Airline Data Research Articles

Seasonal time series forecasting: a comparative study of arima and ann models

This paper addresses the concerns of Faraway and Chatfield (1998) who questioned the forecasting ability of Artificial Neural Networks (ANN). In particular the paper compares the performance of Artificial Neural Networks (ANN) and ARIMA models in forecasting of seasonal (monthly) Time series. Using the Airline data which Faraway and Chatfield (1998) used and two other data sets and taking into consideration their suggestions, we show that ANN are not as bad as Faraway and Chatfield put it. A rule of selecting input lags into the input set based on their relevance/ contribution to the model is also proposed.

An Appraisal of the Use of Airline Data in Assessing the World City Network: A Research Note on Data

Information on air passenger flows is potentially a prime data source for assessing spatial patterns in the world city network, but previous analyses of this issue have been hampered by inadequate and/or partial data. The ensuing analytical deficiencies have reduced the overall value of these analyses. Therefore, this paper examines how some of these deficiencies might be rectified. First, it reviews the rationale for using airline data to analyse the world city network. Secondly, it assesses the problems encountered by previous research. The third section elaborates on the construction of a global intercity matrix based on the so-called Marketing Information Data Transfer database and explains how this matrix can circumvent some previously identified problems.

A COMPARATIVE STUDY OF FORECASTING MODELS FOR TREND AND SEASONAL TIME SERIES DOES COMPLEX MODEL ALWAYS YIELD BETTER FORECAST THAN SIMPLE MODELS

Many business and economic time series are non-stationary time series that contain trend and seasonal variations. Seasonality is a periodic and recurrent pattern caused by factors such as weather, holidays, or repeating promotions. A stochastic trend is often accompanied with the seasonal variations and can have a significant impact on various forecasting methods. In this paper, we will investigate and compare some forecasting methods for modeling time series with both trend and seasonal patterns. These methods are Winter's, Decomposition, Time Series Regression, ARIMA and Neural Networks models. In this empirical research, we study on the effectiveness of the forecasting performance, particularly to answer whether a complex method always give a better forecast than a simpler method. We use a real data, that is airline passenger data. The result shows that the more complex model does not always yield a better result than a simpler one. Additionally, we also find the possibility to do further research especially the use of hybrid model by combining some forecasting method to get better forecast, for example combination between decomposition (as data preprocessing) and neural network model.

On the use of inadequate airline data in mappings of a global urban system

Air transportation flows are potentially a prime data source in assessments of a global urban system. There are, however, two major obstacles that weaken their overall value. First, the potential usefulness of this data source is often assumed rather than properly discussed. Second, although a more refined discussion shows that airline data are indeed potentially useful, previous analyses have been hampered by inadequate and/or partial data.

Dispersed fares within a fare class: How can the reality be harnessed?

Most revenue management seat allocation models require inputs of the revenue value associated with the bookings expected in each fare class. This paper examines the impacts of more realistic fare value assumptions on the revenue performance of a commonly used seat allocation optimisation model (EMSR) and a new heuristic decision rule, called dispersed fare rule (DFR). Typical fare data for airlines, hotels, cruise lines and rental car companies show that the fares in each fare class follow a probability distribution, rather than the typical assumption of a single fare value. The revenue performance of this new DFR approach on a single leg is compared. Simulation analysis of many different scenarios is applied, using actual airline data, to show that the new decision rule can take advantage of the variability of actual fare values around the mean fare values to generate a significant revenue improvement over EMSR (2–6 per cent).

