|
AGIFORS Airline
Operations 2008 |
||||||||||
|
Technical Program (Based on submitted abstracts to date*) These submitted abstracts have been conditionally accepted for inclusion in the AGIFORS Airline Operations conference agenda. Final acceptance is contingent on 1) receipt of complete presentation on/before 30 April 2008, and 2) approval of presentation content by the AGIFORS technical committee. Monday - 19 May 2008
Keynote: KLM VP Operational Plan Management What is really happening at the day of operations? Many experts are year over year involved in designing a new time table. Many sources are used as input, however until today it is still a challenge to track down what is really happening at the day of operations. Still in the planning phase, adjustments to the timetable are still possible since resources are available, there is time enough to find the best solution and changes will be stored in the logbook nicely. However coming closer to the day op operations, flexibility decreases, the number of problems increases and we are running out of time since the passengers have to leave. Only if we are able to really track down what is happening on the day of operation, we can develop a more robust schedule, and therefore improve our daily operations and our customer comfort.
Simulating the Stability of Flight Schedules by the Use of a Capacity
Model For the generation of robust flight schedules it is mandatory to take into account the variablities of the day-to-day operation. The generation of robust flight schedules requires a reliable forecast of the operational interrelations. Simulation is the most powerful instrument to estimate the operational stability for a given schedule. Lufthansa has set up an event-driven simulation model called “SimOps” to analyse rotationally connected flight schedules regarding their punctuality and performance. Recently two more features have been added to the simulation model in order to get a more realistic picture of the main influences and their impact on the stability. Delays are dependent on various external factors such as weather conditions. By generating random process times for the simulation based on historical data, it is assumed that process times within a time interval are stochastically independent. Clustering the data into days of good, average and poor performance and evaluating them seperately leads to more realistic simulation results. Traffic increase at already congested hubs results in further deterioration of the performance and punctuality of the network. Delays caused by high capacity loads are dependent on the demand and the combination of wake-vortex classes in the sequence. Therefore a model has been developed to forecast congestion driven delays for approach and line-up processes.
Maintenance Reachability – Robust Schedule and Route Construction at
Southwest Irregular operations often ruin the best-laid aircraft maintenance plans, as aircraft bound for maintenance bases get re-routed for other reasons. At Southwest Airlines, this problem is further complicated by the need to balance maintenance work across multiple maintenance locations. This presentation will cover some planning techniques and metrics that Southwest Airlines is using to make the flight schedule more robust for handling both planned and unplanned maintenance events, as well as balancing work at the maintenance bases. Tuesday - 20 May 2008
Keynote: The New Dutch Railway Timetable In December 2006, Netherlands Railways introduced a completely new timetable. The timetable was designed to facilitate further growth of passenger and freight transport on a highly utilized railway network on one hand, and to improve the robustness of the timetable in order to reduce the number of train delays in the operation, on the other hand. The construction of a railway timetable from scratch for about 5,500 daily trains was a complex puzzle. To support this whole process, sophisticated Operations Research techniques were used to generate several timetables, one of which was finally selected and implemented. Furthermore, the costs of a railway operator are mainly determined by the rolling stock and crew schedules. Two innovative Operations Research tools were used to come up with efficient schedules for these two resources. The total additional annual profit for the company, as a result of the new timetable and the more efficient resource schedules, can be estimated at 70 Million euros (105 Million dollars). However, the benefits of the new timetable for the Dutch society as a whole are much higher: more trains are transporting more passengers on the same railway infrastructure, and all these trains run more on time than ever before. As a result, future growth in railway transportation demand can be handled by rail transport, allowing cities to remain accessible and reducing the emission of greenhouse gases.
Flight Operations Health Check Many Flight Operations departments try to achieve strategic company goals. Often these Goals fall along the lines of operational, financial and customer service targets set by senior management. In order to measure if these targets are being met, management is focusing on Key Performance Indicators (KPI's) such as OTP. The Challenge for the Flight Operations department is then to maintain or more often improve efficiency in-order to meet these KPI’s. Because most KPI’s are not being met, or at least not every day, Flight Ops management is faced with difficult decisions in how to fix issues resulting from areas not necessarily visible to them. The Health Check methodology implements up to 200 KPI’s into a Flight Operations department which can be measured. These KPI’s can then be compared to industry best practices through which improvement recommendations for gap closure are provided. This presentation will give an overview of the methodology and will cover industry standard KPI examples and industry best practise benchmarks.
Improving Operational Performance Without Increasing Planned Costs Airline plans that are "optimal" on paper may not be optimal in practice. In fact, attempts to optimize the plan often limit the slack in the system, thereby reducing the opportunities to recover when disruptions occur. On the other hand, adding slack to the system means either more resources or fewer flights, which translates to higher (planned) costs or lower (planned) revenues. Thus, it can be difficult to justify in the planning process qualities that are critical in operations. We present a method for re-allocating the existing the slack in a network, maintaining the existing planned costs and revenue projections but decreasing the expected delay propagation in operations.
