 |
Optimized scheduling produces significant savings in crew costs
|
Business impact
A reduction of 5-8% in variable crew-related costs and an increase of 4-7% in crew utilization.
Issue
The second largest operating cost for an airline is the cost of flight crews. El-Al wanted a more efficient way to build schedules that would reduce the variable cost of flight crews and crew utilization as much as possible.
Executive summary
A crew scheduling solution for El-Al considers billions of possibilities before determining crew assignments that meet complex criteria. The solution is easy to use, reduces variable crew costs and improves crew utilization.
What IBM did
El-Al builds its schedule about a month in advance. In addition to which planes fly where and when, round trips for each type of crew member must be planned. The large number of variables that are considered in building the schedules is staggering and includes legal requirements for rest and duty, equipment ratings, home bases, overtime, crew preferences and options for overnight or heading home at the end of a day. And many of these variables have a direct impact on cost.
IBM Research was brought in to develop a crew scheduling application that would satisfy all legal criteria, completely cover the flight schedules, reduce costs, optimize use of resources and be easy to use. For El-Al, the solution was built around three elements:
Using advanced mathematical models and algorithms developed by IBM Research, an optimizer was developed that works through the billions of possible combinations and finds a solution that meets all the criteria. Parallel processing was used to improve computational efficiency.
The application includes a user-friendly interface that seamlessly ties together all the elements of the solution and allows schedule adjustments to be made as needed. In the end, not only did the solution reduce the variable costs related to crews and improve crew utilization, but it also reduced the time necessary to produce a schedule.
Capabilities applied
IBM Research's success in solving this sort of vast problem is the result of years of experience in advancing core optimization methods and architectures for high performance parallel computer processing. This experience allowed IBM to deliver a valuable solution to El-Al's problem of crew scheduling.
Related case studies
More on the issue
Hospitality and leisure 2010: Experience rules
IBM Global Business Services consultants are thinking ahead to a fundamental shift from traditional hospitality and leisure offerings to a new paradigm of individually customized travel experiences.
More on the team
Avraham Harpaz
For over 18 years, Avi Harpaz has been working on optimization algorithms that solve large scale problems involving huge numbers of design variables and constraints.
More on research
ECO.2000
ECO.2000 is a powerful, user-friendly crew pairing and assignment system, designed to meet the rigorous and dynamic requirements of commercial airlines.
More on research
IBM makes advances in airline automation
IBM researchers have played a fundamental role in the development of the underlying mathematical optimization technology for airline scheduling problems.