Constructing robust crew schedules with bicriteria optimization

AbstractOptimization‐based computer systems are used by many airlines to solve crew planning problems by constructing minimal cost tours of duty. However, today airlines do not only require cost effective solutions, but are also very interested in robust solutions. A more robust solution is understood to be one where disruptions in the schedule (due to delays) are less likely to be propagated into the future, causing delays of subsequent flights. Current scheduling systems based solely on cost do not automatically provide robust solutions.These considerations lead to a multiobjective framework, as the maximization of robustness will be in conflict with the minimization of cost. For example crew changing aircraft within a duty period is discouraged if inadequate ground time is provided. We develop a bicriteria optimization framework to generate Pareto optimal schedules for the domestic airline. A Pareto optimal schedule is one which does not allow an improvement in cost and robustness at the same time.We developed a method to solve the bicriteria problem, implemented it and tested it with actual airline data. Our results show that considerable gain in robustness can be achieved with a small increase in cost. The additional cost is mainly due to an increase in overnights, which allows for a reduction of the number of aircraft changes. Copyright © 2003 John Wiley & Sons, Ltd.

Posterior Analysis of Stochastic Frontier Models with Truncated Normal Errors

Previous work in stochastic frontier models with exponentially distributed one-sided errors using both Gibbs sampling and Monte Carlo integration with importance sampling reveals the enormous computational gains that can be achieved using the former. This paper takes up inference in another interesting class of stochastic frontier models, those with truncated normal one-sided error terms, and shows that posterior simulation involves drawing from standard or log-concave distributions, implying that Gibbs sampling is an efficient solution to the Bayesian integration problem. The sampling behavior of the Bayesian procedure is investigated using a Monte Carlo experiment. The method is illustrated using US airline data.

Optimizing aircraft routings in response to groundings and delays

This paper presents the time-band optimization model for reconstructing aircraft routings in response to groundings and delays experienced over the course of the day. Whenever the schedule is disrupted, the immediate objective of the airlines is to minimize the cost of reassigning aircraft to flights taking into account available resources and other system constraints. Associated costs are measured by flight delays and cancellations. The time-band model is constructed by transforming the routing problem into a time-based network in which the time horizon is discretized. The resulting formulation is an integral minimum cost network flow problem with side constraints. Conditions for which an exact solution to the model represents an optimal solution for the original problem are stated. The transformation procedure is polynomial with respect to the number of airports and flights in the schedule. Computational experience shows that the underlying network structure of the transformed problem often leads to integral solutions when solved with a standard linear programming code. Empirical results using Continental Airline data demonstrate that the solutions obtained are either provably optimal or no more than a few percentage points from the lower bound.

Flight graph based genetic algorithm for crew scheduling in airlines

Abstract Crew scheduling is an NP-hard constrained combinatorial optimization problem, very important for the airline industry [G. Yu (Ed.), Operations Research in Airline Industry, Kluwer Academic Publishers, Dordrecht, 1998]. We solve this problem using a genetic algorithm applied to a flight graph representation that represents several problem-specific constraints, unlike previous attempts [D. Levine, Application of a hybrid genetic algorithm to airline crew scheduling, Ph.D. dissertation, Computer Science Department, IIT , Chicago, USA, 1995; J.E. Beasley, P.C. Chu, A genetic algorithm for the set covering problem, Eur. J. Oper. Res. 94 (1996) 392–404; P.C. Chu, J.E. Beasley, A genetic algorithm for the set partitioning problem, Technical report, Imperial College, UK, 1995]. In extensive experimental comparisons on flight data of several airlines, the new approach performed better than other approaches in 17 out of 24 data sets.

IBM journey management library: an arena system for airport simulations

Airline passenger terminal congestion caused by increasing passenger traffic results in unsatisfactory levels of customer service. We discuss a simulation modelling tool to help airlines and airports to use advanced technology to improve service to passengers. The tool consists of custom designed, reusable modules that represent the most common airline and airport system data, logic and processes. A model of an actual airline operation based on this approach is described.

Nonparametric measurement of cost competitiveness and productivity

A nonparametric method of measuring cost competitiveness is empirically applied over international airlines data and a comparison made of the two subsets: competitively efficient and the nonefficient.

A dynamic efficiency model using data envelopment analysis

Changes in productivity due to technological progress or regress are incorporated in models of data envelopment analysis (DEA) in two ways: either through a network technology or in terms of a capacity variable or capital input embodying technical progress. The second way is analyzed here in terms of time-varying capital inputs due to the changes in input prices over time. This provides an optimal control theoretic view of the time path of capital inputs which minimizes a discounted sum of total input costs for each DMU. The cost of risk aversion and the adjustment cost of input price fluctuations may then be minimized in an extended DEA model framework. These extensions of the DEA model are analyzed here and illustrated by an empirical application to international airlines data.