Virtual Spares Many airlines schedule aircraft connections at their hubs on a first arrive-first depart basis. Others deviate from this discipline to correlate aircraft connections with passenger connections, or to equalize aircraft utilization. Recent studies in the field of robust scheduling have shown that more robust schedules can be developed by correlating the ground-times of aircraft with the expected delay "upstream" of connections. However, the resulting schedules are not necessarily amenable to the swapping of aircraft between maintenance routings, a tried and true strategy for dealing with the realized (as opposed to expected) disruptions. I present a novel idea for scheduling aircraft connections that leverages the beneficial features of both approaches to mitigating the effects of disruptions. By scheduling the first arrival from an origin where operations are very reliable, and scheduling that same aircraft as the last departure and to a destination where operations are also very reliable, we create a "virtual" spare aircraft that can be swapped with any aircraft that is delayed. A similar strategy can be employed with the second, third, and other early arrivals to create additional virtual spares, subject of course to marketing considerations.
Simulating Flight Delays in Airport Ground Staff Scheduling Ground staff scheduling still often relies on deterministic optimization models. However, especially in ramp operations, workloads can vary considerably depending on the magnitude of flight delays. In practice, this often leads to problems in service quality or to excessive overtime assignments. Based on empirical flight delays, we derive a model for the simulation of delays in task-based staff scheduling. Based on a real-world case, we show how delays impact on workloads and their coverage. We also give directions on how staff scheduling models can be improved by taking stochastic flight delay information into account. The
Potential of Tail Assignment Optimization The operational robustness of a tail assignment is determined by criteria like crew changes or suitable distribution of ground times buffers. In practice, however, there is a strong focus on simply finding a maintenance feasible solution. It is not uncommon that other quality criteria are only considered by applying local and often manual adjustments. In this talk we present some thoughts on the potential that can be unleashed by adressing the tail assignment problem as a global optimization problem. We report on solution techniques and numerical results, and investigate the trade-off between different optimization criteria.
Panel Discussion: Linking Planning and
Operational Performance Airline plans that are "optimal" on paper do not always perform well operationally. In fact, a planned schedule that attempts to maximize revenue while minimizing cost will typically have limited slack, limiting the opportunities for recovering when disruptions occur. But developing "more robust" plans is easier said than done: What characteristics of a planned schedule make it more robust? How should robustness be measured? How should robustness be valued? How should robustness be validated? This session provides the opportunity for sharing ideas and insights about this challenging topic. Wednesday - 21 May 2008
Trends in Flight Operations. Future HCC vs. OCC Many airlines are trying to improve their transfer management on their main Hubs. In order to have a better visibility and ability to manage the transfer of Costumers, Bags, and Aircraft, dedicated Hub Control Canters have been introduced. Since then the industry has witnessed a shift in many main operational decision making points from Operations Control to Hub Control. The main question, however, is how airlines can best re-align their complete organization and various responsibilities and ensure optimal process visibility consistent with industry trends. If achieved, improved OTP and more efficient turnaround of AC will occur. This presentation will give an overview of current best practice HCC / OCC concepts and will cover industry standard KPI examples and industry best practise benchmarks for continued performance improvement.
Airline Operations Control Modeling Most airlines are organized in functional silos, which are usually constrained by contradicting goals and rules. Resulting fragmented processes proves suboptimal at best and can be costly, slow and inconsistent in operations. In attempt to provide an efficient holistic solution, we have developed an analytical approach and created the airline business process model that maps identified functional areas and processes. Cross-functional processes are showing interrelated tasks at different levels of granularity. The developed model helps to measure efficiency, detect bottlenecks, create what-if scenarios and build robust, integrated mathematical models. Examples that highlight the value of our approach will be provided.
Airbus A-380 Aircraft Boarding Strategies A Simulation Study The boarding of passengers into the double-deck high capacity A-380 aircraft by Airbus presents a major challenge for the airlines. This study is focused at developing a Monte Carlo simulation model to measure the duration of boarding process under different scenarios. The study considers the A-380 seating configuration delivered to and operated by Singapore Airlines. This study considers various boarding strategies with 1,2 and 3 jetways and doors for boarding passengers. We adopt major popular boarding strategies such as back-to-front, window-middle-aisle and random to evaluate the total boarding times. The results are very encouraging and certainly prompts the need to initiate a study at A-380 with different seating capacities. A
Well Oiled Engine with ESCapE Increased work load and closure of similar services abroad combined with understaffing maes personnel planning crucial to maintain and improve the performance of our Engine Services department. For this reason we developed a tool, called ESCapE, that shows the effects of personnel planning on turn-around-time of engine parts. The tool shows that understaffing, leave, etc. for certain skills not only reduce on-time performance, but also imply long idle times for mechanics with other skills. With the help of what-if scenarios management is supported to enhance personnel planning, establish the right mix of skills and to anticipate changes in workload. Workload does not only consist of actual, open orders, but also of a forecast for the coming 4 weeks. Output of the model is expected workload, taking into account individual skills and rosters. Moreover, the program computes statistics of expected throughput times. A
Simulation Study to AirTran Airways Super Tug Operations AirTran Airways is among the very few successful and profitable airlines in US. As the prices of jet fuel continue to increase the airline initiated a study with Embry-Riddle Aeronautical University to conduct a feasibility study aimed at using super-tug to transport aircraft to and from the gates at concourses C and D at Atlanta Hartsfield-Jackson International Airport (ATL) to their maintenance facility. This study investigates the possibility of reducing costs through saving jet fuel. The airline needed this study to analyze and justify the prohibitive expenses associated with purchase and maintenance of these tugs. This study adopts simulation with animation capability to analyze the annual savings by studying the numbers of super-tugs needed, as well as the utilization and operation cost. The results are very encouraging and provided the airline the opportunity to clearly evaluate the cost and benefits of purchasing new super-tugs.
