← Back to Resources
case studyDecember 20th 2022 • 2 m read

22% reduction in aircraft loading time with Assaia alerts

Visibility and coordination of ground events offer aviation industry savings of $600 million per year

Assaia recently undertook a study at a large US airport to understand the impact of using computer vision, AI and alerts to enable ground handlers to better coordinate the loading process on an aircraft. The study covered four airlines at 31 aircraft stands over a period of three months.

In many cases, a number of ground-handlers such as those delivering on-board catering and luggage arrive at the aircraft at the same time, resulting in people in each other's way, uncoordinated activities and a prolonging of both processes. Think of it as the opposite of a Formula 1 pitstop.

We had anecdotal evidence and some observations that forward (fwd) catering and belt loading have this negative interaction when they happen at the same time. Therefore our hypothesis was that the entire cargo loading process (defined as the time between first belt loader connecting and last belt loader disconnecting) takes longer when fwd belt loading and fwd catering operations overlap.

The Setup

The ApronAI system uses existing cameras at the airport to create timestamps for all key turnaround activities. These timestamps are presented to users in real-time via the Turnaround Control application. However, for the purposes of this case study, we downloaded the turnaround data from the application and used Tableau to do the analysis. The analysis showed us that there is significant room for improvement in terms of which ground-handling activity should happen at which time.

The ApronAI creates a real-time understanding of the current state of each turnaround, and can be used to notify different ground handlers about the ideal time to approach the aircraft. These notifications can be sent from the ApronAI system directly or via integration with an existing resource management system.

During the reporting period, we observed over 23,000 turnarounds. For 88% of those turns, the fwd catering and fwd belt loading had some degree of overlap. This confirms our theory that these operations are typically not synchronized. Also, it shows that it is very possible to turn an aircraft around without this overlap (as was the case in 12% of the turns). We checked whether the turns without overlap shared some common characteristics (e.g. aircraft size, stand, airline, turn duration, etc) but could not find any specific characteristic which leads us to conclude that this happens randomly.

Next, we looked at the average total unloading/loading duration for both groups of turns (with/without overlapping fwd catering and fwd belt loading). As you can see from table 1, the loading process on average took 13 minutes (or 22%) shorter for turns without overlapping operations as opposed to turns with overlapping operations.

Even though we have not found any specific flight characteristic for which asynchronous operations would not be possible, it might be the case that the full benefit might not be achieved for all flights. Regardless, the results show that there is a very large potential gain.

The results shown in the previous section are pretty significant but the question remains what the value of knowing this is?

If loading crews can win 13 min on each turn where currently fwd loading and fwd catering are happening at the same time, then in total they can win over 266,000 minutes over the entire three-month period. Per day this would be 2,800 minutes. Since we already know that a loading crew is, on average, occupied with a turn without overlapping operations for 60 minutes, we can divide this gained time by 60 minutes to find that the same loading teams could actually handle up to 48 additional flights per day.

Another way to look at this is to consider a loading team that is working 8 hours per shift and handling 5 departures per shift. Saving 13 minutes on each of those turns yields them additional time which can be used for more flexible planning and would allow them to at least handle one additional departure during their shift. For all loading teams required to handle all traffic in this scenario, this would also accumulate to 49 additional departures per day.

With airlines’ and ground handlers’ focus on keeping cost low, these findings present a clear business case. The average monetary value of the time saved equals $15 per flight.

For an average airport that would accumulate up to $6 million dollars per year. For an average airline that saves up to $3.75 million per year. If these savings would be scaled across all airports and airlines, it would save the industry almost $600 million per year.

If you are eager to learn more about this particular case study or about different ways in which Assaia’s ApronAI software can create value for your airport or airline contact us at info@assaia.com or via the website form.