The create a system of predictive mathematical analysis

Размещено на http://www.allbest.ru/

Content

Introductory part

1. Subject review

2. Previous studies review

3. Subject relevance

4. Purpose and objectives of the study

Chapter 1

1. Data collection

2. Creation of the database

Chapter 2

1. Designing a data warehouse

2. Mathematical Model description

Chapter 3

1. Data analysis and forecasting

2. Applying mathematical tools

Conclusion

Bibliography

Introduction

1. Subject review

Today Russian Federation is suffering from a heavily misbalanced situation in the level of development of its regions. This trend can be noticed in many areas, for instance, social and economic situation in various regions may differ significantly. The Moscow region is far more developed in many ways in comparison with other regions of the country. For example, one study concludes that "the gross regional product per capita in Moscow is more than 2 times higher than the same average in the rest of Russia".

This situation has a knock-out effect on various economic and social activities, including the system and infrastructure of air transport. A huge part of air cargo and passenger traffic goes through Moscow transport hub (MTH). This indicates the development of a centralized and misbalanced structure of transport routes. The ratio of the average outgoing monthly air freight traffic for ten largest cities of Russia (2010-2011, according to Trading Clearing House) are shown in Figure 1.



Figure 1

The diagram clearly shows that Moscow beats its closest pursuer (St. Petersburg) more than 10 times. Also it should be noted that the share of MTH in airline transportation in Russia rose from 30% to 55% over the past 10 years.

Thus, the MTH has turned into a massive trade and distribution center for the entire country. For such a big country as Russia such an unbalanced distribution plays an unfavorable role for sustainable development and creates a number of serious structural economic problems.

2. Previous studies review

data collection mathematical model

This subject has been raised repeatedly. Reports, studies and proposals were submitted. Both governmental authorities and commercial organizations that are members of the cargo and passenger airline business community have shown their stand on the problem. Also, the academic society and even the commercial sector of the economy that is not directly involved in airline business are interested in the problem. All these parties seriously consider and analyze the problem from different perspectives which allow to understand the processes currently going on in the industry.

The report "Some aspects of the regional airline transportations" by CEO of Airline Company «Flight» on the International Air Transport Forum shows how big the cargo traffic is misbalanced between MTH and other regions. The speaker focuses on such important issues as one-way flow of goods, very small traffic volume among the regions and the high competition of air transport with other types of transport. The speaker refers to the data on the density of the population in the federal districts, the data on the amount of warehouse space in regions, as well as the data on the inbound and outbound flows of goods in different cities of Russia. One of the conclusions of the report indicates that business community expects no change in the near future in the current pattern of flows as well as no reduction of the existing misbalance between consumption in the central part of Russia (especially in Moscow), and in the eastern regions.

A look at the problem from the academic point of view is illustrated in the report "Analysis of the status and development of the air transport system (in Russia)". The author of the report repeatedly points out to the fact of deteriorating state of the airfield network in Russia. He provides some interesting statistics, such as "the number of existing airports in the Russian Federation decreased from 1450 to 351 from 1991 to the present" or "in general depreciation of airfield infrastructure is close to 80% "(referring only regional airports). Among the findings of the report is the idea that the government should urgently deal with this problem by improving the infrastructural suitability of regional airports.

The issue was also considered on the governmental level, with some decisions already made. The Ministry of Transport has submitted the "Road map" project on the development of regional air traffic until 2030 that would help to solve a lot of issues in the industry. The project aims at providing financial support for the airports, development of a minimum transport accessibility standard, improvement of government regulation, as well as reducing the cost of regional passenger air fare. In general, the objectives of the project are very relevant to the industry. The question is whether they will be fully implemented and how quickly it will be done.

The article "The business model of air cargo in the Russian Federation" also should be noted. Authors focus on transport relations with other countries. The article refers to the potential of the Russian air transport infrastructure relatively to the international airline companies. The authors say that in order to benefit from this potential Russia needs to introduce an 'e-freight' standard, improve (expand) the airfield network, simplify of admission procedures, as well as changing of the national legal framework on the issue. A central point of discussion refers to extension of the national airport network and in particular to switching from a centralized structure to a distributed structure of airport network consisting of multiple "hubs" for more efficient operation of the airlines network.