Multiobjective Gate Assignment Problem Gate assignment problem has recently received some attention with most of the effort directed towards developing heuristic approaches that would allow capturing multi-objective nature of this problem. We used a more traditional IP formulation and were able to model most of the practical requirements for gate assignment as well as establish a few links with aircraft scheduling process. We consider several different aspects of gate assignment optimization such as resource utilization, passenger satisfaction through walking distance minimization and special constraints formulation, and schedule robustness via safety buffers, flight retiming and multiple gate assignment. We present computational results for several problems of the realistic size and verify feasibility of our solutions. Thursday - 22 May 2008
Optimizing Composition of Flights in Peak Hours at Hub Airports Amsterdam Aviation Economics Hub carriers try to connect markets by scheduling arrivals and departures around waves at their hub airports at several wave moments a day. Generally this leads to a large number of arrivals, followed by a large number of departures in limited time periods. The peak hour capacity of the hub airport is often the limiting factor, when the airline tries to concentrate its arrivals and departures. The question arises for airlines which arrivals and departures (spokes) are critical in the sense that they need to be scheduled in the absolute peak hour or that they can – without loosing too much market share - be scheduled slightly before or after the absolute peak hour. The connecting time is taking a relative large part in total elapsed travel time in short haul markets. As connecting time is perceived as more inconvenient than flight time, a small increase in connecting time leads to a relative large deterioration of the product offered. This deterioration may lead to market loss, if this particular connecting market is served by other competing airlines via their hubs. In case however, there is limited or no competition, the market loss resulting from this deterioration is smaller, and hence the connecting time in this connecting market is less critical. Connecting time in longer haul market is less critical. On the other hand, competition may be more intense, as more competing hubs are involved, which may make connecting time also critical in longer haul markets. The paper identifies for the hub Amsterdam critical spokes, by determining the average competition level of the markets connected to these spokes as well as the average (geographical) length of these markets. This leads to conclusions with regard to the financial robustness and sustainability of the hub and the extent to which some spokes may be removed from the peaks or some other spokes may need to be re-allocated.
Integrated Gate and Bus Assignment at Amsterdam Schiphol Airport At an airport a series of assignment problems need to be solved before aircraft can arrive and depart and passengers can embark and disembark. We consider the gate assignment and the planning of platform busses at Amsterdam Airport Schiphol (AAS). Currently these two problems are solved sequentially. We investigate their integration, which enables the use information regarding the bus planning in the gate assignment. For both problems, we have developed a column generation algorithm. Our main objective is to maximize robustness, which is achieved by minimizing a specific function of the idle times. Recently, we have extended and integrated our two algorithms into an algorithm for the integrated problem. Our algorithm uses stabilized column generation. Computational experiments with real-life data from AAS indicate that our algorithm is able to compute a planning for one day at Schiphol in a reasonable time.
Fairness in the Aircraft Landing Problem The last few decades a lot of attention has been paid to Collaborative Decision Making in Air Traffic Management. We consider the use of airline preferences in the aircraft landing problem (runway scheduling). Fairness (among airlines) becomes an important consideration and is explicitly considered in our model. The results show that large cost savings compared to the FCFS schedule that resembles current practice can be obtained. Different heuristics give schedules with different trade-offs between efficiency, cost, delay and fairness. Hopefully, these results can be a starting point for discussing the fairness issues related to CDM processes among air traffic stakeholders. Vendor Exhibition
Call for Technical Papers Come and share with us your ideas, practical innovations, current trends, philosophies, and latest advances on the topics that matter most to you.
Based on participant feedback and recent survey results, there continues
to be a high level of interest in examining the critical role of
operations throughout the entire “schedule lifecycle” from robust
schedule planning to day-of-operations management to post-operations
evaluation and feedback.
If you are interested in presenting at the study group meeting in Amsterdam, please complete the on-line registration form at www.agifors.org/studygrp/opsctl/2008/present.html. As always, talks are subject to approval, and time slots are available on a first-come, first-serve basis - so if you are interested, act now! Abstract Deadline: 31st March 2008 Complete Presentation Submission Deadline: 30th April 2008 * Technical Program is subject to change The AGIFORS Airline Operations 2008 conference technical program is currently being finalized, for more information please contact tim.niznik@aa.com. Please refer to previous years conference proceedings for a complete listing of technical talks given in the past at AGIFORS Airline Operations meetings. |
|||||||||
Welcome | Agenda | Registration | General Information | Hotel Information | Host City |
||||||||||