Jeppesen Pairing & Rostering Johan Kristofferson Global Program Sales Leader Crew, Ops & Analytics October 5th, 2016
What successful Airlines have in common Network (Commercial) Aircraft planning Fleeting Strategic pairing Pairing Manpower planning Aircraft planning Tail Assignment Rostering 18 months 6 months 2-6 weeks 7 days Flight Deck Management / Cabin Crew Management Tracking Tracking = KPI s/ $$$
What successful Airlines have in common What successful airlines have in common Clear targets. A shared definition of good. KPI s corresponding to the targets Strong focus on data quality Well defined planning process interfaces Follow-up routines including reporting Adaptability. Ability to change processes and targets when needed Aircraft planning Fleeting Strategic pairing Network (Commercial) Manpower planning Aircraft planning Tail Assignment Pairing Flight Deck Management / Cabin Crew Management Rostering Tracking Tracking = KPI s/ $$$
What we provide Efficiency More efficient and robust operation Facts Easy to evaluate detailed scenarios before taking decisions Process Fact based process changes, fast to implement LONG-TERM PLANNING PLANNING OPERATIONS + FOLLOW-UP Lower cost, reliable plan Lower cost, fly more, analysis Lower cost, crew bidding Cost control Manpower Planning Crew Pairing Crew Rostering Crew Tracking Long-term planning. Build duty patterns. and assign to crew Repair plan Tail Assignment Ops Control Assign flights and maintenance to tails Repair plan Fatigue Mobile Lower cost, fly more, robustness Cost & revenue control
Six Differentiators Driving efficiency Control work-life balance Manage change Schedule safety Integrating systems and processes Regular new releases Manage growth and reduce costs while protecting stability Enable crew to influence rosters Analyze and implement business changes fast Measure and improve schedule safety Integrate across company silos Continuous improvement Jeppesen Proprietary - Copyright 2016 Jeppesen. All rights reserved.
Crew Pairing
Jeppesen Crew Pairing Customers
Tactical Execution Strategic Thinking Driving Efficiency Superior Efficiency through best-of-breed optimization Crew Pairing as a strategic tool
The value of best-of-bread Optimization Keeping crew Variable together crewing Qualifications What-if scenarios Trip regularity Retiming Fatigue risk management Augmentation Dynamic aircraft rotations Robustness Rules agreements Base Balancing 9
Evolution of Crew Pairing as a Strategic tool Strategic Thinking Input Timetable Rotations Headcount Hotels Bases Crew Pairing Input Timetable Aircraft Rotations Headcount Hotels Transports Input Timetable Rotations Headcount Hotels Bases Crew Pairing Input Timetable Rotations Headcount Hotels Bases Crew Pairing Very Limited Capabilities Constrained Capabilities Unconstrained Capabilities
Examples Dynamic Aircraft Rotation Identify forced turns that improves the Crew Pairing solution Re-timing Identify small changes to the time table that improves Crew Pairing solution Dynamic Base Balancing Let the optimizer identify optimal base balance
Crew Pairing & Aircraft Rotations Crew usually follows the aircraft part of a day: Increase robustness More efficient pairings Changing aircraft turns can: Improve the robustness Additional cost savings 1
Dynamic Aircraft Rotation Identify forced turns that improves the Crew Pairing solution From a crew perspective we would like 102 and 105 to be connected 102 105 102 105 102 102 105 105 Too short to allow aircraft change for crew A different routing allows crew to stay on aircraft
Aircraft Rotations Workflow 1. Create reference solution with optimizer 2. Create solution with Dynamic Aircraft Rotations 3. Communicate findings to Fleet department 4. Receive new rotations from Fleet 5. Create solution based on new rotations Direct costs Block per duty Duty days Aircraft changes # Aircraft cxn Crew Pairings Reference 1181 k 4:20 1746 354 1689 Crew Pairings Dynamic Aircraft Rotations 1167 k 4:24 1722 328 Crew Pairings New Rotations 1170 k 4:23 1728 332 27 changes identified 24 changes accepted -1% -6% 1
Re-timing Retiming is about changing the timetable (a little bit) so that a better crew schedule can be accomplished Example: Leg A can t connect to leg B due to connection time being to short. A B 0:30 < 0:40 not OK Re-time leg B 10 minutes! +0:10 A B 0:40 >= 0:40 OK!
Re-timing workflow 1 The planner sets possible retiming alternatives down to each individual flight A A A B B B 2 Optimizer considers all alternatives but will chose the alternative that gives the best overall cost 3 Created solution includes flights including retimings
Unconstrained Bace Balancing Add realistic base options Remove base constraints Run Optimization Jeppesen Proprietary - Copyright 2016 Jeppesen. All rights reserved.
Dynamic Base Balancing Original Base Distribution Pattern Optimizer Suggestion
A culture of strategic thinking! Analysis Idea Hours! Optimizatio n Model
Crew Rostering
Jeppesen Crew Rostering Customers
I want a cup of coffee...