3. Subject relevance

From all the facts mentioned above, it can be concluded that current airport network structure in Russia represents a centralized formation with its heavy dominating core in the Moscow transport hub; however, regional airports and airfields in the majority of cases are characterized by lack of development, lack of adequate infrastructure and the extremely strong deterioration of the equipment. Also it should be noted that more than half of traffic by air pass through the MTH, while the flow of goods among the regions immeasurably small. Considering territorial vastness of Russia, such an unbalanced situation in the industry is unacceptable. Issues of restructuring industry are emerging. Some immediate actions must be taken, including analysis and proposals for the changes in the structure of the air transport network. This study aims to address this emerging issue.

4. Purpose and objectives of the study

The main purpose of this study is to create a system of predictive mathematical analysis which will allow to plan the development of air transportation and will provide the possibility to deliver the concept of the alternative structure of airport system in Russia. The main objectives of the research are:

· To investigate the problem and identify the current situation in Russian air industry by reviewing of existing literature sources, media reports, research articles, etc.;

· To create a test sample and identify a data structure for further processing;

· To design the data warehouse and select the mathematical tools that will provide the basic functionality of the analysis system of air transport in Russia;

· To forecast future trends in the Russian aviation industry and make proposals for alternative structure of national airport network using analysis system.

Chapter 1

1. Data collection

Determination of data structure and composition will be the first step in achieving the major objective of this research project. The following aviation industry standard reporting form #14-CA "Traffic volume between major air destinations in Russia" was taken as a basis for providing information. This form requires monthly reporting of air traffic volume and shipments both domestic and international. An example is shown in Table 1.

Table 1.

Airline code (DAL*, IAL**)	Transportation type (REG, IRG)	The pair of cities	Transportations
		From	To	Passengers	Freight, tones	Mail, tones
1	2	3	4	5	6	7
DAL	REG	Saint-Petersburg	Krasnoyarsk	25	0	0
DAL	NRG	Moscow	Sochi	69	0,7	0,01
IAL	REG	Moscow	Berlin	75	0,4	0
DAL	REG	Khabarovsk	Saint-Petersburg	47	0	0,01

*DAL -- domestic airlines; **IAL -- international airlines

This reporting form contains some excessive data not required for further research. Also some relevant data is missing in the form. An unnecessary data include:

- Airline code (DAL, IAL). This project deals with domestic air transport flows in Russia; hence, the inclusion of IAL data in the model is not feasible.

- Transportation type. There is no need to separate traffic into regular and irregular. This paper focuses not on the individual airline company's traffic volumes, but on the volume of traffic between airports in general. Transportation type does not influence on it.

Also there is some data that should be added to the form:

- The distance between airports. This parameter is very important when organizations choose a transport mode. For example, air transportation will be the best way to carry cargo from Moscow to Vladivostok due to the vast distance between the two cities and land infrastructure features of the country.

- Detail of the airports. The table shows only the names of the cities of departure and arrival, but there is no detailed information on airports which will be useful for research.

- A broader territorial identity. It means, that the name of the area / region / republic and the name of the federal district will be added to the table. This information will be also useful in future research.

After we determine the composition and structure of the data it is necessary to identify the method of collecting data on passenger and freight traffic between the airports of Russia for 5-10 years. There are two versions of data collection:

1) Collection of actual statistical (real) data;

2) Creating a model data based on actual trends in social and economic development of the regions.

Unfortunately, the collection of actual data on air transportations is not feasible. In Russia there is no single, objective public source of accurate information in this area. There is no such an open framework, which would include all the necessary information for the investigation. Therefore, the study suggests a method of creating model data for analysis purposes. Such data need to be generated not only based on common sense, but also taking into account factual information that could affect the amount of air traffic. Thus, the model data will be compiled based on:

· the population of the regions;

· gross regional product per capita;

· information on the development of industry in the regions;

· information on the standard of living in the regions;

· natural seasonal fluctuations;

· total traffic volume for each year from 2003 to 2011.

