Sunday, December 2, 2012

Appreciating the role of data in Atc: Part I - Methods of Data Anaylsis

Preview Questions
Part I
How do we analyse data in ATC?
Imagine that you want to buy a TV, you think about it for some time. You consult the internet. You consider the type of TV, the size, and your budget. You imagine the best place for it. You discuss the idea with other occupants of your house or apartment. You talk with friends too. You consult the internet again. When you go to the store you ask the sales' clerk for an opinion and then you decide. Data analyses follow the exact methodology.

Did you know that data analysis forms part of your daily activities?
Your objective is to buy a TV. But before you make your decision, you carefully consider data evidence.   Think about it for a minute. You research the idea first. Then you do a survey. Yes, you do a survey by asking others for their opinions. And that survey is not an ad hoc procedure. 
Your research together with your knowledge and preferences for a TV help you to identify certain investigative variables such as Type, Size, Colour, Features, Cost etc. After you collect your data, you analyze the data by using  at least 1 of the following processes: CBA*; MCA**; Regression Analysis***. 
You analyze the evidence by carefully considering the data to help you determine the feasibility or the practicality of the idea. You even carry out confirmatory tests by checking for comparative information on the internet and chatting with the sales' clerk at the store. You  do all this because you want to make an informed decision about buying a TV. 
The behavioral economists call it rationality. The psychologists call it perception and we can add the neoeconomical concept of social interaction. Whatever our academic thinkers call it, I present this example of buying the TV to show that we automatically use data analysis in our daily lives. Yes we employ data-analysis in our decision-making for daily living. 
The methodology and the concepts surrounding data analysis in the Atc unit can be also applied in an exact manner. We use data evidence in the Atc unit to help us grasp the feasibility or the practicality of the idea.
All studies whether technical or otherwise should include an analysis and not be confined to speculative methods in Atc. In Part II of this post and thereafter we will look at case studies to demonstrate how we can apply the principle of appreciating the role of data in Atc. But first, we need to understand the issue of appropriate data collection.
With the TV purchase, you would not think of asking your friend who lives halfway around the world what brand of TV she uses unless the brand name matters. You also will not ask her age because it is not an important detail for you in deciding on your TV purchase. But you might both discuss features in an effort to find a product that best appeals to your preferences or what you will like about your TV. In like manner, we gather data that we consider relevant to the specific analysis. 
There are different ways of collecting data. I always thought that a survey was the only method. I have since learnt that data can be collected through experimental economics or games. In these games, participants play simulated roles and fill a very small survey. I hope to try that with Atcos one day. 
Data collection should not be limited to gathering information about flights and passenger movement. We all need to digress from the archaic concept that we only need to keep records that will assist in airline economics and air space billing operations. Most of the empirical analyses outside of monetary calculations or counting processes are centered on measurements or analyses of behavior. We can apply the same notion to ATC, ATM and other aspects of aviation.
A survey is the cheapest, quickest method to collect information from participants. Formulating a survey or setting up a game involves much preparation and research. I spent a whole semester learning how to do both formulations. The advantage of using a custom-made survey over a generic one is that the former is suited to the specific information that will help us in the analysis. When we use generic surveys, we may need more than 1 dataset to capture the information that gives us the best understanding.
The study on fatigue among controllers which, will be highlighted on Wednesday, was aimed at isolating the organizational economic effects of fatigue. The first inventory of wellbeing attempted to examine fatigue and the presence of other detractors to wellbeing in the organization. Conditional constructs like fatigue and behavioral constructs like wellbeing are not easy to quantify because they are a collation of overlapping individual perceptions. 
An example of a measurement scale
To measure these and other derivatives of organizational behavior, different types of measurement scales are normally used which, allows the participant to rate an item. This allows the analyst to get an accurate picture of complex constructs or difficult-to-measure constructs such as fatigue and wellbeing. 
The statistician sorts and uses the rated responses from the survey or game to determine the presence of relationships, to validate an observation, to identfy factors or to observe the trends and forecast an outcome****. 
It is imperative that anyone holding a managerial post in the Atc unit possess the solid foundation of matching academic skills in addition to the Atc background so as to be able to understand and appreciate better the process of data analysis in Atc. In this way, the adequately trained manager will be more adept in spearheading any processing of both the good and bad day-to-day issues of the Atc unit. She should thus expect to make better decisions with better outcomes. 
Of course, data management and analysis are costly processes. But any process of analysis is costly. Decision-making in an organization carries a price. In the Atc unit it may be expensive but certainly less costly to adapt orthodox methods of analysis rather than increasing the workload of controllers and their assistants by implementing questionable ethics of organizational operation. 
Do you understand fully the role of data in Atc?
Now that we have established a simple route for data analysis, I think it will be easier to present highlights from several studies. In Wedneday's post, we will consider the highlights of the study on fatigue to demonstrate the concepts of data collection and analysis explained in today's post.
Il pleut des cordes à Trinidad (its raining cats and dogs in Trinidad). Bonsoir à tous! (Have a good evening everyone!)

*Cost Benefit Analysis
**Multi-Criteria Analysis
***Graphical Analysis
****Time Series Analysis

Reddit Digg Share on Tumblr Delicious StumbleUpon Follow Me on Pinterest Print Friendly and PDF

No comments :

Post a Comment