Long queues can decrease NAR by $4.09 per passenger
Airports are successfully recovering their pre-COVID travel levels. Many passengers are excited about their summer vacations and upcoming holiday plans but is your airport ready to handle massive passenger influxes and maintain optimal wait times and service levels?
.webp)
Massive, chaotic queues are the last thing both passengers and airports want to see. These queues results in delays, missed flights, and regretting the travel plans they’ve made. No wonder IATA’s survey revealed that passengers prioritize speed above all else at airports. But how does wait time actually impact airports’ bottom line?
55% of consumers were more likely to buy, the longer they remain in an airport concession (Mintel reports, 2022). The average value of this purchase is $7.44 which is the global non-aeronautical revenue per passenger. Airports can realize $4.09 NAR per PAX (55%*$7.44) when they spend more time in concessions through shorter wait time.
But how much time would PAX need to make this additional purchase? For business or solo travelers, additional 5 minutes would be enough. For families of 5 people, they will likely need more than 10 minutes. To simplify this variability and measure the per minute impact on NAR, a scenario analysis was conducted.
In an optimistic scenario, 5 minutes is enough for PAX to make additional $4.09 spend like coffee or a water bottle. Then 1 minute wait time decrease brings additional $0.82 per PAX ($4.09/5 minutes). In a realistic scenario, PAX would need additional 10 minutes to bump their grab-n-go snacks to a restaurant dine-in option.
With these assumptions in mind, we've constructed 4 scenarios as the following.
- Pessimistic: PAX need 20 minutes
- Conservative: PAX need 15 minutes
- Realistic: PAX need 10 minutes
- Optimistic: PAX only need 5 minutes
Even in a conservative scenario, airports can realize $0.28 additional NAR per passenger per 1 minute wait time decrease.
An airport can generate $560,000 per 1,000,000 PAX with a 2 minute decrease in wait time
Queue management, therefore, is not just about maintaining optimal wait time and airport operations but its the single biggest factor between PAX deciding to head to their flight directly or browse retail and buy that extra cup of coffee or a bag of snack.
Post-COVID trend pressures further with return-to-normal traffic, and airport labor challenges to handle all the traffic for travel processing. Airports should seriously consider investing in their queue management to protect their non-aeronautical revenue bottom-line.
Queue management isn’t just posting wait times
Effective queue management is all about proactive data sharing between stakeholders. Why?
Posting wait time is a simple, easy fallback method that many airports use to justify their efforts to queue management. But how many passengers follow TSA guidelines to arrive 2-3 hours before the flight departure like a diligent student?
For busy immigration officers and airport staff trying to process and manage the queue, how much does having a wait time range really help them prevent long queues or know exactly what to do when the wait gets too long? Without knowing exact cause of long queues and idle resources they can utilize, there’s nothing they can actually do.
%20(1).webp)
1. Queue management is a collaboration
Check-in queues are affected by the number of airline staff processing check-ins, the number of check-in desks assigned to the airline by the airport, and the number of airport staff providing support.
It's a close collaboration between the airport, government partners, and airlines. Security queues at airports near the border require TSA, CBP, and airport staff to work together to achieve optimal wait times.
Queues can result from various factors, such as a lack of staff, improper instructions, or specific passenger characteristics or even last minute flight delays. Information should be shared with relevant stakeholders in real time to address the issue. In particular, when the queue requires additional staff, having full visibility into terminal operations is essential to avoid disrupting operations from staff reallocation.
2. Airport queues are complicated and fluid
It’s been 50 years since TSA set the security wait time goal at 10 minutes per passenger. Yet only 11 out of 130 airports deliver on that promise. Why?
Airports’ complicated queue environments has dozens of factors that contribute to the queue at the same time as well as a changing queue environment. This makes it impossible for one entity to manually gather, synthesize and understand what is going on. Some airports have 17+ queues forming at the same time within the immigration terminal. There are queues constantly merging and separating real-time, not something a human can easily understand and manage.
50,000 passengers passing through 15 different zones with 17 different queues at the same time easily requires a supercomputer or an AI to effectively analyze the queue environment.