Airline failure and distress prediction: a comparison of quantitative and qualitative models

In this paper, an exploratory study of failure and distress prediction models is presented based on a qualitative survey among new-entrant airline managers and a quantitative data source containing financial and traffic data of new-entrant airlines. Using logistic regression, it became evident that qualitative variables are a viable source for model construction. They contribute to a deeper understanding of the underlying distress and failure process. A number of significant predictor variables are discussed in the context of managing new-entrant airlines.

Energy demand and environmental pressures due to the operation of Olympic Airways in Greece

Environmental issues relating to air transport are analyzed. An inventory calculation system for air traffic is developed to determine annual fuel consumption and emissions. It uses airline data on routes, hours of flights, density of traffic, fleet mix, and ratings of engine manufacturers. It was applied to an airline company in Greece. Yearly environmental loads have been calculated and show that significant amounts of pollutants are received in areas around airports.

Testing for technical change: a nonparametric application

A method of evaluating the impact of technical change on productivity in a Farrell-type model is developed here and empirically applied over international airlines data.

The Influence of Top Management Team Heterogeneity on Firms' Competitive Moves

The authors acknowledge support from Columbia University's Management Institute. Warren Boeker, Sara Keck, John Michel, Peter Murmann, Michael Tushman, and Ruth Wageman made helpful suggestions on earlier drafts. We are grateful for access to the airline data base jointly assembled by the third author, Martin J. Gannon, Curtis M. Grimm, and Ken G. Smith. This paper explores the executive origins of firms' competitive moves by focusing on top management team characteristics, specifically on team heterogeneity, rather than on the more often studied environmental and organizational determinants of such behaviors. Arguing that competitive actions and responses represent different decision situations, we develop propositions about how heterogeneity may enhance some competitive behaviors but impair others. With a large sample of actions and responses of 32 U.S. airlines over eight years, we find results that largely conform to our propositions. The top management teams that were diverse, in terms of functional backgrounds, education, and company tenure, exhibited a relatively great propensity for action, and both their actions and responses were of substantial magnitude. Heterogeneous teams, by contrast, were slower in their actions and responses and less likely than homogeneous teams to respond to competitors' initiatives. Thus, although team heterogeneity is a double-edged sword, its overall net effect on airline performance, in terms of changes in market share and profits, was positive.*

Systematic measures of dynamic Farrell efficiency

A method of measuring the systematic component of dynamic efficiency characterized by Farrell's nonparametric method is developed and empirically applied over airlines data.

Dynamic Farrell efficiency: a time series application

Dynamic versions of Farrell efficiency are formulated here and empirically applied to airlines data to estimate productivity growth and technical efficiency changes.

Perishable-asset revenue management: Generic and multiple-price yield management with diversion

In any situation of fixed capacity and a perishable service or product, firms want to avoid spoilage of the service or product and receive the most revenue possible in the face of uncertain demand. Stimulation of demand from price-sensitive customers, however, through discount prices to customers that reserve early can help fill capacity. This is attractive if non-price-sensitive customers can be prevented from diverting to the lower rates. This paper provides generic results for deciding how many discount units to sell (the so-called yield-management problem) and extends the generic problem in several ways: (1) to situations with continuous, not discrete resources, (2) to treatment of yield management jointly with overbooking, and (3) to problems with diversion with more than two price classes. Simulation studies using airline data suggest that our decision rules perform better than previous rules now used in practice.

MULTIPOINT COMPETITION, MARKET RIVALRY AND FIRM PERFORMANCE: A TEST OF THE COMPLETE MEDIATION MODEL.

After controlling for predictors of rivalry and economies of scope, this study uses U.S. airline panel data to test whether multipoint competition has a positive effect on economic performance, and...