Automatic Disruption Handling in Crew Management. Is that possible?
All you have to do is to push the “Find Solution” button and the system will generate for you the best suitable solution. Then, the next step is to push the “Apply Solution” button and the changes will be applied, notification generated and sent to crew members to have them ready to fly.
Awesome, isn’t it?!! But what is the “best suitable solution”? How long does the system take to provide the solution? How can the user check that the solution is really the best one? What happens if in the meantime something changes? Those are only few of the dozen of questions and doubts a crew expert usually asks when someone tries to convince him that it is possible to solve problems automatically during IROPS (Irregular Operations).
Which problems does an Airline face during an IROPS situation? Let´s try to summarize the problems in three main groups:
Aircraft Re-routing – When a disruption occurs, the first problem to face is that one or more flights won’t proceed as planned. The root cause may differ from bad weather condition to maintenance to geographical or political reason. The ops controller normally located in the OCC (Operation Control Center) has the task to restore all the flights, assigning to them the right aircraft, at the right new estimated time and in the right place, reconnecting future flights with available aircraft.
Crew rotations – it is similar to the aircraft re-routing but for each flights several crew members are affected (one flight with a problem means up to 12-14 crew member involved in the long haul flights) and in addition the legality rules for crew are much more complicated than the ones for aircrafts. Furthermore, crews are humans so they need a place to sleep, food and beverage, a way to leave the airport, reach it and go back home.
Passenger Re-accommodation – If a flight is cancelled or delayed, passengers must be rebooked on some other flight or combination of flights in order to reach their final destination. This process can be very expensive and time consuming especially if the Airline needs to re-book passengers on other airlines and if the re-booking is not done fast enough.
What is the right approach to solve these three problems? Which problem needs to be tackled first?
In the ideal world an integrated system, with an integrated database, with the proper amount of senior management support (that is more and more lacking nowadays due to organizational issues) should provide the possibility to look at the three problems simultaneously providing an end to end solution capable to optimize and minimize in terms of time and costs the impact of an IROPS situation.
Nowadays, many vendors offer software products that address IROPS issues to some degree. However, there are strong indications that the market for IROPS solutions is still immature. But Airlines are interested and attentive in such market and lot of investments and activities are actually ongoing with the result that many progresses will be expected over the next few years.
Coming back to the main topic, we should consider that a crew manager normally focuses only on the crew rotations problems. Once the new aircraft rotation is published the crew department starts to rebuild the crew rotations, trying to keep costs down and to employ as less crew as possible. The countermeasures the crew manager normally has can be summarized in the following:
- Swap crews
- Use standby / call from off
- Use ferry flights or positioning flights.
- Ask for additional delays or cancellations
The order of possible solutions to apply may vary depending on the airline policies and the type of problem and network. I often visited airlines where some of those solutions couldn’t be applied for structural reasons and the only possible option was to increase delays and cancellations.
Despite the complexity and the size of the airline, the high amount of crew members normally involved in a disruption should definitely push Airlines to invest in order to interactively support the crew and ops controllers with analytics or artificial intelligence methods to rank and order possibilities by improving significantly the purely manual operation. Such “interactive” approach can already provide high benefits in the main steps of the disruption handling approach and in particular:
- In detecting problems that are affecting the operations (management by exception)
- In simulating/forecasting problems that are going to happen (what-if and scenarios)
- In providing alternatives while trying to solve disruptions (solution proposal)
This is not exactly the “apply solution” button everybody is looking for, but it definitely is a mandatory step before moving into the age of the optimization for the disruption handling.
Will we never get the chance to solve these complex problems in a fully automatic way?
My answer to this is definitely yes. We will soon benefit of the progresses of optimization and general improvements in computing power and costs but it will still take some time to fall in the optimization age of the automatic disruption handling.