A model part of the data relates only to the volume of air traffic, while remaining data is real (for example, information about the airports).

After comprehensive data collection the study moves to the next step which suggests creating a structured database of air transportation that will provide easy handling and storage of information.

2. Creation of the database

At this stage an operational database will be created that will include information about air transportations. This database then will serve for the data warehouse as the primary source. Model data generated in the previous step is also taken to create the database.

In the first step it is necessary to develop a structure of the database. It must match the structure of model data. Based on these assumptions, the tables that will display the information corresponding to the data model should be selected as follows:

· Region -- a table containing information relating to the territory (region name, name of the federal district);

· City of departure - a table containing information on the point of departure;

· City of arrival -- a table containing information on the point of arrival;

· Transportation type - a table containing information about the type of traffic (passenger, cargo, mail);

· Transportation - a table containing information on the amount of traffic between two Russian airports, the distance between them, the month and year of the carriage. Table structure displayed in Figure 2.

Figure 2

So, in the first chapter the issue of data composition for the purposes of the study was discussed, the strategy for data collection was developed, and the relational database has been created and filled in. This database will serve as the source for the data warehouse, which will be developed in the next chapter.

Chapter 2

1. Designing a data warehouse

As stated in the introduction, the purpose of this study is to design an information system of analysis that will allow to plan the development of air transportations and will provide the possibility to suggest an alternative structure of national airport network in Russia. Therefore, this analysis process can be divided into two logical parts. The first will concentrate on planning and the second part will deal with redistribution of air transportation flows.

The first component of the developing analysis tool will be implemented in the form of multi-dimensional data storage system based on Microsoft SQL Server Analysis Services. The data warehouse (definition by Bill Inmon) “is a subject-oriented, integrated, time-specific and immutable set of data which support decision making”. In the case of air transportation - decision making means the prediction of future air freight and passenger traffic volumes. This system will provide data in a multi-dimensional cube with a fact table that reflects the volume of air freight/passenger flows between airports and dimension tables, such as time, city of departure, city of destination and shipment type. The structure of the data warehouse is presented in the form of "star" and it conditionally looks as shown in Figure 3.



Figure 3

The database created in chapter 1 serves as an external data source to the data warehouse.

The next step is to describe the mathematical model and the tools that will be used to show the redistribution of air transportation flows.

2. Mathematical model description

Description of the mathematical model is important for the implementation of the mathematical tools that will serve as the second component of the analytical system.

The whole structure of the air transport flows among Russian airports can be displayed as a complex network or a graph. Vertices in the network will refer to airports, while arcs between nodes reflect the volume of traffic between airports. A fragment of such a network (with four cities) is shown on Figure 4.



Figure 4

The figure vertices signed as cities, and both sides of the arc signed refer to freight volume. Closest arc's number to the vertices indicates the volume of outbound flow. For example, the average freight volume from Moscow to Krasnodar - 97 tons, from Krasnodar to Moscow - 65 tons, etc. Also, this information can be displayed in the table as shown below in Table 2.

Table 2

City In	Moscow	Khabarovsk	Krasnodar	Novosibirsk
City Out
Moscow	0	455	97	243
Khabarovsk	374	0	<2	25
Krasnodar	65	2,5	0	4
Novosibirsk	196	47	3	0

This type of data is collected for 26 vertices, that is, for the 26 largest Russian airports. This is done for the purpose and scope of the model to be created in this study (that network does not represent a real airport structure in Russia). From the table view it is very easy to get a general incoming and outgoing freight volume, summarizing data for rows or columns.

After uploading data from the operational database to our data warehouse, it is necessary to use the method of optimizing the network in terms of redistribution of air traffic. In the model it is assumed that all airports will become clustered into groups with one central airport in each group.

In the second chapter the data warehouse was designed and mathematical methods for the analysis of air transportations in Russia were chosen and described.

