Pushka

r Zag

ade

ASeminar Synopsis

On

“BIG DATA TO AVOID WEATHERRELATED FLIGHT DELAYS ”

Submitted to

JSPM’sJAYAWANTRAO SAWANT COLLEGE OF ENGINEERING

PUNE 28SavitribaiPhule Pune University, Pune

In Partial Fulfillment ofT.E. Computer Engineering

T.E. Semester II

Submitted byMr. Pushkar Girish Zagade

Under the Guidance of

Prof. Neelima Satpute

Department Of Computer Engineering

JSPM’sJAYAWANTRAO SAWANT COLLEGE OF ENGINEERING

Academic Year 2014-15

1

Pushka

r Zag

ade

Abstract

This paper identifies key aviation data sets for operational analytics,presents a methodology for application of big-data analysis methods to oper-ational problems, and offers examples of analytical solutions using an inte-grated aviation data warehouse. Big-data analysis methods have revolution-ized how both government and commercial researchers can analyze mas-sive aviation databases that were previously too cumbersome, inconsistentor irregular to drive high-quality output. Traditional data-mining methodsare effective on uniform data sets such as flight tracking data or weather.Integrating heterogeneous data sets introduces complexity in data standard-ization, normalization, and scalability. The variability of underlying datawarehouse can be leveraged using virtualized cloud infrastructure for scala-bility to identify trends and create actionable information. The applicationsfor big-data analysis in airspace system performance and safety optimiza-tion have high potential because of the availability and diversity of airspacerelated data. Analytical applications to quantitatively review airspace per-formance, operational efficiency and aviation safety require a broad data set.Individual information sets such as radar tracking data or weather reportsprovide slices of relevant data, but do not provide the required context, per-spective and detail on their own to create actionable knowledge. These datasets are published by diverse sources and do not have the standardization,uniformity or defect controls required for simple integration and analysis.At a minimum, aviation big-data research requires the fusion of airline, air-craft, flight, radar, crew, and weather data in a uniform taxonomy, organizedso that queries can be automated by flight, by fleet, or across the airspacesystem.

2

Pushka

r Zag

ade

1 Introduction

1.1 Research In The University Of MichiganThe students from the University of Michigan have started a new research whichhelps in understanding the weather of a particular place. They have taken data ofthe weather of the past 10 years. The analysis of this data helps in understandingthe patterns in the weather. This is a very creative and new process. It could leadto understanding similarities in the weather in the past years. It could be of helpin predicting the weather in the future. This can be very helpful for flights. Withthe help of this data, the flights can be cautious of bad weather in advance. So itwill be usefull.

1.1.1 More About Big Data for Predictive analytics

The data used in this research is available publicly. Since it the hourly data of thelast ten years, the data is huge in quantity. Hence, it has to be managed cleverlyand all of it must be taken into consideration. The study of the weather is carriedout keeping the flights and their journeys in mind. This enables the researchersto understand the effect of weather on a particular journey. This is a very uniquestudy. It will help in predicting the delays or preventing them in certain cases.

3

Pushka

r Zag

ade

2 Brief Working

They’ve gathered more than 10 years of hour-by-hour weather observationsand domestic fight data, and are using advanced data analytics to spot patterns andhelp airlines manage more efficiently. While the project uses public data that hasbeen available for years, its size and scope make it unique and it will help in thisproject.

They gather this data in one place and apply this level of computing powerto it. ”That enables us to do a very sophisticated analysis of how weather andflight delays are connected and go far beyond individual airports. We can seehow the weather in Atlanta may affect flight operations in Detroit later in theday, or exactly how a delayed plane on the East Coast ripples through the systemto California.”The chief goal is to enable airlines to anticipate and deal with de-lays before they happen.Today, most airlines compensate for delays by addingslack to the system. They may schedule in extra flight time during the win-ter or keep additional staff members on call. But generally, they don’t look atlarge-scale weather patterns when they’re building initial schedules. And theirability to shuffle resources to deal with emerging weather patterns is limited.

