Good Data and Good Decisions

The first thing airline schedulers tell us they need to make decisions is “good data” to know which changes are better than others.  They would like to have:

  • profitability data by flight segment,
  • current booking data from the reservations system (daily update), and
  • forecast revenue for each flight.

Some schedulers also mention that they would like data to help them know:

  • the risk of violating an airport’s 80:20 rule and losing a valuable slot, and
  • whether a change violates operational constraints, gate limits or maintenance rules.

You can guess that the second thing schedulers want is a system that enables them to use of all this data in one place.  Can intelligent decision support tools be designed to solve problems within the scheduling system? Today’s planners often have to pull data from multiple systems, use manual Excel sheets, or even send their schedule to other departments. The result can take a lot of time, create conflict and lose much of the commercial value that was planned into the schedule.  When a crisis happens, there usually is little time for this.

Recent examples of situations that required quick decisions:

  • Major issues forcing cancellation of multiple flights (Max issue, COVID-19, etc.)
    • mass edit capabilities are needed to automate the process for a whole period and enable quick execution of decisions
  • Loss of one or more aircraft for any reason
    • booking and profit data are needed to identify the right flights to cancel
    • swaps to put those flights on one tail to cancel should be automated
    • finding the best flight pairs to cancel (one may be unprofitable but the other very profitable) requires seeing the data for both
    • running analytics reports helps compare scenarios for optimal choices
    • avoiding the risk of cancellations impacting historic slot rights requires slot monitoring data to automatically flags flights at risk

Zulu has integrated this data into our scheduling module:

  • Cancellation decisions can be based on data using decision support tools
    • to eliminate the scramble with other systems or manual analysis steps
    • to automate solution development and compare scenario options
  • Potential up-gauge/down-gauge opportunities can be optimized
    • using booking data directly in the editing view or in analytic reports
    • by running fleet optimization tools on the whole schedule
  • Maintenance opportunities can be designed into the schedule early
    • where data and rules are visible for quick decisions or for use within intelligent tools (solvers/optimizers) for more sophisticated solutions.
    • The intelligent rotation optimizer assures all rules and constraints are met
  • Multiple scenarios can be created quickly and compared in Zulu’s analytics module.
    • Any violations of rules or tradeoffs of key metrics are flagged for the user to review and make choices using their experience.
  • Updating the organization on the effects of the changes can be “real-time”
    • Zulu’s analytics tools can compare multiple scenarios in standard reports or by creating user-defined custom analytics in dashboards
  • Discussions with other departments can focus on tradeoffs of goals rather than ending up with a bias toward any one goal and losing significant value or efficiency from the schedule.