Chapter 3

1. Data analysis and forecasting

The objective of Chapter 3 is to deliver an alternative structure of national airport network in Russia based on the analysis results. This will be achieved in two stages. Firstly, it is necessary to assume the amount of air transport flows for 2015. Secondly, mathematical model that was briefly described in the second chapter must be properly applied to be able to work under these assumptions.

So, the first stage is realized in an environment Microsoft SQL Server BI Development Studio. One of the built-in Data Mining models is used for forecasting. Data mining is a virtual structure that stores data used in the performance of data mining in SQL Server. Information in the model is in the same form as in the database, but instead of the actual data it contains rules and patterns for the data stored in the model. These rules and patterns are interpretation of multidimensional data in the form of statistical information, which is then used to predict future behavior and changing certain aspects of the data.

Application of a data mining model shown that the total volume of air traffic in Russia increased, however, the share of the total MTH air flows in 2015 compared to 2011 increased from 55% to 61%. This tendency shows the increasing imbalance between the Moscow region and other regions.

After provision of relevant information on the forecast, data can be processed with usage of mathematical tools.

2. Applying mathematical tools

This section is devoted to applying the mathematical tools to the predicted data on 2015. The volume of air transport flows is distributed over a network, depending on the route and the distances between cities.

Applying the mathematical model resulted in the transformation of current airport network from centralized structure into a distributed network with one large transport hub in Moscow and six other regional hubs spread across Russia. These hubs are supposed to be situated in Khabarovsk, Magadan, Yakutsk, Novosibirsk, Yekaterinburg and Krasnodar. According to system's prognosis these cities shall become the transport nodes in the structure of the air transport system of Russia. As a result, the share of MTH shall decrease from 61% to 20-25%; furthermore, the share of selected hubs should increase in average to 7-15% each.

Conclusion

The study shows how imbalanced is the air transport system in Russia. A very large share of air traffic passes through the Moscow Transportation Hub that excessively overloads Moscow region. On the other hand, the regional air transport infrastructure remains in a poor state and every year loses its logistics potential.

This study attempts to suggest an alternative structure of airport network in Russia by switching from centralized to distributed formation with several hubs. If applied, this approach will help to significantly decrease the logistics burden carried by MTH and at the same time to ensure the development of regional airports, regions itself, as well as their overall transport infrastructure. After the processing and analysis of data with utilization of applied mathematical tools the proposed airport network will incorporate seven segments. Each segment has a major airport, through which most of the cargo and passenger flows will pass. Proposed changes will allow the aviation industry to develop in a more efficient and balanced way. This can be seen by the example of the United States and Germany. Also proposed structural changes would positively affect overall transit flows through Russia.

As a result of this work a proposal can be made to develop infrastructure of hubs (identified in Chapter 3) that moves current airport network in Russia to a distributed model from a centralized structure. Federal authorities and business community in Russia should both make efforts and pay more attention to this emerging problem because decision making, including financing and managing of those projects in the air transport industry heavily depend on the decisions taken by the State.

Bibliography

1. Диго С. М. «Базы данных. Проектирование и создание. Учебно-методический комплекс», Москва: Изд. Центр ЕАОИ, 2008

2. А.А. Барсегян, М.С. Куприянов, В.В. Степаненко, И.И. Холод «Методы и модели анализа данных: OLAP и Data Mining», С-П: БХВ-Петербург, 2004

3. MSDN: MS SQL Server, <http://msdn.microsoft.com/ru-ru/library/bb545450(v=msdn.10%20).aspx>

4. А. Карпов «Некоторые аспекты региональных авиаперевозок», <http://www.aex.ru/docs/2/2011/7/4/1364/print/>, 2011

5. Р. Ф. Дружаева, Е. И. Меркулова (2012) Бизнес-модель развития грузовых авиаперевозок в Российской Федерации, Наука и транспорт. Гражданская авиация, №1, 20-22

6. М. В. Сацик (2010) «Анализ состояния и развития авиатранспортной системы», реферат

7. Federal State Statistics Service web-site <http://www.gks.ru/>

8. Trading Clearing House web-site http://www.tch.ru/

Размещено на Allbest.ru