Airlines generally deal with weather delays after they happen.”They want togive them the ability to be a bit more proactive. When they’re able to predictdelays further in advance, they’ll be able to do a much better job of communicat-ing with passengers and optimizing resources.This project may be used to buildcomputer-modeling software that could predict the outcome of an infinite numberof hypothetical flight and weather scenarios, helping airlines spot likely weatherdelays in advance.

That knowledge could enable airlines to adjust their schedules to account forweather patterns. It may also lead to new options for passengers. For example, air-lines could look several steps ahead to predict a future flight delay, then offer pas-sengers a pre-emptive rebooking to avoid it.Imagine youre scheduled to fly out ofDetroit four hours from now and theres a storm in Atlanta.. The airline could usethis data to determine that the storm in Atlanta is likely to delay your plane. Theycould then contact you and offer you a seat on an alternate flight. You save time,and the airline doesnt have to accommodate you on a later flight after the delayhappens.

Airlines could also use the advance warning to allocate their own resourcesmore efficiently, shuffling ground crews, flight crews and other assets to minimize

4

Pushka

r Zag

ade

disruption.The project draws on resources from a wide swath of disciplines in-cluding engineering, computer science and others. He says that breadth of knowl-edge has brought a fresh perspective that could hold far-reaching benefits for bothairlines and passengers.Our background in industrial engineering and computerscience enables us to put existing data together in new ways and ask a whole newset of questions. For me, thats what has been really exciting about this research.

Some of the first changes that passengers see are likely to be simple ones, liketweaks to flight times and more proactive communication.

3 Application

• We can use big data to collect large set of informations and use it for data warehouse purpose.

• Using this technique we can predict the weather information of next 2,3 monthin advance . Using this we can avoid the delays as well as cancellation offlight due to bad weather.

• Big data can also use for data mining purpose.

• By using big data we can analyse the large set of information.

4 Conclusion

I think these new analytics will enable passengers as well as airlines to bettermanage the whole travel process. If airlines can offer more options and passengerscan educate themselves on how to use those options, well see fewer delays and aless stressful travel experience in the years to come.

5

Pushka

r Zag

ade

References

[1] Manyika JChui MBrown Bet al Big data:The next frontier for innovation,competition,and productivity McKinsey Global Institute

[2] Research Trends Issue 30 September 2012

[3] The Fourth Paradigm Data-Intensive Scientific Discoveryedited by Tony Hey, Stewart Tansley,andKrist in Tolle pp 12-15

[4] Big Data for Development: From Information- toKnowledge Societies Martin Hilbert (Dr. PhD.) Hilbert,Big Data for Dev.; pre-published version, Jan. 2013

[5] S. Consulting. The New York CityTaxicab Fact Book.[Online]. Available:http://www.schallerconsult.com/taxi/taxifb.pdf, accessed 2006.

[6] Taxi of Tomorrow Survey, New York City Taxi andLimousine Commission, New York, NY, USA, 2011.

[7] W. Wu, W. S. Ng, S. Krishnaswamy, and A. Sinha, “Totaxi or not to taxi? Enablingpersonalised and real-time trans-portation deci- sions for mobile users,” in Proc. IEEE 13thInt. Conf. Mobile Data Manage. (MDM), Jul. 2012, pp. 320 323.

[8] Y. Ge, H. Xiong, A. Tuzhilin, K. Xiao, M. Gruteser,and M. Paz-zani, “An energy effcient mobile recom-mender system,” in Proc. 16th ACM SIGKDD Int. Conf.Knowl. Discovery Data Mining (KDD), 2010,pp. 899 908.

[9] H. Yang, C. S. Fung, K. I. Wong, and S. C.Wong, “Nonlinear pricing of taxi services,”Transp. Res.A, Policy Pract., vol. 44, no. 5, pp. 337v348, 2010.

[10] K. Yamamoto, K. Uesugi, and T. Watanabe, “Adap-tive routing of cruising taxis by mutual exchange of path-ways,” in Knowledge-Based Intelligent Information and En-gineering Systems. Berlin, Germany: Springer-Verlag, 2010.

6

Pushka

r Zag

ade

7