|
FAQ |
Columns |
Companies |
Events |
Govt. Initiatives |
Fraud |
knowledgeBase Mailing List | News | Research Reports | Vendors | VC Funding | Search |
|
||
|
|
| |
|
||
|
||
|
||
|
||
|
||
|
 |
Home => Research Quality Service Index's impact on Airline IndustryThe airline industry is in the midst of tremendous change, due to multiple bankruptcies resulting from the airlines’ inability to gain market share and contain costs. Larger airlines such as Delta, Northwest and United have found it impossible to stay afloat partly due to the proliferation of newer, more nimble lowcost competitors, as well as higher fuel prices, ballooning healthcare costs and tougher relations with unions. The industry as a whole lost nearly $9 billion US Dollars in 2005, according to the Air Transport Association. To overcome these difficulties, airlines have implemented severe cost-cutting measures, as well as strategies designed to increase revenues. Within the industry, we are already seeing the beginnings of a wave of consolidation. The 2005 merger between US Air and America West and rumors of a likely Continental-United merger are only the tip of the iceberg. Notwithstanding the fact that larger airlines are still struggling to get the right low-cost model to work, we are seeing an increasing number of these airlines promoting low-cost and no-frill airlines. All this points towards a desperate dash towards increasing market share, because the stark reality is that cost cutting can only go so far. Current Industry Modeling Market share has a strong influence on revenues generated by an airline. And, because of inherent difficulties involved in collection and analysis of data, many airlines use a statistical model to calculate current market share in any region, or even between given airport-s. This model, known as Quality of Service Index (QSI), relates the number of passengers traveling in a particular itinerary to the “quality” in relation to other itineraries between the same airports. Quality in a particular itinerary is defined as a function of various service attributes of that itinerary and the importance given to those service attributes by passengers traveling in that itinerary. This degree of importance or “preference” is measured in terms of “Preference Weights.” For a given QSI model, these preference weights are obtained using statistical techniques and/or analyst intuition. Once the preference weights are obtained, QSI values are calculated for each itinerary. Passenger Preference Weights are derived using a log-linear (Logit) model that captures the interaction of passenger preference and service attribute. The Logit model employs a wide range of independent variables (itinerary characteristics), such as: level-of-service (nonstop, direct, single-connect, and doubleconnect) with respect to the best level-of-service available in the airport-pair, carrier, time-of-day, secondbest- connection indicator, second-best-connection-time-difference, elapsed-time-difference, distance, fare, point-of-sale city presence, codeshare indicator, commuter indicator, regional jet indicator and seat variables. Logit models use a technique called Maximum Likelihood Estimation (MLE) to simultaneously estimate parameter weights for all of the above-mentioned independent variables. Using these parameter weights and the characteristics of each itinerary, a QSI value is assigned to it. However, QSI models are problematic on two fronts. A downside to this model is that some preference weights, or a combination, may be dependant on others that are used in the model. This creates a problem known as Heteroskedasticity and leads to the commonly found error in statistical analysis where the same effect is counted more than once, leading to inaccurate results. For example, elapsed itinerary trip time and equipment-type are two service attributes, which are related to each other, as are elapsed– itinerary-trip-time and number-of-stops. In addition, QSI models do not measure the underlying real-world competitive dynamics that may exist among passenger itineraries. A Case Study Asian economies, particularly India, China and to some extent, the GCC (Gulf Cooperation Council), are experiencing an unprecedented boom, driven by newly open markets and the liberalization of trade barriers. This has triggered an increase in travel to these regions, which is good news for the airlines.. As a result, we set out to conduct an analysis of which carriers are benefiting from this boom and which ones need to increase market share. For our study, we chose the Dubai-Mumbai-Singapore-Hong Kong region as it straddles the high-growth Asian economies. Another unintended benefit of choosing this region is that air travel data is not as readily available as it is in the Western hemisphere (U.S. and Europe). Data for the U.S. and European markets is available in what is known as the Market Information Data Tapes (MIDT), a comprehensive archive of passenger travel statistics in the region. MIDT data is either not available for Asian markets or is very sparse. As a result, we do not have the liberty of using statistical regression techniques (QSI or Logit) on MIDT data in order to build a model and estimate demands and market share for individual airlines. We found that the following parameters or service attributes have a significant impact on market share of airlines:
The present study concentrates on three Origin and Destination pair (O&D) markets namely ‘Dubai- Singapore’ (DXB-SIN), ‘Dubai-Hong Kong’ (DXB-HKG) and ‘Dubai-Mumbai’ (DXB-MUM). Assumptions and Description of Variables In the analysis, we have made some practical assumptions. For instance, multiple carrier attribute variables are aggregated into a single variable called “Rank”, specifying the rank of an individual airline based on different parameters, as found in the website http://www.airlinequality.com. As the study involved international markets which normally fly only jet aircrafts, aircraft type variable has been excluded from our analysis. The data for remaining variables were obtained from individual airline websites. The parameter “type of connection” (CT) is coded as ‘1’ if the connection type is Non-stop and ‘2’ otherwise. The departure time (DT) was coded in different values for different time points as per passenger preference. The time-of-daycurve variable is a high-order polynomial equation which is simplified here as a step function.  The above figure shows time of day preferences graph for the Dubai-Hong Kong market. We can see that a non-stop flight between Dubai and Hong Kong travels for around 7 hours while the time difference between the two cities is 4 hours. Therefore, when a flight departs from Dubai and arrives at Hong Kong, the local times for departure and arrival differ by as much as 11 hours. As shown by the above graph, variable DT is assigned a value ‘1’ if the departure is between 6:00 pm to 11:00 pm present day and arrival is between 5:00 am to 9:00 am next day. This itinerary is considered as a night journey, which an international passenger is more likely to choose. The variable is assigned a value ‘2’ if departure time is the same as the last example. But, if the arrival is the afternoon of the next day, a value ‘3’ is assigned if the journey is slated as day travel and ‘4’ otherwise. The preferences also depend on the time of arrival in Hong Kong. Commuter factor is used to penalize itineraries involving small aircraft, which passengers normally tend to avoid. However, in our study, we consider only long distance international flights where commuter service is generally not available. Point-of-sale (POS) city presence is a variable that signifies whether the airline has its hub between citypairs (e.g., if Emirates has a strong hub presence in Dubai (DXB) then it is more likely that the airline will have a strong POS city presence for the DXB-SIN itinerary). This information was not available, and therefore, not considered in our analysis at this stage. In any case, this factor has a strong correlation with number-of- departures variable for the hub airport (which may also be an origin or destination airport) and therefore excluding it from consideration is not expected to influence market share values significantly. Calculations As the actual passenger movement between the O&D was not available, we estimated the market share for each itinerary ‘i’ ranging from 1 to ‘n’ as a multiplicative function of above parameters.  n - Total number of itinar/routes Freq - Frequency Cap - Capacity Dist - Distance CT - Dummy variable indicating Connection type DT - Dummy variable indicating Departure time We assumed the values of k1=k2=k3=1 and k4=k5=k6=k7=k8= -1 based on the present market conditions, and the market shares for individual airlines were calculated. Results By taking into consideration all the factors mentioned above, the market shares for airlines flying between in a given Origin-Destination pair were calculated.
Table1: Market Share of “Dubai-Singapore” Airline Market
Table 2: Market Share of “Dubai-Hong Kong” Airline Market
Table 3: Market Share of “Dubai-Mumbai” Airline Market Conclusion From our analysis, we find that Emirates leads the way in the ‘Dubai-Singapore’ and ‘Dubai-Mumbai’ markets, with a market share of 39% and 44% respectively, while Cathay Pacific Airways tops the Dubai- Hong Kong market with a market share of 29%. This analysis presents an objective, data-driven, “as-is” picture of a carrier’s competitive standing and is valuable input for senior management in the airline industry. Airlines can use these results to plan and allocate available resources to proactively meet passenger demand and provide quality service, and thereby increase market share and revenues. Airlines that do not perform well in the model can either take steps to improve those service attributes of the itinerary in which they performed poorly, or can decide whether or not they should drop a particular itinerary and start a new one. Our approach works well where there is paucity of MIDT and related data and would also benefit smaller carriers who do not have the wherewithal to spend the large amounts required to access MIDT data. About KARVY Global Services The Knowledge Process Outsourcing group at KARVY Global Services Ltd provides value-added outsourced services to clients in the transportation industry. We help large and small airlines use complex Operations Research and Statically driven techniques that help them determine their position in the competitive landscape, using models such as QSI. Our process-driven approach, governed by best practices such as Six-Sigma, delivers to clients cost savings of nearly 40% depending on specific engagements, while making no compromise in quality. KARVY Global Services is a market leader in providing world class BPO services to clients in Financial Services, Retail, Healthcare, Transportation and High Tech industries. These services include Knowledge Process Outsourcing, Transaction Processing, Finance & Accounts Outsourcing, Human Resources Outsourcing and Voice Services. |
|
|
|
|
Home | About Us
| Introduction To BPO
|
Feedback | Disclaimer | Public Notice | OneIndia
Copyright © 2001-2008 BPOIndia.org |
|
