CREW SCHEDULING AND MANAGEMENT SOLUTIONS

We offer a range of solutions to suit a variety of crew scheduling situations

Local South African carrier, FlySafair has been ranked the world’s most on-time airline for 2017 by air travel intelligence specialist, OAG. The airline achieved an annual on time performance (OTP) of 95.94%. It was also one of just three airlines to receive a five-star rating. This is the first time that OAG has introduced star ratings based on carriers’ ontime performance. FlySafair uses the Webb Elgin Associates WECOMS Crew Ops Management System

The key benefits of the WECOMS solution are its simplicity, availability, and economy.

The system now also tracks the following:
• Aircraft assignments
• Aircraft maintenance checks
• Fuel uplifts and burnoffs
• Ground service providers
• Visual flight progress

  
  Read more about the WECOMS crew scheduling system

  Go to the WECOMS Crew Scheduling system demo playground

  Request a WECOMS system Brochure
  

Try our handy Flight and Duty calculator

1. Introduction

Crew scheduling optimisation is an area that receives much attention in the airline industry due to the high manpower cost of crew. Optimisation, and in particular crew pairing optimisation, is offered as a high-ROI investment which promises significant crew productivity benefits. In more recent conferences on this subject there appears to be an emergence of a strong body of opinion that supports the view that more emphasis needs to be placed on operational recovery rather than a priori optimisation.

2. Pairing Optimisation Overview

The goal of pairing optimisation in summary is to create strings of flight duties from the airline flight schedule that can make maximum use of crew legal duty periods, thereby creating the most “productive” schedule. Historically, the concept of creating crew pairings probably arose from the need to simplify the crew rostering task. In the days before cost-effective computing power was available in this domain, the manual crew-scheduling task could be massive, and complicated. Creating pairings that repeated themselves in daily, weekly, or monthly patterns was an effective way to reduce the amount of manual calculation. Simply put, an airline schedule containing 200 sectors per day would contain 6000 sectors per month. The number of pairings that could cover such a schedule could be as low as 200 if there is high daily repetition. It is therefore easier and quicker to roster crew to pairings than to flights.

During the 1990’s a number of software vendors launched Pairing Optimisation initiatives, usually on the back of academic research. Sweden and Canada were particularly active in this arena. The vendors have developed very expensive sophisticated algorithms that attempt to build the best pattern of crew pairings. These systems claim high ROI by generating low percentage savings potential in a very high cost area. A single percentage point improvement in crew productivity could generate $2-3 million of annual savings potential in a medium sized airline.

3. Optimisation Realities

Whilst it is true that Pairing Optimisation systems offer the potential to create more efficient crew pairings, there are a number of factors that must be considered when analysing the true business benefits of pairing optimisation:

Ø
The mathematics involved in airline crew pairing optimisation is far from trivial. Most of the vendors operating in this area employ a fair number of doctorate level mathematicians and operations researchers to lead and develop such systems. This leads to complex system maintenance and high cost of development.

Ø
Airline crewing analysts can often manually, or semi-automatically, create pairing solutions which seem to be better than the system generated solutions

Ø
Even with today’s technology, solutions can take days to produce. In a world where airline schedules change frequently, the optimisation activities cannot always keep up with pace of change.

Ø
Pairings have a human or emotional dimension, with mathematics cannot always explain, and therefore a good set of pairings in the view of the airline and its crewmembers may not be the mathematically optimised set of pairings.

Ø
Optimising pairings without considering the “rosterability” of the pairings is too narrow a view. There is a need for pairing optimisation to also consider the broader problem of roster optimisation. Pairing solutions are usually not created with a real-world view of actual crew availability in a particular period, or sub-period.

Ø
The optimisation “savings” are often not realised in practice due to the operated schedule differing from the planned schedule. Due to airline business processes, and the time taken to generate solutions, it will often be the case that the actual flown schedule contains many deviations from the schedule used to generate the optimised pairings. Even one or two small changes to operational pairings can negate the planned optimisation savings.

Ø
Optimised pairings are usually very tight and have extensive interactions, often resulting in severe promulgation of disruptions. The planned optimisation savings can be negated in these cases by the costs of repairing the crew patterns. Recent case history in Europe shows that over-optimisation can even result in extensive schedule cancellations due to crew unavailability.

4. Recommended Strategy

We recommend an holistic approach along the following lines:

Business alignment – an airline must determine its trade-off sensitivities at a strategic level. Optimised pairings will offer a quantified financial benefit potential at the cost of schedule reliability, crew satisfaction, and rosterability. A low-cost carrier will take a different view on these tradeoffs than will a state-owned flag carrier. A highly unionised airline will have different pressures. There is also the decision to be made concerning the level of “slack” to be provided in the number of crew employed.

Agree a set of crew scheduling rules – an airline should define a set of schedule shaping rules which “personalise” the optimised pairing solution to its unique circumstances, and capture some of the intuitive rules that the crew analysts often apply in their manual solutions. Such rules can also weed-out those pairings that have proven to be operationally unflyable, or otherwise unacceptable, for various reasons.

Disruption modelling – models which will use past delay history to build a simulation routine for future schedules and therefore attempt to predict the disruption potential of a future schedule based on past experience, and to assign a predicted completion or reliability index to each planned pairing.

Define buffers – based on operational circumstances, an airline should specify the use of rule buffers, which can isolate the crew schedule from some of the commonly encountered schedule disruptions. These buffers must be applied differently to different pairings depending on the disruption probability for each pairing, and the knock-on potential of the pairing.

Operational repair – ensure that any systems used provide the capability to re-optimise as the schedule changes or becomes disrupted, in order to preserve the cost savings.

Roster interaction – the final cost of a crew schedule and its reliability, as well as the crew quality of life, are a function of both pairings and roster, and neither of these must be looked at in isolation.