I want a cup of coffee... Decaf Caffè Latte Caffè Mocha White Chocolate Mocha Caffè Americano Vanilla Latte Syrup Flavored Latte Cinnamon Dolce Latte Espresso Caffè Au Lait Cappuccino Espresso Macchiato Caramel Macchiato Cinnamon Spice Mocha Pumpkin Spice Latte
I want a bidding system... Bid ratio system Golden / joker / trump days (off) Bidding window Bid groups Requests Auction bidding Rolling seniority History Fairness Bid lines Strict seniority
Crew Rostering - Overview Strict Seniority Fly below rank Robustness Fatigue risk management Fairness Mixed fleets/ranks Rules agreements What-if scenarios Crew bids Recency Base variants Recurrent training Keeping crew together Vertical constraints Qualifications 25
Strict Seniority PBS Weighted bid model Allows the crewmember to weigh different properties of trips and rosters Each bid is associated with a weight AVOID work 25-26Dec AWARD check in > 10:00 AWARD check out > 8:00 AWARD layover in CDG 500 points 300 points 100 points 10 points So for each crewmember in seniority order: as long as overall constraints are met! The total sum of the weights is maximized for that crewmember All trips not yet assigned to senior crew are considered 26
Strict Seniority PBS Bid groups model Allows the crewmember to bid for a number of different scenarios - Bid Groups Bid Group 1 AVOID work 24-26Dec AWARD layover in CDG AWARD layover in FCO HIGH LOW Bid Group 2 Bid Group 2 AVOID Bid Group work 25-26Dec2 AVOID work 25-26Dec AWARD AVOID Time work Away 25-26Dec from Base > 30:00 HIGH AWARD Time Away from Base > 30:00 HIGH AWARD Time Time Away Away from from Base Base > 20:00 > 30:00MEDIUM HIGH AWARD Time Away from Base > 20:00 MEDIUM AWARD Time Away from Base > 20:00 MEDIUM So for each crewmember in seniority order: and as long as overall constraints can be met! Bid Groups are evaluated in a sequential order until a legal roster can be awarded A Bid Group is awarded (i.e. fulfilled) if a legal roster can be built using ONLY trips satisfying the bids in the Bid Group All trips not yet assigned to senior crew are considered 27
Jeppesen Award Process Strict Seniority Initial Award Constrained Solution Sequential optimization process awarding lines to crew in seniority order Objective Maximize bid fulfilment in seniority order Affects all crew
How Can the World s Best Optimization Help a Crew Member? Crew Member 1 Bid Group 1 Award Trip number 3045 Limit 6 Award Trip number 3028 Limit 3 Award Trip number 3033 Limit 2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 3045 3045 3045 3045 3045 3045 3033 3028 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Crew Member 2 Bid Group 1 Award Trip number 2356 Award Trip number 3028 Limit 5 Award Trip number 3045 Limit 1 3045 3045 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 3045 3045 2356 3045 3045... 3045 3028 3028 3028 3028 3033 3028 Jeppesen Proprietary - Copyright 2016 Jeppesen. All rights reserved.
Senior Alternates Results Jeppesen Proprietary - Copyright 2016 Jeppesen. All rights reserved.
Jeppesen Award Process Strict Seniority Initial Award Constrained Solution Global optimization Objective Ensure global constraints (Stacking, Reserve Levels, etc...) are met Only affects required number of junior crew members in reverse seniority order
Constrained Solution Primary objective - Optimization process focusing on the overall constraints Secondary objective Maximum consideration of bid preferences The most senior crew affected by the constrained solution is defined by the agreement. Best-of-bread optimizer ensures least amount of crew is affected in reverse seniority order #1 #2... VAC V
Constrained Solution Primary objective - Optimization process focusing on the overall constraints Secondary objective Maximum consideration of bid preferences The most senior crew affected by the constrained solution is defined by the agreement. Best-of-bread optimizer ensures least amount of crew is affected in reverse seniority order #1 #2... VAC V
Constrained Solution Primary objective - Optimization process focusing on the overall constraints Secondary objective Maximum consideration of bid preferences The most senior crew affected by the constrained solution is defined by the agreement. Best-of-bread optimizer ensures least amount of crew is affected in reverse seniority order #1 #2... VAC V
Crew Web & Mobile
Next Generation Crew Web & Mobile Crew Access (PAL 2014) Crew Bid (GC 2016) Crew Request (KLM 2017) Crew Exchange (Alaska 2016) Crew Bid Bid Group (Undisclosed 2017) Crew Leave Crew Career
Crew Modules vision Anywhere at Anytime user experience made easy, so that bidding and trading can be done when & where needed. Simplicity configuration so that the focus is business logic integration, but the User Interface remains generic. Decline one off complicated solutions rather than squeezing them in. Up to date a simple way to deploy and migrate to new core features modules in production. 3
CrewBid Point Model Crew Bid The weighted model for strict seniority is used and recommended for large planning areas. Point model same module used for WFS & Strict Seniority Combination bids are standard in Crew Bid Strict Seniority preferences Trip properties Roster properties expressed as Rule Relaxations / Global Bids Crew can create backup bids so that bids work for both reserve lines and flight lines.
CrewBid - Bid Groups The Bid group model is used for small to medium planning areas in the strict seniority market. Bid Groups are powerful to build and express different scenarios. Process support Possibility to express preferences for flight line or reserve line as pre-process Crew creates backup bids so that bids work for both cases Dynamic Bid Model Bid groups Roster property defined in top Bid Group level, so that crew can decide and express which Bid Groups are for Flight Lines and which ones are for Reserve lines. Expressing preferences as bid groups using bids with Award/Avoid logic within each Bid Group Create, Sort and Copy bid groups Create and Copy bids within the bid groups Copy a bid from one bid group to another bid group Qualifiers Rule Relaxations & Preferences (Global Bids) Bid group legality validation
Crew Bid Bid Groups Jeppesen Proprietary - Copyright 2016 Jeppesen. All rights reserved.
Thank You Jeppesen Proprietary - Copyright 2016 Jeppesen. All rights reserved.