Having a fixed solution fails to support these airport queue dynamics resulting in data not accurate enough to use for queue management strategies and decisions.
.png)
3. The bigger the airport, the harder queue management is
Queue management becomes even more challenging in larger airports. Unlike a manufacturing assembly line with identical products and a defined flow, passenger flow is much more random and can easily get out of control. Passengers exhibit different behaviors, characteristics, and personalities, which can prolong the queues, particularly for immobile passengers and families (Janssen, 2020). Some passengers do not follow the compliance guidelines, further delaying the process with manual baggage checks.
The more passengers there are, the more frequent these delays occur. Juggling all these variables in a spreadsheet is how you end up with a sub-optimal plan.
Just using flight schedule data to plan staffing to meet the optimal wait time is not enough because of these unpredictable variations.
Additionally, having more airlines complicates the queue even further. Some airlines have a floating check-in desk assignments which makes it hard for airports to define queue area to capture data per airline. Airports can avoid this level of complexities by choosing an AI that supports this real-world complexity and fluidity. Access our full AI buying guide here.
What to look for in a queue management solution
Although queue management is complicated and seems impossible, there are still solutions that will successfully keep PAX wait optimal and actually improve NAR monetization potential. This section covers what to look for to avoid failed queue management project.
1. Quickly suggest staff reallocation
The biggest hurdle in queue management is how to reallocate staff without hurting the operation of other terminals, especially with airport labor challenges. Innovative airports use AI to setup virtual manager alerts. An airport had two zones each with 3 staff processing the immigration queue. Zone A had 3 minutes of observed wait time, while Zone B had 20 minutes of observed wait time.
With mobile dashboard data and AI’s virtual manager alert, the airport can easily send one staff from Zone A to Zone B to manage long queues without hurting the operation of Zone A.
2. Provide daily queue level of service report
Responding to the queues real-time is important, but it’s no use if there’s not enough staff in the first place. Daily rundown of what happened and how the wait time deviated can answer questions like:
- Do we have enough capacity structurally for the volumes of passengers that well in the area?
- What’s our pain points with regards to the queueing process and times
- What opportunities are there to improve efficiencies and passenger experience along their journey
With these daily level of service (LoS) reports, terminal operations team can grasp the real-world complexities like passenger behavior and characteristics. Using this data on top of the flight schedule data leads to more accurate staffing prediction and planning:
- Schedule staff around peak previous day passenger processing times not just fixed flight schedules
- Consider the trend of passenger characteristics per airlines in staffing with information like tend to arrive 20 minutes before the flight or arrive 3 hours before the flight to enjoy drinks
- Identify sustaining queue trends that require infrastructure investments that staffing strategies can’t solve
Continuously capturing these data will show you the trend of the queues within and across zones, terminals. You can discover abnormalities and how to reallocate resources looking at this trend data.
3. Accurately predict wait time
Using AI-based queue management solution allows airports to predict future wait time. Instead of being reactive, they can now be proactive.
Here’s how Zensors AI predicts future wait time for our airport clients using a neural net:
- Train AI on previous queue data
- Input the number of people in the queue, the entrance flows, the exit flows, and the observed wait time for the queue
- Predict what the wait time will do for the next n minutes (n can be 1 minute to far out in the future)
For more in-depth computation of observed wait time, look forward to our next post on how to build an airport queue management system with existing cameras.
Zensors AI ensures accuracy of the predicted wait time through providing rich variety of high-quality queue data to train the AI model. Zensors AI is currently processing 3,000 hours of videos every day that adds up to the accuracy. Zensors team also provides Human-AI collaboration training to interpret results and provide constructive feedback.
Make sure your choice of AI solution incorporates these features listed out in our comprehensive AI buying guide for airports.
Need more AI inspirations?
Looking for more resources to help you kickstart your AI queue management journey? Here are some additional posts for more inspiration:
- Complete AI buying guide: How to choose the right AI for your airport
- How to build an airport queue management system with existing cameras
- A full AI solution stack for aviation
.webp)