can be setup to consider the location’s inventory levels and exclude items that are out of stock
These are some of the ways AI technologies can be applied to customize the menu based on the specific guest(s) during the drive-thru process. There are also many ways to leverage AI technologies above-store to help with new product development, defining which items should be available per location, season, and day-part, and to optimize supply chains and inventory levels at each location.
Ordering
Many drive-thrus have a menu and a separate order confirmation board. Once guest(s) are identified and have considered the custom menus, they now begin placing their order. Here are some options to leverage AI technologies in the ordering process:
- Ordering Bots: AI bots can be used for some or all of the order-taking process. Over time this technology will improve to the point of being difficult to tell if it’s a human or bot speaking
- Item Modifiers: AI algorithms can be leveraged to propose modifiers most likely to be selected by guest(s) using order history and pattern recognition
- Up-selling: AI algorithms can be used to identify the order upgrades or add-ons that are most probable to be selected by the guest(s)
- Cross-selling: AI algorithms can be used to identify the potential related items that are most probable to be selected by the guest(s)
Just as described above in the Menus section, AI technologies can be leveraged above-store to help define optimal item modifiers and add-on options that should be available in specific day-parts, locations, and regions.
Payment
The primary use of AI technology in the drive-thru payment process will be in the area of guest identification and authentication. Payment technology companies will continue to innovate in this area to leverage facial and voice recognition as additional methods to authenticate that the guest is authorized to use the payment method being presented for payment.
Fulfilment
AI can be leveraged both proactively and reactively in the QSR kitchens to optimize efficiency and accuracy. Some proactive options include:
- Predictive Preparation: AI technologies can be leveraged to predict order item volumes at specific times so that order prep can be started in advance, optimizing drive-thru times
Some reactive options (reacting to a specific order coming from the drive-thru) include:
- Quality Assurance: AI technologies can be used by having cameras monitor that an order is being prepared properly and that mistakes aren’t made (ex: flagging when a modifier like “add cheese” isn’t performed)
- Order Prep: AI technologies could be leveraged to determine statistically which condiments, napkins, straws, etc and how many to include to ensure guest satisfaction while also minimizing costs
Post-Order
Once the guest exits the drive-thru, QSRs can use AI technologies to optimize guest follow-up activities. Here are a few examples:
- Guest Satisfaction: AI technologies could be leveraged to determine guest satisfaction based on facial and voice recognition of emotions. Chat bots could also be leveraged in post-order guest satisfaction surveys
- Loyalty: AI technologies could be leveraged to provide post-order rewards and incentives most likely to cause that specific guest to increase their visit frequency
- Promotions: AI technologies could be leveraged to define the optimal promotions on a guest-by-guest basis or for a specific location or region
As you can see from the list above, there are a significant number of areas in the drive-thru process that can be optimized or enhanced with AI technologies.
In the future, competitive advantages in QSR will be significantly determined by the level of creativity applied in the application of AI technologies.
Allen Eskelin, CEO of Peak Portfolio
I would like to continue to grow this list of potential applications of AI technology in the drive-thru so let me know if the comments or via email if you think of any that should be included. I would also appreciate it if you let me know of any QSRs who have implemented AI in their drive-thrus.
